Re: [VOTE] DUCC contribution to the UIMA sandbox

[Lou DeGenaro]

+1

On Fri, Oct 12, 2012 at 3:12 PM, Jaroslaw Cwiklik  wrote:

> Hi, I'd like to move the DUCC [1] proposal along.
>
> To that end, I'm asking if you will please VOTE on bringing DUCC into the
> Sandbox, after we get a software grant for it.
>
> [ ] +1 Approve the contribution
> [ ] 0  Don't care
> [ ] -1 Veto the contribution (please provide specific comments)
>
> You're all welcome to cast your vote. You dont have to be UIMA committer.
> If you decide to veto the contribution,
> please explain why.
>
>
> [1]
>
> http://markmail.org/search/?q=DUCC#query:DUCC+page:1+mid:hc6xuecvnhb2hzdb+state:results
>

https://cwiki.apache.org/confluence/display/UIMA/powered-by-apache-uima.html

[Lou DeGenaro]

[image: Your Home
Page] Page
Not FoundJust me?

I've found other anomalies too...this the right place to complain (or to be
instructed in the errors of my ways)?

Lou.

Re: Citable UIMA-DUCC reference

[Lou DeGenaro]

Richard,

It's not easy to find. Try
https://cwiki.apache.org/confluence/display/UIMA/DUCC

After the current UIMA-AS release is completed the first release of DUCC
(yay!) will follow, hopefully in the next few weeks.

Lou.


On Fri, Oct 18, 2013 at 5:28 AM, Richard Eckart de Castilho
wrote:

> Hi there,
>
> I'd like to refer to UIMA-DUCC in a publication. Unfortunately, there
> isn't any URL yet that can be referred to and I also didn't find any
> publications on it. Is there anything citable or are there any suggestions
> how to refer to DUCC?
>
> Cheers,
>
> -- Richard

Re: Early review of DUCC pre-release - test code in non-standard place

[Lou DeGenaro]

Marshall,

I'll look into the "MachineFactsTest" and "ReservationFactsTest" and 9
others issue.  Thx!

Lou.


On Wed, Nov 20, 2013 at 9:05 PM, Marshall Schor  wrote:

> I see some classes that look like tests in the main (non-test) source tree
> of at
> least one project: uima-ducc-cli (see for instance,
> org.apache.uima.ducc.cli.ws.json, there are classes "MachineFactsTest" and
> "ReservationFactsTest".
>
> I think these should be moved to the normal maven spot (convention over
> configuration :-) ), so they get run automatically during the build (they
> would
> need to be turned into junit tests too).
>
> A scan of the code shows 9 classes with the word "test" as part of the
> name, and
> none of these are in the src/test/java conventional spot, so none of the
> tests
> are being run as part of the build, I think (I didn't check...).
>
>
>
> -Marshall
>

Re: Early review of DUCC pre-release - general approach for standard javascript libraries

[Lou DeGenaro]

I think there are two cases: build-time and run-time.

For the run-time: I think we have to assume that external access is not
possible for the target cluster.  Hence, at run-time users employing the
DUCC web-server need only access to it to obtain the necessary javascripts.

For the build-time: if the javascripts and images can be obtained by maven
at external sites on-the-fly and put in the proper web-server location that
is perfectly acceptable.  Note, however that some images have been reduced
in size by your truly using image manipulation tools, and thus their
inclusion in our SVN is warranted.

Finally, as to which javascripts to use, the human-readable ones or the
compacted ones, I'm indifferent since I don't spend much time looking at
the guts of these things (other than our own ducc.js).  I don't know that
there is much benefit to using the compacted versions, but am willing to
consider this post-Rel 1.0.

Lou.


On Wed, Nov 20, 2013 at 5:47 PM, Marshall Schor  wrote:

> I may have missed this discussion before, if so, please point me to the
> thread ...
>
> DUCC includes a website that is provided by running a web server (Jetty, I
> think
> - is that right?).
>
> The site makes use of many javascript libraries (jquery and so forth).
>
> When building the website, the current design I think incorporates the
> javascript libraries as part of the web project's "binary" build
> artifacts.  It
> does this by building from a copy of the javascript sources in our svn
> source
> tree; these have been put there manually; as I understand it, this is
> because
> there wasn't a "maven central" kind of artifact that could be incorporated
> at
> build time.  (I do see some jquery maven artifacts, but I'm not sure these
> are
> the "packaging" that you want).
>
> Reading on the internet indicates that jquery comes with some "packaging"
> things
> that let you incorporate /exclude the parts you need, and then you can do
> some
> kind of optimization (I'm guessing) to prepare a faster-to-load version of
> it
> for your application.  I'm guessing we don't do any of that (yet) (maybe
> for a
> future release?).
>
> If we don't do any of that, I'm guessing there are several alternatives for
> arranging to use these kinds of javascript libraries:
> 1) copy the library sources into our SVN, and package these in the binary
> build.  In this case the web pages DUCC serves would use references to the
> "built" subdirectories where these reside to load from.
>
> 2) Have the web pages DUCC serves use references to some standard places
> on the
> web where these libraries reside.  For instance, using 

Re: Early review of DUCC pre-release - test code in non-standard place

[Lou DeGenaro]

My claim is that as of now, all hits on the string "test" or "Test" when
grepping .java's in DUCC are legit.  Hopefully this resolves
MachineFactsTest" and "ReservationFactsTest" and 9 others?

Note that there is a PamTest.java.  If its name is an issue I can change it
to perhaps PamCheck.java.  I happen to like it as it is, but will change if
that is the convention.

Also there are some packages in the examples project having "test" as a
part of the package name.  That's OK?

Lou.


On Thu, Nov 21, 2013 at 8:06 AM, Lou DeGenaro wrote:

> Marshall,
>
> I'll look into the "MachineFactsTest" and "ReservationFactsTest" and 9
> others issue.  Thx!
>
> Lou.
>
>
> On Wed, Nov 20, 2013 at 9:05 PM, Marshall Schor  wrote:
>
>> I see some classes that look like tests in the main (non-test) source
>> tree of at
>> least one project: uima-ducc-cli (see for instance,
>> org.apache.uima.ducc.cli.ws.json, there are classes "MachineFactsTest" and
>> "ReservationFactsTest".
>>
>> I think these should be moved to the normal maven spot (convention over
>> configuration :-) ), so they get run automatically during the build (they
>> would
>> need to be turned into junit tests too).
>>
>> A scan of the code shows 9 classes with the word "test" as part of the
>> name, and
>> none of these are in the src/test/java conventional spot, so none of the
>> tests
>> are being run as part of the build, I think (I didn't check...).
>>
>>
>>
>> -Marshall
>>
>
>

Re: Early review of DUCC pre-release - general approach for standard javascript libraries

[Lou DeGenaro]

I delivered Jira 3447 to remove extraneous material from the webserver's
opensources directory.

Lou.


On Fri, Nov 22, 2013 at 9:04 AM, Lou DeGenaro wrote:

> I think there are two cases: build-time and run-time.
>
> For the run-time: I think we have to assume that external access is not
> possible for the target cluster.  Hence, at run-time users employing the
> DUCC web-server need only access to it to obtain the necessary javascripts.
>
> For the build-time: if the javascripts and images can be obtained by maven
> at external sites on-the-fly and put in the proper web-server location that
> is perfectly acceptable.  Note, however that some images have been reduced
> in size by your truly using image manipulation tools, and thus their
> inclusion in our SVN is warranted.
>
> Finally, as to which javascripts to use, the human-readable ones or the
> compacted ones, I'm indifferent since I don't spend much time looking at
> the guts of these things (other than our own ducc.js).  I don't know that
> there is much benefit to using the compacted versions, but am willing to
> consider this post-Rel 1.0.
>
> Lou.
>
>
> On Wed, Nov 20, 2013 at 5:47 PM, Marshall Schor  wrote:
>
>> I may have missed this discussion before, if so, please point me to the
>> thread ...
>>
>> DUCC includes a website that is provided by running a web server (Jetty,
>> I think
>> - is that right?).
>>
>> The site makes use of many javascript libraries (jquery and so forth).
>>
>> When building the website, the current design I think incorporates the
>> javascript libraries as part of the web project's "binary" build
>> artifacts.  It
>> does this by building from a copy of the javascript sources in our svn
>> source
>> tree; these have been put there manually; as I understand it, this is
>> because
>> there wasn't a "maven central" kind of artifact that could be
>> incorporated at
>> build time.  (I do see some jquery maven artifacts, but I'm not sure
>> these are
>> the "packaging" that you want).
>>
>> Reading on the internet indicates that jquery comes with some "packaging"
>> things
>> that let you incorporate /exclude the parts you need, and then you can do
>> some
>> kind of optimization (I'm guessing) to prepare a faster-to-load version
>> of it
>> for your application.  I'm guessing we don't do any of that (yet) (maybe
>> for a
>> future release?).
>>
>> If we don't do any of that, I'm guessing there are several alternatives
>> for
>> arranging to use these kinds of javascript libraries:
>> 1) copy the library sources into our SVN, and package these in the binary
>> build.  In this case the web pages DUCC serves would use references to the
>> "built" subdirectories where these reside to load from.
>>
>> 2) Have the web pages DUCC serves use references to some standard places
>> on the
>> web where these libraries reside.  For instance, using 

Re: Early review of DUCC pre-release - sign_jconsole_jar fails

[Lou DeGenaro]

Jira 3454 is delivered to detect improper ducc.properties specification of
ducc.jvm and to give helpful message.


On Thu, Nov 21, 2013 at 1:33 PM, Marshall Schor  wrote:

> On a linux box where I ran this after running post_install (which seemed
> to run
> OK), I do not have $JAVA_HOME set:
>
> .ducc_runtime/admin$ echo $JAVA_HOME
> 
>
> Running the sign_jconsole_jar command says it thinks java-home is /usr,
> and then
> it fails to find the file
> /usr/lib/jconsole.jar
>
> -Marshall
>

Re: Early review of DUCC pre-release - building on non-linux platform

[Lou DeGenaro]

Where exactly is this breaking...install, start?  DUCC does not officially
(or unofficially) support Windows OS.  Would adding such code modifications
indicate otherwise?

Lou.


On Mon, Nov 25, 2013 at 4:59 PM, Marshall Schor  wrote:

> I put into my local copy a test inside the chmod method which does:
>
>   String osName = System.getProperty("os.name");
>   if (osName.startsWith("Windows")) {
> return;
>   }
>
> to bypass this on Windows...  Should I commit this?
>
> -Marshall
>
> On 11/25/2013 3:40 PM, Marshall Schor wrote:
> > I see that Java has some methods that claim to set permissions for users
> and
> > "other" (but not group - that needs some Java 7 support).  Perhaps these
> methods
> > could be use in the Crypto class, to be more platform independent?
> >
> > -Marshall
> >
> > On 11/25/2013 2:07 PM, Marshall Schor wrote:
> >> for testing it is more convenient for me to be able to build on
> non-Linux
> >> platforms, such as Windows.
> >>
> >> This used to work, but a recent change now runs a test which tries to
> issue the
> >> "chmod" command, and fails.  Could this test be altered or somehow made
> to work
> >> or be bypassed on Windows?
> >>
> >> -Marshall
> >
>
>

Re: Early review of DUCC pre-release - building on non-linux platform

[Lou DeGenaro]

How about something like this in CryptoTest?:

// ** Windows is not supported **

@Before
public void setUp() throws Exception {
Assume.assumeTrue(!System.getProperty( "os.name" ).startsWith(
"Windows" ));
}


On Tue, Nov 26, 2013 at 8:56 AM, Marshall Schor  wrote:

> The "build" is breaking, not the running :-)
>
> The build (after you fixed it) now runs the test cases :-)  So one of the
> tests
> fails because chmod routine throws an exception (it can't find the
> executable,
> duh, on windows).
>
> I don't think making this modification would indicate that running is
> supported
> on windows; we could add a comment, also, just to be clear.
>
> -Marshall
> On 11/26/2013 7:48 AM, Lou DeGenaro wrote:
> > Where exactly is this breaking...install, start?  DUCC does not
> officially
> > (or unofficially) support Windows OS.  Would adding such code
> modifications
> > indicate otherwise?
> >
> > Lou.
> >
> >
> > On Mon, Nov 25, 2013 at 4:59 PM, Marshall Schor  wrote:
> >
> >> I put into my local copy a test inside the chmod method which does:
> >>
> >>   String osName = System.getProperty("os.name");
> >>   if (osName.startsWith("Windows")) {
> >> return;
> >>   }
> >>
> >> to bypass this on Windows...  Should I commit this?
> >>
> >> -Marshall
> >>
> >> On 11/25/2013 3:40 PM, Marshall Schor wrote:
> >>> I see that Java has some methods that claim to set permissions for
> users
> >> and
> >>> "other" (but not group - that needs some Java 7 support).  Perhaps
> these
> >> methods
> >>> could be use in the Crypto class, to be more platform independent?
> >>>
> >>> -Marshall
> >>>
> >>> On 11/25/2013 2:07 PM, Marshall Schor wrote:
> >>>> for testing it is more convenient for me to be able to build on
> >> non-Linux
> >>>> platforms, such as Windows.
> >>>>
> >>>> This used to work, but a recent change now runs a test which tries to
> >> issue the
> >>>> "chmod" command, and fails.  Could this test be altered or somehow
> made
> >> to work
> >>>> or be bypassed on Windows?
> >>>>
> >>>> -Marshall
> >>
>
>

Re: Early review of DUCC pre-release - building on non-linux platform

[Lou DeGenaro]

Ok, so go ahead and change as you see fit.  My main concern is to not imply
Windows support of any kind.


On Tue, Nov 26, 2013 at 10:45 AM, Marshall Schor  wrote:

> Nice, I didn't know about Assume - I guess it's a feature of JUnit 4.x.y.
>
> This change would skip the whole test.  Since the test tests many things,
> such
> as generating key pairs, and using them to encode and decode a message
> (comparing the result), it would be better I think to be able to run the
> test.
>
> -Marshall
>
>
> On 11/26/2013 10:00 AM, Lou DeGenaro wrote:
> > How about something like this in CryptoTest?:
> >
> > // ** Windows is not supported **
> >
> > @Before
> > public void setUp() throws Exception {
> > Assume.assumeTrue(!System.getProperty( "os.name" ).startsWith(
> > "Windows" ));
> > }
> >
> >
> > On Tue, Nov 26, 2013 at 8:56 AM, Marshall Schor  wrote:
> >
> >> The "build" is breaking, not the running :-)
> >>
> >> The build (after you fixed it) now runs the test cases :-)  So one of
> the
> >> tests
> >> fails because chmod routine throws an exception (it can't find the
> >> executable,
> >> duh, on windows).
> >>
> >> I don't think making this modification would indicate that running is
> >> supported
> >> on windows; we could add a comment, also, just to be clear.
> >>
> >> -Marshall
> >> On 11/26/2013 7:48 AM, Lou DeGenaro wrote:
> >>> Where exactly is this breaking...install, start?  DUCC does not
> >> officially
> >>> (or unofficially) support Windows OS.  Would adding such code
> >> modifications
> >>> indicate otherwise?
> >>>
> >>> Lou.
> >>>
> >>>
> >>> On Mon, Nov 25, 2013 at 4:59 PM, Marshall Schor  wrote:
> >>>
> >>>> I put into my local copy a test inside the chmod method which does:
> >>>>
> >>>>   String osName = System.getProperty("os.name");
> >>>>   if (osName.startsWith("Windows")) {
> >>>> return;
> >>>>   }
> >>>>
> >>>> to bypass this on Windows...  Should I commit this?
> >>>>
> >>>> -Marshall
> >>>>
> >>>> On 11/25/2013 3:40 PM, Marshall Schor wrote:
> >>>>> I see that Java has some methods that claim to set permissions for
> >> users
> >>>> and
> >>>>> "other" (but not group - that needs some Java 7 support).  Perhaps
> >> these
> >>>> methods
> >>>>> could be use in the Crypto class, to be more platform independent?
> >>>>>
> >>>>> -Marshall
> >>>>>
> >>>>> On 11/25/2013 2:07 PM, Marshall Schor wrote:
> >>>>>> for testing it is more convenient for me to be able to build on
> >>>> non-Linux
> >>>>>> platforms, such as Windows.
> >>>>>>
> >>>>>> This used to work, but a recent change now runs a test which tries
> to
> >>>> issue the
> >>>>>> "chmod" command, and fails.  Could this test be altered or somehow
> >> made
> >>>> to work
> >>>>>> or be bypassed on Windows?
> >>>>>>
> >>>>>> -Marshall
> >>
>
>

Re: [jira] [Commented] (UIMA-3452) DUCCs jars should include a version number

[Lou DeGenaro]

In order of preference, I'd be in favor of:

uima-ducc
ducc
uima

and I'd be much less if favor of (meaning against):

lib
ducc-uima

I like the idea of all the ducc jars in their own folder, happily quacking
together!

Lou.




On Tue, Nov 26, 2013 at 11:38 AM, Jerry Cwiklik (JIRA)
wrote:

>
> [
> https://issues.apache.org/jira/browse/UIMA-3452?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13832743#comment-13832743]
>
> Jerry Cwiklik commented on UIMA-3452:
> -
>
> In addition to poms, Ducc's  python scripts also need to change to make
> this work. Currently, the ducc_util.py builds a CP using explicit jar
> names. I think this is because all Ducc jars are directly in the lib
> folder. If we add versions to jars, this would become a maintenance issue.
>
> I propose that we move Ducc jars to a separate folder to be able to take
> advantage of java's wildcard support. Now, there are two choices, we can
> move jars to lib/uima or create a new folder say lib/ducc-uima. Actually,
> there is a third choice - leave Ducc jars where they are (under lib) and
> construct the CP with /lib/*. I believe java does not use recursion (not
> sure if this will change in the future), so only Ducc's jars will be picked
> up.
>
>
> > DUCCs jars should include a version number
> > --
> >
> > Key: UIMA-3452
> > URL: https://issues.apache.org/jira/browse/UIMA-3452
> > Project: UIMA
> >  Issue Type: Improvement
> >  Components: DUCC
> >Affects Versions: 1.0-Ducc
> >Reporter: Jerry Cwiklik
> >Assignee: Jerry Cwiklik
> >
> > Seems like DUCC follows the wrong example set by UIMA which is not to
> include version number in a jar name. Most Apache projects follow a
> convention which includes the version number. Modify poms to append a
> version number. Change CLI pom to name jars with versions.
>
>
>
> --
> This message was sent by Atlassian JIRA
> (v6.1#6144)
>

Re: Early review of DUCC pre-release - Jars in the src tree?

[Lou DeGenaro]

The ones in uima-ducc-examples have been removed via Jira 3461.


On Tue, Nov 26, 2013 at 3:22 PM, Marshall Schor  wrote:

> The source tree has 4 Jars in it; normally the source tree is supposed to
> only
> have "sources" in it. Two are in the uima-ducc-examples/lib, and two are
> in the
> top level uima-ducc/src/main in cglib and saxon.  We should Investigate
> and see
> if some other approach could work that inserts these at build time into the
> binary distribution (if needed).
>
> -Marshall
>
>

Re: Early review of DUCC pre-release - examples project has undocumented scripts

[Lou DeGenaro]

Fixed with Jira 3461.

Lou.


On Tue, Nov 26, 2013 at 3:59 PM, Marshall Schor  wrote:

> The examples project has scripts: build-ducc-authenticator-jar etc., which
> are
> not mentioned anywhere, and which have no internal documentation as to what
> they're for and when you might want to run them.
>
> -Marshall
>

Re: Early review of DUCC pre-release - Example code missing comments

[Lou DeGenaro]

The examples pertaining to the webserver have been commented.  I've
re-assigned Jira 3461 to Eddie for the remaining examples.

Lou.


On Tue, Nov 26, 2013 at 3:51 PM, Marshall Schor  wrote:

> DUCC has a project, uima-ducc-examples, which I'm guessing has examples
> for the
> user to learn from and build upon.  Many of the things here are Java
> classes;
> many have no comments in them.  It would be good to have at least a "class"
> comment - motivating what this class is about and why it is in the
> uima-ducc-examples project.
>
> -Marshall
>

Re: Early review of DUCC pre-release - DUCC Web Server

[Lou DeGenaro]

Thanks for the several issues you raised here.  I believe that all have now
been addressed by Jira 3465.

Lou.


On Fri, Nov 22, 2013 at 11:36 AM, Marshall Schor  wrote:

> The docs say this is accessed at port 42133.  Is this configurable?
> --
>
> The docs mention two display modes: scroll mode and classic mode.  The
> website
> has "Table Style" for scroll/classic.  Is Table Style the same as "Display
> Modes"?
>
> --
> A recent version of the ducc webserver I tried (not sure if it's the
> identical-to-trunk level), had some issues:
> In Google Chrome browser,  switching to Scroll mode made the header row
> have
> column widths that were much wider than the rest of the data in the table,
> so
> they didn't line up (even after trying both kinds of "reloading");  Firefox
> didn't have this issue.  Is there a "recommended/required" browser people
> need
> to use?
>
> Another issue was (in Firefox) the header columns don't re-size (in scroll
> mode)
> when you change the browser width, even after refreshing with the "refresh"
> button in the web page (but they do resize if you use firefox's own
> refresh button).
>
> After picking a sort index, the direction arrows become very hard to see
> on that
> column (the color / size makes it hard to see them).  There's no apparant
> way to
> "undo" the sort picking; you can sort on another column, of course.  (And,
> I
> discovered that if you do a firefox-refresh (not the special in-the-page
> refresh
> button), it seems to undo the sort - not sure this is working as designed
> though).
>
> Is any kind of multi-column sorting supported?  If I sort on one column,
> and
> then pick on another column as the sort key, do items that have now have
> the
> same sort key value, retain the ordering they had before the resort?
>
>
> -Marshall
> On 11/20/2013 2:22 PM, Marshall Schor wrote:
> > I've started to do some preliminary review of the DUCC, in preparation
> for its
> > release.
> >
> > A scan of the code for the phrase "GPL" turns up some things:
> >
> > Maven detects there's some dependencies on JAXB in the project duccdocs,
> (which
> > seems unusual?), and finds that JAXB is a dependency (??) which is dual
> licensed
> > under CDDL and GPL. Does duccdocs really need to have a dependency on
> JAXB?
> >
> > DataTables-1.9.1 has a 3rd version of jquery.js which includes Sizzle.js
> > licensed under the MIT, BSD, GPL.  I think this probably means "triple
> licensed"
> > but I guess it could mean that all of them might apply to different
> parts?  If
> > you look at the current Sizzle.js it is only licensed under the MIT
> license.
> >
> > The jquery.js file is in the source svn 6 times, at multiple (3 or more)
> > versions.  Is this desired?
> >
> > -Marshall
> >
> >
> >
>
>

Re: Early review of DUCC pre-release - duccdocs - more screenshots?

[Lou DeGenaro]

Screen shots added.

Lou.


On Fri, Nov 22, 2013 at 12:58 PM, Marshall Schor  wrote:

> There's quite a bit of documentation on the web server.  I think this
> could be
> enhanced if there was a picture readers could look at.  I think the docbook
> system has some kind of callouts linking things on the picture to text -
> does
> LaTeX have something similar?
>
> -Marshall
>

Re: Early review of DUCC potential release concerns - jconsole, aopalliance, poms

[Lou DeGenaro]

Bogus version of jconsole is removed.

Lou.


On Wed, Nov 20, 2013 at 2:29 PM, Marshall Schor  wrote:

> I see there's a bogus version of jconsole (says for compiling).  I hope we
> can
> find a better way and get rid of this.
> ---
> In the parent pom, the aopalliance.version property is duplicated.  Do we
> really
> have a dependency on this?
> ---
> The uima-ducc pom has a superfluous stanza:
> 
> 
> Jira
> https://issues.apache.org/jira/browse/UIMA
> 
>
> This is in the main pom already, so this can be deleted.  It would be good
> to go
> thru the poms to find other things like this and remove them...
> ---
> The parent pom says
> 
> and then proceeds to list a bunch of things.  I wonder if this is really
> correct.  I looked at one project, for instance: the uima-ducc-sm, and it
> seems
> to (itself) declare 3 dependencies (but of course, since it inherits from
> the
> parent pom, all the others are also pulled in).  I think only the
> dependencies
> that are required for all modules should be listed in the parent.
>
> -Marshall
>

Re: [VOTE] Release UIMA-DUCC 1.0.0 RC2

[Lou DeGenaro]

This was fixed for RC2. I'll close the Jira.

Lou.


On Fri, Dec 27, 2013 at 2:21 PM, Marshall Schor  wrote:

> I see there's an unresolved Major Jira for Ducc with Fix version 1.0.0:
>
> https://issues.apache.org/jira/browse/UIMA-3517
>
> Should this be fixed before doing the release candidate?
>
> -Marshall
>
> On 12/26/2013 4:35 PM, Jaroslaw Cwiklik wrote:
> > Hi,
> >
> > the second release candidate of the sandbox project Apache UIMA DUCC
> v1.0.0
> > is ready for voting.
> >
> > Issues fixed in RC2:
> >
> > 1) Fixed NOTICE and LICENSE files (missing saxon legal info, corrected
> BSD
> > based products)
> > 2) Fixed WebServer issues
> > 3) Fixed typo in README
> >
> > DUCC stands for Distributed UIMA Cluster Computing. DUCC is a cluster
> > management system providing tooling, management, and scheduling
> facilities
> > to automate the scale-out of applications written to the UIMA framework.
> > Core UIMA provides a generalized framework for applications that process
> > unstructured information such as human language, but does not provide a
> > scale-out mechanism. UIMA-AS provides a scale-out mechanism to distribute
> > UIMA pipelines over a cluster of computing resources, but does not
> provide
> > job or cluster management of the resources. DUCC defines a formal job
> model
> > that closely maps to a standard UIMA pipeline. Around this job model DUCC
> > provides cluster management services to automate the scale-out of UIMA
> > pipelines over computing clusters.
> >
> > For a complete list of issues please visit:
> >
> https://issues.apache.org/jira/browse/UIMA-3520?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%221.0-Ducc%22%20ORDER%20BY%20updated%20DESC%2C%20priority%20DESC%2C%20created%20ASC
> >
> > The Maven artifacts are here:
> >
> https://repository.apache.org/content/repositories/orgapacheuima-021/org/apache/uima/
> >
> > The source and binary zip/tars are here:
> > http://people.apache.org/~cwiklik/releases/uima-ducc/1.0.0/RC2/
> >
> > The SVN Tag Checkout:
> >
> http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc-1.0.0/
> >
> > See: http://uima.apache.org/testing-builds.html for instructions how to
> > test release candidate.
> >
> > Please vote on release:
> >
> > [ ] +1 OK to release
> > [ ]  0 Don't care
> > [ ] -1 Not OK to release, because ...
> >
>
>

Re: [VOTE] Release UIMA-DUCC 1.0.0 RC8

[Lou DeGenaro]

I have DUCC RC5 running in 2 places both Linux following (my own)
instructions on the wiki, one from source (Quick Start) and the other from
binary (Install WebServer Demo). See
https://cwiki.apache.org/confluence/display/UIMA/DUCC.  None of the
post-RC5 changes would affect my success.

Lou.


On Tue, Jan 21, 2014 at 6:32 PM, Burn Lewis  wrote:

> Download binary, verified md5sum, followed INSTALL instructions to create a
> multi-user installation on our small test cluster, verified 1.job worked.
> Now running more complex set of real jobs (try3).
>
>
>
> On Tue, Jan 21, 2014 at 4:46 PM, Marshall Schor  wrote:
>
> > downloaded sources and bin distribution, checked out tag.
> >
> > Signatures OK on sources and bin distr (didn't check the maven ones).
> >
> > tag compare to source.zip - OK
> >
> > build from source (on windows, full build with -Papache-release) - OK -
> > but I
> > note there's this persistent warning about "Multiply defined labels":
> >LaTeX Warning: Label `overflow' multiply defined.
> >
> > Checked the license and notice files in the source and binary
> distribution
> > -
> > look OK.
> >
> > The Notice files in the Jars have the wrong copyright date - need to set
> >  (which is not a property, but a top level Maven element -
> > see
> > full discussion about this in other thread).
> >
> > It would be good if someone can do a quick-start verification test on a
> > linux
> > platform.  I'll see what I can do on Windows (I know it's not supported
> on
> > that
> > platform, but perhaps I won't find any issues :-) )
> >
> > -Marshall
> >
>

Re: [VOTE] Release UIMA-DUCC 1.0.0 RC10

[Lou DeGenaro]

Download and unzipped binary tarball
Installed on uima-ducc-vm.apache.org following instructions for "Install
WebServer Demo" on wiki @
https://cwiki.apache.org/confluence/display/UIMA/DUCC#DUCC-t2
Ran 1.job

Also, followed notes to run with simulated agents
Ran 1.job
Registered and started a service

All OK.

[X] +1 OK to release

See it live @ http://uima-ducc-vm.apache.org:42133

Lou.


On Mon, Jan 27, 2014 at 3:07 PM, Jaroslaw Cwiklik  wrote:

> Downloaded and unzipped source tarball - OK
> Built from source - OK
> Ran 1.job - OK
>
> Based on RC9 testing and the above:
>
> [X] +1 OK to release
>
>
>
>
>
> On Mon, Jan 27, 2014 at 11:45 AM, Marshall Schor  wrote:
>
> > Downloaded artifacts, did spot checking on signatures - OK
> >
> > Compared binary download to rc9 - only found time stamp diffs, and the 1
> > change
> > in the ducc.properties file for ducc.agent.launcher.cgroups.enable.
> >  READMEs,
> > LICENSE/NOTICE compared the same.
> >
> > Compare of sources for rc9 / rc10 show only the ducc.properties change,
> > too.
> >
> > built from source - OK
> >
> > Based on this and the previous RC9 checking,
> >
> > [X] +1 OK to release
> >
> > -Marshall Schor
> >
>

re: [ANNOUNCE] Apache UIMA DUCC 1.0.0 Released

[Lou DeGenaro]

Is it premature or could/should the UIMA Home page be revised to include a
link to DUCC.  One suggestion is to follow the sentence on UIMA_AS:

The UIMA-AS Scaleout Framework
is an add-on to the base
Java framework, supporting a very flexible
scaleout capability based on JMS (Java Messaging Services) and ActiveMQ.

with one sentence on DUCC:

DUCC is a cluster management system providing tooling, management, and
scheduling facilities to automate the scale-out of applications written to
the UIMA framework.

with "DUCC" linked to http://uima.apache.org/doc-uimaducc-whatitam.html

Also, perhaps DUCC should be integrated into the diagram that shows the
relationships of the various UIMA pieces-parts in the "What is UIMA?"
section.

Lou.

Re: [ANNOUNCE] Apache UIMA DUCC 1.0.0 Released

[Lou DeGenaro]

Doing as Jerry suggests is fine by me.  That said, however...

It seems that DUCC deserves equal billing in the diagram by either changing
UIMA-AS to UIMA-AS & UIMA-DUCC or removing the UIMA-AS as a qualifier.  In
either case, it is now plural: Scaleout Frameworks, no?  You could get
fancy and just say "Scaleout Frameworks" in the diagram but then show
UIMA-DUCC and UIMA-AS (links?) when hovering.

And I again suggest that some of the text in the changed paragraph be hot
links (same as as those in the left sidebar) to the corresponding -AS and
-DUCC pages, accordingly.

Lou.


On Mon, Feb 3, 2014 at 1:13 PM, Jaroslaw Cwiklik  wrote:

> To review, the proposal is to add a new section to  the left sidebar:
> Scaleout Frameworks
> and add two links:
>
>UIMA-AS
>DUCC
>
> We already have dedicated pages for each of the above.
>
> I can also extend the content (as Lou suggested) to change paragraph that
> currently reads
>
> "The UIMA-AS Scaleout Framework<
> http://uima.apache.org/doc-uimaas-what.html> is
> an add-on to the base Java framework, supporting a very flexible scaleout
> capability based on JMS (Java Messaging Services) and ActiveMQ."
>
> to
>
> "The UIMA-AS and DUCC are both Scaleout Frameworks and are addons to the
> base Java framework. The UIMA-AS supports
> very flexible scaleout capability based on JMS (Java Messaging Services)
> and ActiveMQ. The UIMA-DUCC extends UIMA-AS
> by providing cluster management services to automate the scale-out of UIMA
> pipelines over computing clusters."
>
> DUCC is not mentioned in the diagram so not totally clear if we need to say
> more about DUCC and its relation to the whole
> picture.
>
> Comments?
>
>
>
>
> On Fri, Jan 31, 2014 at 2:38 PM, Jaroslaw Cwiklik 
> wrote:
>
> > I volunteer. Will start working on this next week.
> >
> >
> > On Fri, Jan 31, 2014 at 2:27 PM, Marshall Schor  wrote:
> >
> >>
> >> On 1/31/2014 6:19 AM, Richard Eckart de Castilho wrote:
> >> > I think the diagram may get a bit overloaded if we add more stuff
> >> > like DUCC, uimaFIT and RUTA. I'd suggest to change "UIMA-AS Scaleout
> >> > Framework" to "Scaleout frameworks" instead to subsume UIMA-AS and
> DUCC.
> >> > uimaFIT could fall under "Java Framework" and Ruta would fall under
> >> "Tooling".
> >> >
> >> > How about setting up dedicated pages for UIMA-AS and UIMA-DUCC, as
> they
> >> > exist for uimaFIT and RUTA and link them in the sidebar? I would have
> >> > liked to link to a dedicated UIMA-DUCC landing page in the freecode
> >> > announcement.
> >>
> >> This sounds reasonable to me.  Any volunteers?
> >>
> >> -Marshall
> >>
> >> >
> >> > -- Richard
> >> >
> >> > On 31.01.2014, at 12:13, Lou DeGenaro  wrote:
> >> >
> >> >> Is it premature or could/should the UIMA Home page be revised to
> >> include a
> >> >> link to DUCC.  One suggestion is to follow the sentence on UIMA_AS:
> >> >>
> >> >> The UIMA-AS Scaleout Framework
> >> >> <http://uima.apache.org/doc-uimaas-what.html>is an add-on to the
> base
> >> >> Java framework, supporting a very flexible
> >> >> scaleout capability based on JMS (Java Messaging Services) and
> >> ActiveMQ.
> >> >>
> >> >> with one sentence on DUCC:
> >> >>
> >> >> DUCC is a cluster management system providing tooling, management,
> and
> >> >> scheduling facilities to automate the scale-out of applications
> >> written to
> >> >> the UIMA framework.
> >> >>
> >> >> with "DUCC" linked to
> >> http://uima.apache.org/doc-uimaducc-whatitam.html
> >> >>
> >> >> Also, perhaps DUCC should be integrated into the diagram that shows
> the
> >> >> relationships of the various UIMA pieces-parts in the "What is UIMA?"
> >> >> section.
> >> >>
> >> >> Lou.
> >> >
> >>
> >>
> >
>

Re: [ANNOUNCE] Apache UIMA DUCC 1.0.0 Released

[Lou DeGenaro]

Very good.  I'm all in!

Lou.


On Mon, Feb 3, 2014 at 5:24 PM, Jaroslaw Cwiklik  wrote:

> Yes, both UIMA-AS and DUCC will be hot links to relevant pages in the new
> content. If there are
> no objections I can change the diagram to say Scaleout Frameworks.
>
>
> On Mon, Feb 3, 2014 at 3:23 PM, Lou DeGenaro 
> wrote:
>
> > Doing as Jerry suggests is fine by me.  That said, however...
> >
> > It seems that DUCC deserves equal billing in the diagram by either
> changing
> > UIMA-AS to UIMA-AS & UIMA-DUCC or removing the UIMA-AS as a qualifier.
>  In
> > either case, it is now plural: Scaleout Frameworks, no?  You could get
> > fancy and just say "Scaleout Frameworks" in the diagram but then show
> > UIMA-DUCC and UIMA-AS (links?) when hovering.
> >
> > And I again suggest that some of the text in the changed paragraph be hot
> > links (same as as those in the left sidebar) to the corresponding -AS and
> > -DUCC pages, accordingly.
> >
> > Lou.
> >
> >
> > On Mon, Feb 3, 2014 at 1:13 PM, Jaroslaw Cwiklik 
> wrote:
> >
> > > To review, the proposal is to add a new section to  the left sidebar:
> > > Scaleout Frameworks
> > > and add two links:
> > >
> > >UIMA-AS
> > >DUCC
> > >
> > > We already have dedicated pages for each of the above.
> > >
> > > I can also extend the content (as Lou suggested) to change paragraph
> that
> > > currently reads
> > >
> > > "The UIMA-AS Scaleout Framework<
> > > http://uima.apache.org/doc-uimaas-what.html> is
> > > an add-on to the base Java framework, supporting a very flexible
> scaleout
> > > capability based on JMS (Java Messaging Services) and ActiveMQ."
> > >
> > > to
> > >
> > > "The UIMA-AS and DUCC are both Scaleout Frameworks and are addons to
> the
> > > base Java framework. The UIMA-AS supports
> > > very flexible scaleout capability based on JMS (Java Messaging
> Services)
> > > and ActiveMQ. The UIMA-DUCC extends UIMA-AS
> > > by providing cluster management services to automate the scale-out of
> > UIMA
> > > pipelines over computing clusters."
> > >
> > > DUCC is not mentioned in the diagram so not totally clear if we need to
> > say
> > > more about DUCC and its relation to the whole
> > > picture.
> > >
> > > Comments?
> > >
> > >
> > >
> > >
> > > On Fri, Jan 31, 2014 at 2:38 PM, Jaroslaw Cwiklik 
> > > wrote:
> > >
> > > > I volunteer. Will start working on this next week.
> > > >
> > > >
> > > > On Fri, Jan 31, 2014 at 2:27 PM, Marshall Schor 
> wrote:
> > > >
> > > >>
> > > >> On 1/31/2014 6:19 AM, Richard Eckart de Castilho wrote:
> > > >> > I think the diagram may get a bit overloaded if we add more stuff
> > > >> > like DUCC, uimaFIT and RUTA. I'd suggest to change "UIMA-AS
> Scaleout
> > > >> > Framework" to "Scaleout frameworks" instead to subsume UIMA-AS and
> > > DUCC.
> > > >> > uimaFIT could fall under "Java Framework" and Ruta would fall
> under
> > > >> "Tooling".
> > > >> >
> > > >> > How about setting up dedicated pages for UIMA-AS and UIMA-DUCC, as
> > > they
> > > >> > exist for uimaFIT and RUTA and link them in the sidebar? I would
> > have
> > > >> > liked to link to a dedicated UIMA-DUCC landing page in the
> freecode
> > > >> > announcement.
> > > >>
> > > >> This sounds reasonable to me.  Any volunteers?
> > > >>
> > > >> -Marshall
> > > >>
> > > >> >
> > > >> > -- Richard
> > > >> >
> > > >> > On 31.01.2014, at 12:13, Lou DeGenaro 
> > wrote:
> > > >> >
> > > >> >> Is it premature or could/should the UIMA Home page be revised to
> > > >> include a
> > > >> >> link to DUCC.  One suggestion is to follow the sentence on
> UIMA_AS:
> > > >> >>
> > > >> >> The UIMA-AS Scaleout Framework
> > > >> >> <http://uima.apache.org/doc-uimaas-what.html>is an add-on to the
> > > base
> > > >> >> Java framework, supporting a very flexible
> > > >> >> scaleout capability based on JMS (Java Messaging Services) and
> > > >> ActiveMQ.
> > > >> >>
> > > >> >> with one sentence on DUCC:
> > > >> >>
> > > >> >> DUCC is a cluster management system providing tooling,
> management,
> > > and
> > > >> >> scheduling facilities to automate the scale-out of applications
> > > >> written to
> > > >> >> the UIMA framework.
> > > >> >>
> > > >> >> with "DUCC" linked to
> > > >> http://uima.apache.org/doc-uimaducc-whatitam.html
> > > >> >>
> > > >> >> Also, perhaps DUCC should be integrated into the diagram that
> shows
> > > the
> > > >> >> relationships of the various UIMA pieces-parts in the "What is
> > UIMA?"
> > > >> >> section.
> > > >> >>
> > > >> >> Lou.
> > > >> >
> > > >>
> > > >>
> > > >
> > >
> >
>

Re: [jira] [Commented] (UIMA-3595) Update UIMA Home page to include DUCC

[Lou DeGenaro]

Maybe Scaleout Frameworks should be UIMA Scaleout Frameworks, to keep up
with the neighboring boxes?

Lou.


On Tue, Feb 4, 2014 at 1:24 PM, Jerry Cwiklik (JIRA) wrote:

>
> [
> https://issues.apache.org/jira/browse/UIMA-3595?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13890968#comment-13890968]
>
> Jerry Cwiklik commented on UIMA-3595:
> -
>
> Attached are two images containing changes (not clickable yet). I've
> edited both in old Powerpoint and saved them as png.
>
> First contains changes for Scaleout Frameworks and UIMA FIT (no Ruta).
>
> The second adds UIMA Ruta. This thing (as best as I can tell) belongs to
> tooling (Ruta Workbench) and Annotators (Rute AE). I wasn't sure how
> to capture this so this is my best guess as to where to place it.
>
> Comments?
>
> > Update UIMA Home page to include DUCC
> > -
> >
> > Key: UIMA-3595
> > URL: https://issues.apache.org/jira/browse/UIMA-3595
> > Project: UIMA
> >  Issue Type: Improvement
> >  Components: Website
> >Reporter: Jerry Cwiklik
> >Assignee: Jerry Cwiklik
> >Priority: Minor
> > Attachments: UimaIs.png, UimaIs2.png
> >
> >
> > The proposal is to change the UIMA home page to accommodate newly
> release DUCC.
> > Here are suggested changes:
> > 1) Modify the diagram to replace "UIMA-AS Scaleout Framework" with
> "Scaleout Frameworks".
> > 2) Add new section to the left nav bar with a title: Scaleout
> Frameworks" containing links to both UIMA-AS and DUCC.
> > 3) Change the main content replacing paragraph which currently reads:
> > "The UIMA-AS Scaleout Framework is an add-on to the base Java framework,
> supporting a very flexible scaleout capability based on JMS (Java Messaging
> Services) and ActiveMQ."
> > to
> > "The UIMA-AS and DUCC are both Scaleout Frameworks and are addons to the
> base Java framework. The UIMA-AS supports
> > very flexible scaleout capability based on JMS (Java Messaging Services)
> and ActiveMQ. The UIMA-DUCC extends UIMA-AS
> > by providing cluster management services to automate the scale-out of
> UIMA pipelines over computing clusters."
> > Use hyperlinks in the above text to link to UIMA-AS and DUCC pages.
> > Comments?
>
>
>
> --
> This message was sent by Atlassian JIRA
> (v6.1.5#6160)
>

Re: [jira] [Commented] (UIMA-3595) Update UIMA Home page to include DUCC, Ruta, and UIMA FIT

[Lou DeGenaro]

Jerry, the landing page looks great!  In your copious free time you may
wish to consider making the diagram clickable.

Lou.


On Thu, Feb 6, 2014 at 9:42 AM, Jaroslaw Cwiklik  wrote:

> Thanks to all for the input. The changes are live. Take a look and let me
> know if you see a problem.
>
> http://uima.apache.org/index.html
>
> Jerry
>
>
> On Wed, Feb 5, 2014 at 2:40 PM, Jaroslaw Cwiklik  wrote:
>
> > The diagram originates from PowerPoint and saved as png file. Not sure if
> > I can improve the quality. Perhaps I can save it a jpg.
> > Will play with this to try to improve it. Any reason why this diagram
> > needs to be in png format instead of say jpg?
> >
> >
> > On Wed, Feb 5, 2014 at 10:40 AM, Jens Grivolla 
> wrote:
> >
> >> Just a minor comment: the diagram appears very pixelated, probably
> >> because of low resolution and lack of antialiasing on the text.  Since
> >> you're redoing the diagram this might be a good opportunity to make it a
> >> bit prettier.
> >>
> >> -- Jens
> >>
> >>
> >> On 05/02/14 15:21, Jaroslaw Cwiklik wrote:
> >>
> >>> Great. I will wait till the end of today to give others a chance to
> chime
> >>> in and if there are no objections I will commit my changes and publish
> >>> the
> >>> new home page.
> >>>
> >>>
> >>>
> >>> On Tue, Feb 4, 2014 at 5:30 PM, Marshall Schor (JIRA)
> >>> wrote:
> >>>
> >>>
>   [
>  https://issues.apache.org/jira/browse/UIMA-3595?page=
>  com.atlassian.jira.plugin.system.issuetabpanels:comment-
>  tabpanel&focusedCommentId=13891317#comment-13891317]
> 
>  Marshall Schor commented on UIMA-3595:
>  --
> 
>  nice work, Jerry :-)
> 
>   Update UIMA Home page to include DUCC, Ruta, and UIMA FIT
> > -
> >
> >  Key: UIMA-3595
> >  URL:
> https://issues.apache.org/jira/browse/UIMA-3595
> >  Project: UIMA
> >   Issue Type: Improvement
> >   Components: Website
> > Reporter: Jerry Cwiklik
> > Assignee: Jerry Cwiklik
> > Priority: Minor
> >  Attachments: UimaIs.png, UimaIs2.png, UimaIs3.png
> >
> >
> > The proposal is to change the UIMA home page to accommodate newly
> >
>  release DUCC.
> 
> > Here are suggested changes:
> > 1) Modify the diagram to replace "UIMA-AS Scaleout Framework" with
> >
>  "Scaleout Frameworks".
> 
> > 2) Add UIMA Ruta and UIMA FIT to the diagram
> > 3) Add new section to the left nav bar with a title: Scaleout
> >
>  Frameworks" containing links to both UIMA-AS and DUCC.
> 
> > 4) Change the main content replacing paragraph which currently reads:
> > "The UIMA-AS Scaleout Framework is an add-on to the base Java
> > framework,
> >
>  supporting a very flexible scaleout capability based on JMS (Java
>  Messaging
>  Services) and ActiveMQ."
> 
> > to
> > "The UIMA-AS and DUCC are both Scaleout Frameworks and are addons to
> > the
> >
>  base Java framework. The UIMA-AS supports
> 
> > very flexible scaleout capability based on JMS (Java Messaging
> > Services)
> >
>  and ActiveMQ. The UIMA-DUCC extends UIMA-AS
> 
> > by providing cluster management services to automate the scale-out of
> >
>  UIMA pipelines over computing clusters."
> 
> > Use hyperlinks in the above text to link to UIMA-AS and DUCC pages.
> > Comments?
> >
> 
> 
> 
>  --
>  This message was sent by Atlassian JIRA
>  (v6.1.5#6160)
> 
> 
> >>>
> >>
> >>
> >
>

Re: Board report due in 1 1/2 weeks

[Lou DeGenaro]

Hi Marshall,

Are the web site changes noteworthy enough?  Ditto for the demo DUCC
WebServer bring-up at http://uima-ducc-vm.apache.org:42133/jobs.jsp.

Btw, the official DUCC web page
http://uima.apache.org/doc-uimaducc-whatitam.html says "DUCC is short for
Distributed UIMA Cluster Computing.".

Lou.


On Sat, Mar 1, 2014 at 10:23 AM, Marshall Schor  wrote:

> Hi,
>
> Items for the report that I can put in include:
>
> 1) 2 releases since last report: uima sdk 2.5.0 and ducc 1.0.0
>
> 2) 1 new committer
>
> Ongoing activities:
> Ruta (Rule-based Text Annotation) heading towards next release with a new
> person
> stepping up to serve as release manager.
> DUCC (Distributed UIMA Cluster Controller) continues to have lots of
> activity,
> after its first release.
> Some new work in Core UIMA is focusing on improving operations in highly
> multi-core/multi-threaded deployments.
> And there is the usual ongoing bug reports and fixes.
>
> Anything else we should include in the report to the board?
>
> -Marshall
>

Re: Board report due in 1 1/2 weeks

[Lou DeGenaro]

re: website changes

Yes.

re: "DUCC is short for Distributed UIMA Cluster Computing."

I presume you mean revising the report, not the website.

Lou.


On Mon, Mar 3, 2014 at 10:29 AM, Marshall Schor  wrote:

>
> On 3/3/2014 6:12 AM, Lou DeGenaro wrote:
> > Hi Marshall,
> >
> > Are the web site changes noteworthy enough?  Ditto for the demo DUCC
> > WebServer bring-up at http://uima-ducc-vm.apache.org:42133/jobs.jsp.
>
> Good points.  Re: the web site changes - I presume you mean changing the
> 1st
> (main) page, text and pictures, to reflect the expanding number of things
> associated with UIMA?  The WebServer bringup is also worth a note I think,
> since
> it involved a bit of extra support from Infra.
>
> >
> > Btw, the official DUCC web page
> > http://uima.apache.org/doc-uimaducc-whatitam.html says "DUCC is short
> for
> > Distributed UIMA Cluster Computing.".
>
> Thanks!  I'll fix this.
>
> -Marshall
> >
> > Lou.
> >
> >
> > On Sat, Mar 1, 2014 at 10:23 AM, Marshall Schor  wrote:
> >
> >> Hi,
> >>
> >> Items for the report that I can put in include:
> >>
> >> 1) 2 releases since last report: uima sdk 2.5.0 and ducc 1.0.0
> >>
> >> 2) 1 new committer
> >>
> >> Ongoing activities:
> >> Ruta (Rule-based Text Annotation) heading towards next release with a
> new
> >> person
> >> stepping up to serve as release manager.
> >> DUCC (Distributed UIMA Cluster Controller) continues to have lots of
> >> activity,
> >> after its first release.
> >> Some new work in Core UIMA is focusing on improving operations in highly
> >> multi-core/multi-threaded deployments.
> >> And there is the usual ongoing bug reports and fixes.
> >>
> >> Anything else we should include in the report to the board?
> >>
> >> -Marshall
> >>
>
>

Re: Wrapping up releases: DUCC 1.0 / Ruta 2.2.0 / parent-pom 9

[Lou DeGenaro]

Hi Richard,

Jerry will have a look at the DUCC 1.0 issue.

Lou.


On Fri, Apr 18, 2014 at 2:41 PM, Richard Eckart de Castilho
wrote:

> Hi all,
>
> I just noticed that some past releases have not been marked as "released"
> in Jira:
>
> - DUCC 1.0
> - Ruta 2.2.0
> - parent-pom-9
>
> I took the liberty of marking Ruta 2.2.0 and parent-pom-9 as released and
> of setting their release dates.
>
> @DUCC: DUCC 1.0 still has an open issue. Could please somebody from the
>DUCC team (Lou?) have a look?
>
> @Peter: We also do not yet have a news item for Ruta 2.2.0.
>
> Cheers,
>
> -- Richard

Re: Wrapping up releases: DUCC 1.0 / Ruta 2.2.0 / parent-pom 9

[Lou DeGenaro]

Thanks!


On Mon, Apr 21, 2014 at 11:37 AM, Jaroslaw Cwiklik  wrote:

> Lou, your jira https://issues.apache.org/jira/browse/UIMA-3682 was
> assigned
> to 1-Ducc on 18/Mar/14. Certainly post 1-Ducc release.
> I've changed it 2-Ducc.
> Jerry
>
>
>
> On Mon, Apr 21, 2014 at 9:55 AM, Lou DeGenaro  >wrote:
>
> > Hi Richard,
> >
> > Jerry will have a look at the DUCC 1.0 issue.
> >
> > Lou.
> >
> >
> > On Fri, Apr 18, 2014 at 2:41 PM, Richard Eckart de Castilho
> > wrote:
> >
> > > Hi all,
> > >
> > > I just noticed that some past releases have not been marked as
> "released"
> > > in Jira:
> > >
> > > - DUCC 1.0
> > > - Ruta 2.2.0
> > > - parent-pom-9
> > >
> > > I took the liberty of marking Ruta 2.2.0 and parent-pom-9 as released
> and
> > > of setting their release dates.
> > >
> > > @DUCC: DUCC 1.0 still has an open issue. Could please somebody from the
> > >DUCC team (Lou?) have a look?
> > >
> > > @Peter: We also do not yet have a news item for Ruta 2.2.0.
> > >
> > > Cheers,
> > >
> > > -- Richard
> >
>

Re: Board report due in 1 week and 2 days

[Lou DeGenaro]

With respect to UIMA-DUCC, I would say that:

*DUCC (Distributed UIMA Cluster Computing) continues to have lots of
activity,*
* after its first release.*

continues to be true from last report.  Also, do recent web site changes
regarding DUCC count?  Such as layout changes and additional content for
"Scalable Frameworks" and "DUCC Demo Page/Live"?

Do upcoming releases count?  I believe that both UIMA-AS and UIMA-DUCC
intend to have another release before the next report is due.

Lou.


On Mon, Jun 2, 2014 at 2:55 PM, Marshall Schor  wrote:

> Hi,
>
> Item for the report that I can put in:
>
> -- Items from our website news:  Ruta 2.2.0 and UIMA SDK 2.6.0 releases,
> and the UIMA Workshop at COLING 2014 (maybe I'll not mention this for this
> report - since this event is in August.)
>
> Anything else we should include in the report to the board?
>
> -Marshall
>
>

Re: [VOTE] Release UIMA-DUCC 1.1.0 RC1

[Lou DeGenaro]

Performed the following:

1. svn co into fresh directory
2. mvn clean install -Pbuild-duccdocs
3. followed wiki instructions to install apache demo

Problems:

1. Several Python scripts have improperly inserted apache headers
2.
uima-ducc-examples/src/main/uima-ducc-vm/config/resources/ducc.properties
and ducc.properties.original have incorrect value for ducc.private.resources
3. Folder resources/private is unnecessary?

Lou.

On Mon, Oct 6, 2014 at 3:37 PM, Jaroslaw Cwiklik  wrote:

> Hi,
>
> the Apache UIMA DUCC v1.1.0 RC1 is ready for voting.
>
> DUCC stands for Distributed UIMA Cluster Computing. DUCC is a cluster
> management system providing tooling, management, and scheduling facilities
> to automate the scale-out of applications written to the UIMA framework.
> Core UIMA provides a generalized framework for applications that process
> unstructured information such as human language, but does not provide a
> scale-out mechanism. UIMA-AS provides a scale-out mechanism to distribute
> UIMA pipelines over a cluster of computing resources, but does not provide
> job or cluster management of the resources. DUCC defines a formal job model
> that closely maps to a standard UIMA pipeline. Around this job model DUCC
> provides cluster management services to automate the scale-out of UIMA
> pipelines over computing clusters.
>
> There were 184 issues fixed in this release:
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%221.1.0-Ducc%22%20ORDER%20BY%20key%20ASC
>
> The Maven artifacts are here:
>
> https://repository.apache.org/content/repositories/orgapacheuima-1036/org/apache/uima/
>
> The source and binary zip/tars are here:
> http://people.apache.org/~cwiklik/releases/uima-ducc/1.1.0/RC1/
>
> The SVN Tag Checkout:
>
> http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc-1.1.0/
>
> See: http://uima.apache.org/testing-builds.html for instructions how to
> test release candidate.
>
> Vote open for 72 hours.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ]  0 Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks.
>

Re: [VOTE] Release UIMA-DUCC 1.1.0 RC5

[Lou DeGenaro]

1. svn co https://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/trunk .
2. mvn clean install -Pbuild-duccdocs
3. followed instructions for (re-)installing line DUCC demo
https://cwiki.apache.org/confluence/display/UIMA/DUCC#DUCC-t2

Looks (mostly) good.  Demo is running here:
http://uima-ducc-demo.apache.org:42133/jobs.jsp

I note the following:

a) the name of the produced .gz file in step 1 was:
uima-ducc-2.0.0-SNAPSHOT-bin.tar.gz
b) the Job Details page Processes tab seemingly takes a non-trivial amount
of time to initially display and later update; I've not seen this before or
elsewhere - so for now I presume it is a problem local to the demo host.

Lou.


On Fri, Oct 17, 2014 at 9:43 AM, Jaroslaw Cwiklik  wrote:

> Hi,
>
> the Apache UIMA DUCC v1.1.0 RC5 is ready for voting.
>
> DUCC stands for Distributed UIMA Cluster Computing. DUCC is a cluster
> management system providing tooling, management, and scheduling facilities
> to automate the scale-out of applications written to the UIMA framework.
> Core UIMA provides a generalized framework for applications that process
> unstructured information such as human language, but does not provide a
> scale-out mechanism. UIMA-AS provides a scale-out mechanism to distribute
> UIMA pipelines over a cluster of computing resources, but does not provide
> job or cluster management of the resources. DUCC defines a formal job model
> that closely maps to a standard UIMA pipeline. Around this job model DUCC
> provides cluster management services to automate the scale-out of UIMA
> pipelines over computing clusters.
>
> There were two issues fixed since RC4
>
> - Manually generated JIRA report to include in the release. Maven changes
> plugin only pulls in the first 100 jiras. Ducc 1.1.0  has 189 bugs so about
> half were missing.
>
>  - Minor documentation fixes in ducc book.
>
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%221.1.0-Ducc%22%20ORDER%20BY%20key%20ASC
>
> The Maven artifacts are here:
>
> https://repository.apache.org/content/repositories/orgapacheuima-1040/org/apache/uima/
>
> The source and binary zip/tars are here:
> http://people.apache.org/~cwiklik/releases/uima-ducc/1.1.0/RC5/
>
> The SVN Tag Checkout:
>
> http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc-1.1.0/
>
> See: http://uima.apache.org/testing-builds.html for instructions how to
> test release candidate.
>
> Vote open for 72 hours.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ]  0 Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks.
>

Re: [VOTE] Release UIMA-DUCC 1.1.0 RC5

[Lou DeGenaro]

1. svn co
http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc-1.1.0/
2. mvn clean install -Pbuild-duccdocs
3. followed instructions for (re-)installing live DUCC demo
https://cwiki.apache.org/confluence/display/UIMA/DUCC#DUCC-t2

Looks (mostly) good.  Demo is running here:
http://uima-ducc-demo.apache.org:42133/jobs.jsp

I note the following:

a) the "name problem" reported previously is resolved by specifying the
correct svn co, as Jerry suggests
b) the Job Details page Processes tab seemingly takes a non-trivial amount
of time to initially display and later update; I've not seen this before or
elsewhere - so for now I presume it is a performance problem local to the
demo host.

My (non-binding) vote is: +1 OK to release

Lou.

On Mon, Oct 20, 2014 at 9:52 AM, Jaroslaw Cwiklik  wrote:

> The reason why you get uima-ducc-2.0.0-SNAPSHOT-bin.tar.gz is because you
> checked out from trunk
> instead of RC5 tag. Technically the same code base but the be 100% sure
> just check out from tag and
> redo your testing as we vote on the RC (tag) not whats in the trunk.
>
> Jerry C
>
> On Mon, Oct 20, 2014 at 9:09 AM, Lou DeGenaro 
> wrote:
>
> > 1. svn co https://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/trunk
> .
> > 2. mvn clean install -Pbuild-duccdocs
> > 3. followed instructions for (re-)installing line DUCC demo
> > https://cwiki.apache.org/confluence/display/UIMA/DUCC#DUCC-t2
> >
> > Looks (mostly) good.  Demo is running here:
> > http://uima-ducc-demo.apache.org:42133/jobs.jsp
> >
> > I note the following:
> >
> > a) the name of the produced .gz file in step 1 was:
> > uima-ducc-2.0.0-SNAPSHOT-bin.tar.gz
> > b) the Job Details page Processes tab seemingly takes a non-trivial
> amount
> > of time to initially display and later update; I've not seen this before
> or
> > elsewhere - so for now I presume it is a problem local to the demo host.
> >
> > Lou.
> >
> >
> > On Fri, Oct 17, 2014 at 9:43 AM, Jaroslaw Cwiklik 
> > wrote:
> >
> > > Hi,
> > >
> > > the Apache UIMA DUCC v1.1.0 RC5 is ready for voting.
> > >
> > > DUCC stands for Distributed UIMA Cluster Computing. DUCC is a cluster
> > > management system providing tooling, management, and scheduling
> > facilities
> > > to automate the scale-out of applications written to the UIMA
> framework.
> > > Core UIMA provides a generalized framework for applications that
> process
> > > unstructured information such as human language, but does not provide a
> > > scale-out mechanism. UIMA-AS provides a scale-out mechanism to
> distribute
> > > UIMA pipelines over a cluster of computing resources, but does not
> > provide
> > > job or cluster management of the resources. DUCC defines a formal job
> > model
> > > that closely maps to a standard UIMA pipeline. Around this job model
> DUCC
> > > provides cluster management services to automate the scale-out of UIMA
> > > pipelines over computing clusters.
> > >
> > > There were two issues fixed since RC4
> > >
> > > - Manually generated JIRA report to include in the release. Maven
> changes
> > > plugin only pulls in the first 100 jiras. Ducc 1.1.0  has 189 bugs so
> > about
> > > half were missing.
> > >
> > >  - Minor documentation fixes in ducc book.
> > >
> > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%221.1.0-Ducc%22%20ORDER%20BY%20key%20ASC
> > >
> > > The Maven artifacts are here:
> > >
> > >
> >
> https://repository.apache.org/content/repositories/orgapacheuima-1040/org/apache/uima/
> > >
> > > The source and binary zip/tars are here:
> > > http://people.apache.org/~cwiklik/releases/uima-ducc/1.1.0/RC5/
> > >
> > > The SVN Tag Checkout:
> > >
> > >
> >
> http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc-1.1.0/
> > >
> > > See: http://uima.apache.org/testing-builds.html for instructions how
> to
> > > test release candidate.
> > >
> > > Vote open for 72 hours.
> > >
> > > Please vote on release:
> > >
> > > [ ] +1 OK to release
> > > [ ]  0 Don't care
> > > [ ] -1 Not OK to release, because ...
> > >
> > > Thanks.
> > >
> >
>

Re: Board report due tomorrow - please reply with improvements to this draft

[Lou DeGenaro]

+1

On Wed, Jun 10, 2015 at 5:35 AM, Richard Eckart de Castilho 
wrote:

> +1
>
> -- Richard
>
> On 10.06.2015, at 09:43, Peter Klügl  wrote:
>
> > +1, nothing to add from me side.
> >
> > Peter
> >
> > Am 10.06.2015 um 04:08 schrieb Marshall Schor:
> >> I actually sent a notice of this out earlier, but due to the wonders of
> >> autocompletion, it ended up going to the wrong mailing list...
> >>
> >> The reminder also came out late...
> >>
> >> Here's what I have so far:
> >>
> >> Board report for Apache UIMA, for June 2015.
> >>
> >> Apache UIMA's mission: the creation and maintenance of open-source
> >> software related to the analysis of unstructured data, guided by the
> >> UIMA Oasis Standard.
> >>
> >> Dates:
> >> 08 Jun 2015  (new)last release
> >> 12 Mar 2013  (no change)  last PMC addition
> >> 24 Dec 2013  (no change)  last Committer addition
> >>
> >> Releases:
> >>  UIMA Ruta 2.3.0 released, with better integration with Maven,
> >>  extensions and simplification of the rules language,
> >>  improvements to analysis engines including the HtmlConverter,
> >>  and many bug fixes. http://uima.apache.org/news.html#08 June 2015
> >>
> >> Other Activity:
> >>
> >> A major potential contribution was posted to our Jira, which
> reimplements
> >> much of the "innards" of UIMA core using more standard Java
> constructions,
> >> in an attempt to support better concurrancy on processors with large
> >> numbers (e.g. 64) of cores.  This work is a proof-of-concept, and is
> >> being evaluated.  The community has asked the donor to submit a
> >> software grant so we can take this through IP clearance and put it into
> our
> >> SVN for further study and evaluation.
> >>
> >> This has prompted some new wiki pages documenting various ideas,
> tradeoffs,
> >> and diagrams explaining some of the more subtle points about UIMA
> innards.
> >> These are on our confluence wiki.  As a side note, a big thank you to
> >> infra for the upgrades to the wiki, in support of nice diagramming
> tooling!
> >>
> >> 3 people including 2 PMC members gave UIMA-related talks at Apachecon
> 2015
> >> in Austin. We heard from one person that they attended ApacheCon just
> to be
> >> able to meet with the UIMA team.  One person subsequently has started
> >> contributing to the project.
> >>
> >> Based on an ApacheCon talk which suggested talking with your community
> >> where they like to talk, more people have started responding to UIMA
> >> questions on StackOverflow, a popular website for these kinds of
> questions.
> >>
> >> We have agreed to participate in the documentary video about the ASF;
> >> a video call is planned for later this month.
> >>
> >> Besides UIMA Ruta (which had a release), UIMA DUCC is progressing toward
> >> a next release, and core UIMA development continues to be quite active.
> >>
> >> Community:
> >>
> >> The community continues to be moderately active.  A new person has
> started
> >> committing improvements to one of the core UIMA tools.
> >>
> >> 
> >>
> >> Please reply with improvements, etc!  -Marshall
> >>
> >
>
>

Re: [VOTE] Release UIMA-DUCC 2.0.0 RC1

[Lou DeGenaro]

Before you make the next RC, check the header on the WS pages.  I see:

*Version: 2.0.0-beta *

Lou.

On Tue, Jul 14, 2015 at 11:38 AM, Jaroslaw Cwiklik  wrote:

> Cancelling DUCC 2.0.0 vote. The source tarball README and RELEASE_NOTES
> project version tokens are not being substituted. Should use maven's
> resources plugin to do the replace.
>
> On Tue, Jul 14, 2015 at 9:52 AM, Jaroslaw Cwiklik 
> wrote:
>
> > Hi,
> >
> > the Apache UIMA DUCC v2.0.0 RC1 is ready for voting (previous vote
> > announcement contained wrong link to the source and binary tarballs)
> >
> > DUCC stands for Distributed UIMA Cluster Computing. DUCC is a cluster
> > management system providing tooling, management, and scheduling
> facilities
> > to automate the scale-out of applications written to the UIMA framework.
> > Core UIMA provides a generalized framework for applications that process
> > unstructured information such as human language, but does not provide a
> > scale-out mechanism. UIMA-AS provides a scale-out mechanism to distribute
> > UIMA pipelines over a cluster of computing resources, but does not
> provide
> > job or cluster management of the resources. DUCC defines a formal job
> > model that closely maps to a standard UIMA pipeline. Around this job
> model
> > DUCC provides cluster management services to automate the scale-out of
> > UIMA pipelines over computing clusters.
> >
> > Issues fixed in this release:
> >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.0.0-
> > Ducc%22%20ORDER%20BY%20key%20ASC
> >
> > The Maven artifacts are here:
> >
> >
> https://repository.apache.org/content/repositories/orgapacheuima-1061/org/apache/uima/
> >
> > The source and binary zip/tars are here:
> > http://people.apache.org/~cwiklik/releases/uima-ducc/2.0.0/RC1/
> >
> > The SVN Tag Checkout:
> > http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc
> > -2.0.0/
> >
> > See: http://uima.apache.org/testing-builds.html for instructions how to
> > test release candidate.
> >
> > Vote open for 72 hours.
> >
> > Please vote on release:
> >
> > [ ] +1 OK to release
> > [ ]  0 Don't care
> > [ ] -1 Not OK to release, because ...
> >
> > Thanks.
> >
>

Re: [VOTE] Release UIMA-DUCC 2.0.0 RC3

[Lou DeGenaro]

I have two complaints (so far):

1. When I build from source the resulting zip is:
uima-ducc-2.0.1-SNAPSHOT-bin.tar.gz
2. After installing, a visit to the WS classes page with scroll mode
preference results in: DataTables warning: table id=system-classes -
Invalid JSON response

I will open a Jira for item #2 and deliver a fix shortly.

Lou.

On Thu, Jul 16, 2015 at 5:19 PM, Jaroslaw Cwiklik  wrote:

> Hi,
>
> the Apache UIMA DUCC v2.0.0 RC3 is ready for voting.
>
> Same as RC2 with a few fixes:
> - Disabled maven changes plugin due to its limitation to generate a maximum
> of 100 JIRAs
> - Improved error handling to detect unavailable ports
> - Changed maven compile plugin to use java 1.7 instead of 1.6
> - Fixed JIRA reports for both binary and source distributions
>
>
> Issues fixed in this release:
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.0.0-Ducc%22%20ORDER%20BY%20key%20ASC
>
> The Maven artifacts are here:
>
> https://repository.apache.org/content/repositories/orgapacheuima-1063/org/apache/uima/
>
> The source and binary zip/tars are here:
> http://people.apache.org/~cwiklik/releases/uima-ducc/2.0.0/RC3/
>
> The SVN Tag Checkout:
>
> http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc-2.0.0/
>
> See: http://uima.apache.org/testing-builds.html for instructions how to
> test release candidate.
>
> Vote open for 72 hours.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ]  0 Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks.
>

Re: [VOTE] Release UIMA-DUCC 2.0.0 RC3

[Lou DeGenaro]

I had to maven update all of my eclipse projects to get rid of red
x's...perhaps
that is why I see 2.0.1?

Lou.

On Fri, Jul 17, 2015 at 10:50 AM, Burn Lewis  wrote:

> FWIW when I build from the RC3 source I get the expected
> uima-ducc-2.0.0-bin.tar.gz
>
> On Fri, Jul 17, 2015 at 10:13 AM, Jaroslaw Cwiklik 
> wrote:
>
> > Hi, cancelling DUCC RC3 due to a problem found by Lou.
> > -Jerry
> >
> > On Fri, Jul 17, 2015 at 9:45 AM, Jaroslaw Cwiklik 
> > wrote:
> >
> > > Do you mean when you build from trunk you get
> > > uima-ducc-2.0.1-SNAPSHOT-bin.tar.gz? The 2.0.1 is the *next* version of
> > > DUCC. When a RC is built all poms are updated to the next level and
> > > committed to the svn. If your complaint is about the version 2.0.1
> > instead
> > > of say 2.1.0 that is a different matter. Trivial fix. On the next RC I
> > can
> > > choose 2.1.0 as the next version for DUCC. But this really is not
> related
> > > to the RC. We can change our poms to whatever version we want post
> > release.
> > >
> > > On Fri, Jul 17, 2015 at 8:56 AM, Lou DeGenaro 
> > > wrote:
> > >
> > >> I have two complaints (so far):
> > >>
> > >> 1. When I build from source the resulting zip is:
> > >> uima-ducc-2.0.1-SNAPSHOT-bin.tar.gz
> > >> 2. After installing, a visit to the WS classes page with scroll mode
> > >> preference results in: DataTables warning: table id=system-classes -
> > >> Invalid JSON response
> > >>
> > >> I will open a Jira for item #2 and deliver a fix shortly.
> > >>
> > >> Lou.
> > >>
> > >> On Thu, Jul 16, 2015 at 5:19 PM, Jaroslaw Cwiklik 
> > >> wrote:
> > >>
> > >> > Hi,
> > >> >
> > >> > the Apache UIMA DUCC v2.0.0 RC3 is ready for voting.
> > >> >
> > >> > Same as RC2 with a few fixes:
> > >> > - Disabled maven changes plugin due to its limitation to generate a
> > >> maximum
> > >> > of 100 JIRAs
> > >> > - Improved error handling to detect unavailable ports
> > >> > - Changed maven compile plugin to use java 1.7 instead of 1.6
> > >> > - Fixed JIRA reports for both binary and source distributions
> > >> >
> > >> >
> > >> > Issues fixed in this release:
> > >> >
> > >> >
> > >>
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.0.0-Ducc%22%20ORDER%20BY%20key%20ASC
> > >> >
> > >> > The Maven artifacts are here:
> > >> >
> > >> >
> > >>
> >
> https://repository.apache.org/content/repositories/orgapacheuima-1063/org/apache/uima/
> > >> >
> > >> > The source and binary zip/tars are here:
> > >> > http://people.apache.org/~cwiklik/releases/uima-ducc/2.0.0/RC3/
> > >> >
> > >> > The SVN Tag Checkout:
> > >> >
> > >> >
> > >>
> >
> http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc-2.0.0/
> > >> >
> > >> > See: http://uima.apache.org/testing-builds.html for instructions
> how
> > to
> > >> > test release candidate.
> > >> >
> > >> > Vote open for 72 hours.
> > >> >
> > >> > Please vote on release:
> > >> >
> > >> > [ ] +1 OK to release
> > >> > [ ]  0 Don't care
> > >> > [ ] -1 Not OK to release, because ...
> > >> >
> > >> > Thanks.
> > >> >
> > >>
> > >
> > >
> >
>

Re: [VOTE] Release UIMA-DUCC 2.0.0 RC5

[Lou DeGenaro]

Installed from binary tarball and ran 1.job - OK (see tutorial on DUCC wiki
https://cwiki.apache.org/confluence/display/UIMA/DUCC#DUCC-t0 which
anticipates 2.0.0)
Spot checked License and Notice - OK, but it seems
(webserver)jetty-jsp-2.1-7.4.4.v20110707.jar
should not be listed as it is covered by (webserver)jetty 7.4.4.v20110707
Built from source - OK
Installed from source build on apache-ducc (at RC4) - OK (see http://uima-
ducc-demo.apache.org:42133/jobs.jsp)

[X ] +1 OK to release (non-binding vote)

Lou.

On Wed, Jul 22, 2015 at 9:54 AM, Jaroslaw Cwiklik  wrote:

> Built from source - OK
> Installed from binary tarball and ran 1.job - OK
> JIRA Report - OK
> Spot checked for version 2.0.0 (README, RELEASE_NOTES.html, DuccDocs) - OK
> Checked signatures on both source and binary - OK
> Spot checked License and Notice - OK
>
> [X ] +1 OK to release
>
> On Wed, Jul 22, 2015 at 9:30 AM, Burn Lewis  wrote:
>
> > - Downloaded binary and source and checked signatures
> > - Installed binary on a 25-node cluster and ran some jobs
> > - Built from source and installed on a single-user 12-node cluster with a
> > non-privileged ducc_ling and ran some jobs
> > - Imported source into Eclipse and ran some CLI tests
> > - Spot check licenses, notices, and documentation
> >
> > [X ] +1 OK to release
> >
> > On Tue, Jul 21, 2015 at 4:30 PM, Jaroslaw Cwiklik 
> > wrote:
> >
> > > Hi,
> > >
> > > the Apache UIMA DUCC v2.0.0 RC5 is ready for voting.
> > >
> > > Same as RC4 with a few fixes:
> > > - Fixed Orchestrator ConcurrentModificationException
> > > - Fixed CLI won't accept "debug" as a scheduling class
> > > - Fixed JIRA report sorting
> > > - Fixed DuccSchedulerClasses is not thread-safe
> > > - Fixed README (replaced references to Sun with Oracle)
> > > - Cleaned up DUCC docs (removed chapters with no content)
> > > - Changed permissions in issuesFixed folder
> > > - Changed Agent's skip interval for publishing inventory
> > > - Fixed LICENSE to refer to correct versions on httpcore and httpclient
> > > jars
> > > - Updated build_duccling
> > >
> > > Issues fixed in this release:
> > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.0.0-Ducc%22
> > >
> > > The Maven artifacts are here:
> > >
> > >
> >
> https://repository.apache.org/content/repositories/orgapacheuima-1065/org/apache/uima/
> > >
> > > The source and binary zip/tars are here:
> > > http://people.apache.org/~cwiklik/releases/uima-ducc/2.0.0/RC5/
> > >
> > > The SVN Tag Checkout:
> > >
> > >
> >
> http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc-2.0.0/
> > >
> > > See: http://uima.apache.org/testing-builds.html for instructions how
> to
> > > test release candidate.
> > >
> > > Vote open for 72 hours.
> > >
> > > Please vote on release:
> > >
> > > [ ] +1 OK to release
> > > [ ]  0 Don't care
> > > [ ] -1 Not OK to release, because ...
> > >
> > > Thanks.
> > >
> >
>

Re: [VOTE] Release UIMA-DUCC 2.0.0 RC6

[Lou DeGenaro]

- built from source w/doc
- installed on apache-ducc (demo)
http://uima-ducc-demo.apache.org:42133/jobs.jsp

+1 OK to release (non-binding vote)

Lou.

On Thu, Jul 23, 2015 at 3:49 PM, Jaroslaw Cwiklik  wrote:

> Hi,
>
> the Apache UIMA DUCC v2.0.0 RC6 is ready for voting.
>
> Same as RC5 with a few fixes:
>
> - Added xmlbeans attributions to DUCC NOTICE file
> - Updated RELEASE_NOTES
> - Updated LICENSE ( removed jetty jar )
> - Updated Ducc Docs
> - Fixed WS wrong day of week
> - Fixed WS Specification should not be displayed when user is not logged-in
> - Fixed JIRA report (number of jiras not assigned to 2.0.0-Ducc)
>
> Issues fixed in this release:
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.0.0-Ducc%22
>
> The Maven artifacts are here:
>
> https://repository.apache.org/content/repositories/orgapacheuima-1066/org/apache/uima/
>
> The source and binary zip/tars are here:
> http://people.apache.org/~cwiklik/releases/uima-ducc/2.0.0/RC6/
>
> The SVN Tag Checkout:
>
> http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc-2.0.0/
>
> See: http://uima.apache.org/testing-builds.html for instructions how to
> test release candidate.
>
> Vote open for 72 hours.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ]  0 Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks.
>

Re: [VOTE] Release UIMA-DUCC 2.0.0 RC7

[Lou DeGenaro]

- built from source
- followed instructions on wiki [https://cwiki.apache
.org/confluence/display/UIMA/DUCC#DUCC-t2] to deploy demo [http://uima-ducc
-demo.apache.org:42133/jobs.jsp]
- spot check WebServer links
- spot check LICENCE, NOTICE, RELEASE_NOTES.html

+1 OK to release

Lou.

On Mon, Aug 3, 2015 at 6:28 PM, Jaroslaw Cwiklik  wrote:

> Hi,
>
> the Apache UIMA DUCC v2.0.0 RC7 is ready for voting.
>
> Same as RC6 with a few fixes:
>
> - Fixed file permissions in apache-uima folder
> - Fixed TimeInit window which showed AE in initializing state when JP was
> in Running state
> - Updated Ducc Docs
> - Fixed default properties to enable Cgroups
> - use ducc.orchestrator.state.publish.rate instead of deprecated
> ducc.orchestrator.abbreciated.state.publish.rate
> - Correct  calculation of allotment for non-preemptble shares
>
> Issues fixed in this release:
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.0.0-Ducc%22
>
> The Maven artifacts are here:
>
> https://repository.apache.org/content/repositories/orgapacheuima-1068/org/apache/uima/
>
> The source and binary zip/tars are here:
> http://people.apache.org/~cwiklik/releases/uima-ducc/2.0.0/RC7/
>
> The SVN Tag Checkout:
>
> http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc-2.0.0/
>
> See: http://uima.apache.org/testing-builds.html for instructions how to
> test release candidate.
>
> Vote open for 72 hours.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ]  0 Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks.
> -jerry
>

Re: Running UIMA Apps on Hadoop

[Lou DeGenaro]

Perhaps you have no choice, but if you do then consider DUCC https://uima.
apache.org/doc-uimaducc-whatitam.html for UIMA scale-out.

Lou.

On Fri, Oct 9, 2015 at 5:54 PM, Yi-Wen Liu  wrote:

> Hi all,
>
> I am a USC student working on professor Mattmann's project "Integration of
> cTAKES/UIMA and Apache Hadoop for the Shangridocs system", the proposal is
> attached.
> I searched many relevant resources of running UIMA on Hadoop, this is one
> of them:
>
> https://cwiki.apache.org/confluence/display/UIMA/Running+UIMA+Apps+on+Hadoop
>
> but it only provides very general explanations.
>
> Has anybody gone through all the steps and succeeded? Hope you could
> provide me some examples. Thanks!
>
> And in the *Important Considerations*,
> "1. The jar file created should *shave* all the classes, descriptors of
> the UIMA app along with the map/reduce and job main class"
>
> I think you mean "The jar file created should *have* all the classes...".
> Is that correct?
>
>
> Thanks,
> Yi-Wen
>

Re: [VOTE] Release UIMA-DUCC 2.0.1 RC2

[Lou DeGenaro]

* fetched tar.gz from link
* employed approximate instructions on wiki
[https://cwiki.apache.org/confluence/display/UIMA/DUCC#DUCC-t2] to
deploy demo [http://uima-ducc-demo.apache.org:42133/jobs.jsp]
* spot check WebServer links

  DUCCBOOK says: Version 2.0.0

* spot check LICENCE, NOTICE, RELEASE_NOTES.html

-1 Not OK to release, because DUCCBOOK says: Version 2.0.0 but should be 2.1.0

Lou.


On Wed, Oct 14, 2015 at 1:50 PM, Jaroslaw Cwiklik  wrote:

> Hi,
>
> the Apache UIMA DUCC v2.0.1 RC2 is ready for voting.
>
> Issues fixed since RC1:
>
> - UIMA-4578 Agent should kill children
> - UIMA-4585 Fix %CPU when JP has child procs & fix for FD leak in agent
> - UIMA-4630 Fix for DD perf metrics not in current UIMA-AS
> - UIMA-4588 - JP JD AP should not listen on admin channel
> - UIMA-4608 Disable automatic svn updates
>
> DUCC stands for Distributed UIMA Cluster Computing. DUCC is a cluster
> management system providing tooling, management, and scheduling facilities
> to automate the scale-out of applications written to the UIMA framework.
> Core UIMA provides a generalized framework for applications that process
> unstructured information such as human language, but does not provide a
> scale-out mechanism. UIMA-AS provides a scale-out mechanism to distribute
> UIMA pipelines over a cluster of computing resources, but does not provide
> job or cluster management of the resources. DUCC defines a formal job model
> that closely maps to a standard UIMA pipeline. Around this job model DUCC
> provides cluster management services to automate the scale-out of UIMA
> pipelines over computing clusters.
>
> Issues fixed in this release:
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.0.1-Ducc%22%20ORDER%20BY%20key%20ASC
>
> The Maven artifacts are here:
>
> https://repository.apache.org/content/repositories/orgapacheuima-1076/org/apache/uima/
>
> The source and binary zip/tars are here:
> http://people.apache.org/~cwiklik/releases/uima-ducc/2.0.1/RC2/
>
> The SVN Tag Checkout:
>
> http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc-2.0.1/
>
> See: http://uima.apache.org/testing-builds.html for instructions how to
> test release candidate.
>
> Vote open for 72 hours.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ]  0 Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks.
>

Re: [VOTE] Release UIMA-DUCC 2.0.1 RC2

[Lou DeGenaro]

oops: should be 2.0.1.

Lou.

On Thu, Oct 15, 2015 at 11:04 AM, Lou DeGenaro 
wrote:

> * fetched tar.gz from link
> * employed approximate instructions on wiki 
> [https://cwiki.apache.org/confluence/display/UIMA/DUCC#DUCC-t2] to deploy 
> demo [http://uima-ducc-demo.apache.org:42133/jobs.jsp]
> * spot check WebServer links
>
>   DUCCBOOK says: Version 2.0.0
>
> * spot check LICENCE, NOTICE, RELEASE_NOTES.html
>
> -1 Not OK to release, because DUCCBOOK says: Version 2.0.0 but should be 2.1.0
>
> Lou.
>
>
> On Wed, Oct 14, 2015 at 1:50 PM, Jaroslaw Cwiklik 
> wrote:
>
>> Hi,
>>
>> the Apache UIMA DUCC v2.0.1 RC2 is ready for voting.
>>
>> Issues fixed since RC1:
>>
>> - UIMA-4578 Agent should kill children
>> - UIMA-4585 Fix %CPU when JP has child procs & fix for FD leak in agent
>> - UIMA-4630 Fix for DD perf metrics not in current UIMA-AS
>> - UIMA-4588 - JP JD AP should not listen on admin channel
>> - UIMA-4608 Disable automatic svn updates
>>
>> DUCC stands for Distributed UIMA Cluster Computing. DUCC is a cluster
>> management system providing tooling, management, and scheduling facilities
>> to automate the scale-out of applications written to the UIMA framework.
>> Core UIMA provides a generalized framework for applications that process
>> unstructured information such as human language, but does not provide a
>> scale-out mechanism. UIMA-AS provides a scale-out mechanism to distribute
>> UIMA pipelines over a cluster of computing resources, but does not provide
>> job or cluster management of the resources. DUCC defines a formal job
>> model
>> that closely maps to a standard UIMA pipeline. Around this job model DUCC
>> provides cluster management services to automate the scale-out of UIMA
>> pipelines over computing clusters.
>>
>> Issues fixed in this release:
>>
>> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.0.1-Ducc%22%20ORDER%20BY%20key%20ASC
>>
>> The Maven artifacts are here:
>>
>> https://repository.apache.org/content/repositories/orgapacheuima-1076/org/apache/uima/
>>
>> The source and binary zip/tars are here:
>> http://people.apache.org/~cwiklik/releases/uima-ducc/2.0.1/RC2/
>>
>> The SVN Tag Checkout:
>>
>> http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc-2.0.1/
>>
>> See: http://uima.apache.org/testing-builds.html for instructions how to
>> test release candidate.
>>
>> Vote open for 72 hours.
>>
>> Please vote on release:
>>
>> [ ] +1 OK to release
>> [ ]  0 Don't care
>> [ ] -1 Not OK to release, because ...
>>
>> Thanks.
>>
>
>

Re: [VOTE] Release UIMA-DUCC 2.0.1 RC3

[Lou DeGenaro]

* fetched tar.gz from link
* employed approximate instructions on wiki
[https://cwiki.apache.org/confluence/display/UIMA/DUCC#DUCC-t2] to
deploy demo [http://uima-ducc-demo.apache.org:42133/jobs.jsp]
* spot check WebServer links
* spot check LICENSE, NOTICE, RELEASE_NOTES.html

+1 OK to release

Lou.


On Thu, Oct 15, 2015 at 2:04 PM, Jaroslaw Cwiklik  wrote:

> Hi,
>
> the Apache UIMA DUCC v2.0.1 RC3 is ready for voting.
>
> Issues fixed since RC2:
>
> - fixed wrong Ducc version in Ducc docs (pdf and html)
>
> DUCC stands for Distributed UIMA Cluster Computing. DUCC is a cluster
> management system providing tooling, management, and scheduling facilities
> to automate the scale-out of applications written to the UIMA framework.
> Core UIMA provides a generalized framework for applications that process
> unstructured information such as human language, but does not provide a
> scale-out mechanism. UIMA-AS provides a scale-out mechanism to distribute
> UIMA pipelines over a cluster of computing resources, but does not provide
> job or cluster management of the resources. DUCC defines a formal job model
> that closely maps to a standard UIMA pipeline. Around this job model DUCC
> provides cluster management services to automate the scale-out of UIMA
> pipelines over computing clusters.
>
> Issues fixed in this release:
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.0.1-Ducc%22%20ORDER%20BY%20key%20ASC
>
> The Maven artifacts are here:
>
> https://repository.apache.org/content/repositories/orgapacheuima-1077/org/apache/uima/
>
> The source and binary zip/tars are here:
> http://people.apache.org/~cwiklik/releases/uima-ducc/2.0.1/RC3/
>
> The SVN Tag Checkout:
>
> http://svn.apache.org/repos/asf/uima/sandbox/uima-ducc/tags/uima-ducc-2.0.1/
>
> See: http://uima.apache.org/testing-builds.html for instructions how to
> test release candidate.
>
> Vote open for 72 hours.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ]  0 Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks.
>

Re: uima ducc installation

[Lou DeGenaro]

Yi-Wen,

Do you have a firewall blocking ports?

Is the port already in-use by another process?

Anything interesting in $DUCC_HOME/apache-uima/apache-activemq/data/activemg
.log?

Lou.


On Tue, Oct 20, 2015 at 2:28 AM, Yi-Wen Liu  wrote:

> Hi,
>
> I was trying to install UIMA DUCC 2.0.0 on my machine (follow
> https://cwiki.apache.org/confluence/display/UIMA/DUCC), and came across
> some problems(ActiveMQ), hope somebody can help, thanks!
>
>
> The ducc_post_install and ducc_check look fine, for ducc_post_install I
> got:
> ActiveMQ is automanaged on node  yiwenlius-mbp
> ...
> ducc_ling is installed for architecture x86_64. See the installation guide
> for additional modifications needed to run user processes with their OS
> user credentials.
> Initial DUCC setup complete.
>
>
> and results from ducc_check which looks very similar to the doc:
> JVM: /usr/bin/java
> ENV: Java is configured as: /usr/bin/java
> ENV java full version "1.8.0_60-b27"
> ENV: Threading enabled: True
> ENV: system is Darwin
> ActiveMQ broker is not running on tcp://yiwenlius-mbp:61617
> Checking 1 nodes
> Waiting for Completion
> Checking yiwenlius-mbp ...
> no processes found.
> Node health checks return.
> All threads returned
>
>
> The reason why I point out activeMQ is that when I do start_ducc, it seems
> the broker is not running:
>
> allnodes /Users/ducc/apache-uima-ducc-2.0.0/resources/ducc.nodes
> Class definition file is ducc.classes
> OK: Class and node definitions validated.
> OK: Class configuration checked
> Starting broker on yiwenlius-mbp
> Waiting for broker . 0
> Waiting for broker . 1
> Waiting for broker . 2
> Waiting for broker . 3
> Waiting for broker . 4
> Waiting for broker . 5
> Waiting for broker . 6
> Waiting for broker . 7
> Waiting for broker . 8
> ActiveMQ broker cannot be found on yiwenlius-mbp:61617
>
>
> And it stopped.
>
> Does somebody know what it is going on and how to fix it?
>
>
> Thanks,
> Yi-Wen
>

Re: uima ducc installation

[Lou DeGenaro]

>From $DUCC_HOME/admin/start_ducc here's how the broker is started:

 41 def start_broker(self):
 42
 43 broker_host = self.ducc_properties.get('ducc.broker.hostname')
 44 print 'Starting broker on', broker_host
 45 lines = self.ssh(broker_host, True, "'", self.DUCC_HOME +
'/admin/ducc.py', '-c', 'broker', "'")

Is ducc.broker.hostname defined correctly in $DUCC_HOME/resources/ducc
.properties?

Can you ssh to that host?  You may need to set up password-less ssh if you
haven't already.

Lou.

On Tue, Oct 20, 2015 at 5:10 PM, Yi-Wen Liu  wrote:

> No I didn't turn on firewall, and no process is using 61617.
>
> However in $DUCC_HOME/apache-uima/apache-activemq, I don't have the
> activemq.log and even the data directory.
> I checked other folders but didn't find activemq.log, does that mean I
> probably missed some steps?
>
> Thanks,
> Yi-Wen
>
> On Tue, Oct 20, 2015 at 4:36 AM, Lou DeGenaro 
> wrote:
>
> > Yi-Wen,
> >
> > Do you have a firewall blocking ports?
> >
> > Is the port already in-use by another process?
> >
> > Anything interesting in
> > $DUCC_HOME/apache-uima/apache-activemq/data/activemg
> > .log?
> >
> > Lou.
> >
> >
> > On Tue, Oct 20, 2015 at 2:28 AM, Yi-Wen Liu  wrote:
> >
> > > Hi,
> > >
> > > I was trying to install UIMA DUCC 2.0.0 on my machine (follow
> > > https://cwiki.apache.org/confluence/display/UIMA/DUCC), and came
> across
> > > some problems(ActiveMQ), hope somebody can help, thanks!
> > >
> > >
> > > The ducc_post_install and ducc_check look fine, for ducc_post_install I
> > > got:
> > > ActiveMQ is automanaged on node  yiwenlius-mbp
> > > ...
> > > ducc_ling is installed for architecture x86_64. See the installation
> > guide
> > > for additional modifications needed to run user processes with their OS
> > > user credentials.
> > > Initial DUCC setup complete.
> > >
> > >
> > > and results from ducc_check which looks very similar to the doc:
> > > JVM: /usr/bin/java
> > > ENV: Java is configured as: /usr/bin/java
> > > ENV java full version "1.8.0_60-b27"
> > > ENV: Threading enabled: True
> > > ENV: system is Darwin
> > > ActiveMQ broker is not running on tcp://yiwenlius-mbp:61617
> > > Checking 1 nodes
> > > Waiting for Completion
> > > Checking yiwenlius-mbp ...
> > > no processes found.
> > > Node health checks return.
> > > All threads returned
> > >
> > >
> > > The reason why I point out activeMQ is that when I do start_ducc, it
> > seems
> > > the broker is not running:
> > >
> > > allnodes /Users/ducc/apache-uima-ducc-2.0.0/resources/ducc.nodes
> > > Class definition file is ducc.classes
> > > OK: Class and node definitions validated.
> > > OK: Class configuration checked
> > > Starting broker on yiwenlius-mbp
> > > Waiting for broker . 0
> > > Waiting for broker . 1
> > > Waiting for broker . 2
> > > Waiting for broker . 3
> > > Waiting for broker . 4
> > > Waiting for broker . 5
> > > Waiting for broker . 6
> > > Waiting for broker . 7
> > > Waiting for broker . 8
> > > ActiveMQ broker cannot be found on yiwenlius-mbp:61617
> > >
> > >
> > > And it stopped.
> > >
> > > Does somebody know what it is going on and how to fix it?
> > >
> > >
> > > Thanks,
> > > Yi-Wen
> > >
> >
>

Re: uima ducc installation

[Lou DeGenaro]

re: PWC6033: Unable to compile class for JSP

In $DUCC_HOME/resources/ducc.properties you can find this:

# Specify the ducc jsp comp dir
ducc.ws.jsp.compilation.directory = /tmp/ducc/jsp

Your choices are:

1. insure that your "ducc" user has r/w access to that directory
2. make an entry in site.ducc.properties (NOT ducc.properties!!!) with a
different directory that is r/w by user "ducc"

Lou.

On Thu, Oct 22, 2015 at 3:55 AM, Yi-Wen Liu  wrote:

> Hi,
>
> Thanks for those helps! They work :) Now I am able to submit job file.
> When I submitted the job, it was always showing WaitingForDriver, and in
> the $DUCC_HOME/logs/ws..log file I got multiple
>
> 21 Oct 2015 22:09:47,133  INFO WS.DuccHandlerProxy - N/A
> handleServletJobStatus  [GET
> /ducc-servlet/proxy-job-status?id=3]@1627539128
> org.eclipse.jetty.server.Request@610246b8
>
> and the job status didn't move forward to Waiting For Services, Waiting For
> Resources, Initializing etc, it was waiting for the driver like forever.
>
> I tried to check DUCC web server, which should be at http://
> :42133
> from the docs.
> However I got this:
> HTTP ERROR: 500
>
> Problem accessing /jobs.jsp. Reason:
>
> PWC6033: Unable to compile class for JSP
>
> PWC6197: An error occurred at line: 24 in the jsp file: /$banner/$image.jsp
> PWC6199: Generated servlet error:
> The type java.util.Map$Entry cannot be resolved. It is indirectly
> referenced from required .class files
>
>
> Do I need one more step to browse this page, or should start the driver
> first?
>
>
> Thanks,
> Yi-Wen
>
> On Wed, Oct 21, 2015 at 6:31 AM, Burn Lewis  wrote:
>
> > Are you starting DUCC from the head node?
> > You could try starting just the broker with "./ducc.py -c broker" and see
> > if any error messages appear on the console.
> > Or try starting a default broker using the AMQ commands directly, e.g. cd
> > apache-uima/apache-activemq; ./bin/activemq start
> > This should produce an error in the logs about a missing activemq.xml,
> but
> > would at least show that AMQ is installed.  AMQ creates the data
> directory
> > and logs when the broker starts, and also saves the pid.
> >
> > ~Burn
> >
> > On Wed, Oct 21, 2015 at 6:12 AM, Lou DeGenaro 
> > wrote:
> >
> > > From $DUCC_HOME/admin/start_ducc here's how the broker is started:
> > >
> > >  41 def start_broker(self):
> > >  42
> > >  43 broker_host =
> > self.ducc_properties.get('ducc.broker.hostname')
> > >  44 print 'Starting broker on', broker_host
> > >  45 lines = self.ssh(broker_host, True, "'", self.DUCC_HOME +
> > > '/admin/ducc.py', '-c', 'broker', "'")
> > >
> > > Is ducc.broker.hostname defined correctly in $DUCC_HOME/resources/ducc
> > > .properties?
> > >
> > > Can you ssh to that host?  You may need to set up password-less ssh if
> > you
> > > haven't already.
> > >
> > > Lou.
> > >
> > > On Tue, Oct 20, 2015 at 5:10 PM, Yi-Wen Liu  wrote:
> > >
> > > > No I didn't turn on firewall, and no process is using 61617.
> > > >
> > > > However in $DUCC_HOME/apache-uima/apache-activemq, I don't have the
> > > > activemq.log and even the data directory.
> > > > I checked other folders but didn't find activemq.log, does that mean
> I
> > > > probably missed some steps?
> > > >
> > > > Thanks,
> > > > Yi-Wen
> > > >
> > > > On Tue, Oct 20, 2015 at 4:36 AM, Lou DeGenaro <
> lou.degen...@gmail.com>
> > > > wrote:
> > > >
> > > > > Yi-Wen,
> > > > >
> > > > > Do you have a firewall blocking ports?
> > > > >
> > > > > Is the port already in-use by another process?
> > > > >
> > > > > Anything interesting in
> > > > > $DUCC_HOME/apache-uima/apache-activemq/data/activemg
> > > > > .log?
> > > > >
> > > > > Lou.
> > > > >
> > > > >
> > > > > On Tue, Oct 20, 2015 at 2:28 AM, Yi-Wen Liu 
> > wrote:
> > > > >
> > > > > > Hi,
> > > > > >
> > > > > > I was trying to install UIMA DUCC 2.0.0 on my machine (follow
> > > > > > https://cwiki.apache.org/confluence/display/UIMA/DUCC), and came
> > > > across
> > &g

Re: uima ducc installation

[Lou DeGenaro]

I just tried using Java 8 and I observe the following in the WebServer's
log file:

=

Problem accessing /system.daemons.jsp. Reason:

PWC6033: Unable to compile class for JSP

PWC6197: An error occurred at line: 24 in the jsp file:
/$banner/$testpattern.jsp
PWC6199: Generated servlet error:
The type java.util.Map$Entry cannot be resolved. It is indirectly
referenced from required .class files
=

Please try Java 7.  The Java used is specified in
$DUCC_HOME/resources/site.ducc.properties, for example:

ducc.jvm = /share/ibm-jdk1.7/bin/java

Lou.





On Thu, Oct 22, 2015 at 3:59 PM, Burn Lewis  wrote:

> Being stuck at WaitingForDriver may mean that the Agent nodes have not come
> online so the RM is not able to assign resources.
> Your "WS.DuccHandlerProxy" error is probably because the webserver has
> failed to start
> I have not been able to reproduce your compile jsp problem by specifying an
> inaccessible directory, but that is certainly worth checking.
>
> Also have you run ./check_ducc and verified that all the daemons are
> running on the head node (or ws pm rm sm & maybe an agent)
>
> ~Burn
>
>
> On Thu, Oct 22, 2015 at 7:50 AM, Lou DeGenaro 
> wrote:
>
> > re: PWC6033: Unable to compile class for JSP
> >
> > In $DUCC_HOME/resources/ducc.properties you can find this:
> >
> > # Specify the ducc jsp comp dir
> > ducc.ws.jsp.compilation.directory = /tmp/ducc/jsp
> >
> > Your choices are:
> >
> > 1. insure that your "ducc" user has r/w access to that directory
> > 2. make an entry in site.ducc.properties (NOT ducc.properties!!!) with a
> > different directory that is r/w by user "ducc"
> >
> > Lou.
> >
> > On Thu, Oct 22, 2015 at 3:55 AM, Yi-Wen Liu  wrote:
> >
> > > Hi,
> > >
> > > Thanks for those helps! They work :) Now I am able to submit job file.
> > > When I submitted the job, it was always showing WaitingForDriver, and
> in
> > > the $DUCC_HOME/logs/ws..log file I got multiple
> > >
> > > 21 Oct 2015 22:09:47,133  INFO WS.DuccHandlerProxy - N/A
> > > handleServletJobStatus  [GET
> > > /ducc-servlet/proxy-job-status?id=3]@1627539128
> > > org.eclipse.jetty.server.Request@610246b8
> > >
> > > and the job status didn't move forward to Waiting For Services, Waiting
> > For
> > > Resources, Initializing etc, it was waiting for the driver like
> forever.
> > >
> > > I tried to check DUCC web server, which should be at http://
> > > :42133
> > > from the docs.
> > > However I got this:
> > > HTTP ERROR: 500
> > >
> > > Problem accessing /jobs.jsp. Reason:
> > >
> > > PWC6033: Unable to compile class for JSP
> > >
> > > PWC6197: An error occurred at line: 24 in the jsp file:
> > /$banner/$image.jsp
> > > PWC6199: Generated servlet error:
> > > The type java.util.Map$Entry cannot be resolved. It is indirectly
> > > referenced from required .class files
> > >
> > >
> > > Do I need one more step to browse this page, or should start the driver
> > > first?
> > >
> > >
> > > Thanks,
> > > Yi-Wen
> > >
> > > On Wed, Oct 21, 2015 at 6:31 AM, Burn Lewis 
> wrote:
> > >
> > > > Are you starting DUCC from the head node?
> > > > You could try starting just the broker with "./ducc.py -c broker" and
> > see
> > > > if any error messages appear on the console.
> > > > Or try starting a default broker using the AMQ commands directly,
> e.g.
> > cd
> > > > apache-uima/apache-activemq; ./bin/activemq start
> > > > This should produce an error in the logs about a missing
> activemq.xml,
> > > but
> > > > would at least show that AMQ is installed.  AMQ creates the data
> > > directory
> > > > and logs when the broker starts, and also saves the pid.
> > > >
> > > > ~Burn
> > > >
> > > > On Wed, Oct 21, 2015 at 6:12 AM, Lou DeGenaro <
> lou.degen...@gmail.com>
> > > > wrote:
> > > >
> > > > > From $DUCC_HOME/admin/start_ducc here's how the broker is started:
> > > > >
> > > > >  41 def start_broker(self):
> > > > >  42
> > > > >  43 broker_host =
> > > > self.ducc_properties.get('ducc.broker.hostname')
> > > > >  44 print 'Starting broker on', broker_host
> > > > >  45

Re: run fails

[Lou DeGenaro]

Yi-Wen,

The DUCC job during runtime comprises two parts, one Job Driver and one or
more Job Processes.  The log file snippit you posted comprises the latter.
Have you examined the Job Driver log file for any errors?  The Job Driver
log file is found in the same directory and starts with "JD-".

Also I believe you said in an earlier post that you have successfully run
one of the sample jobs, such as 1.job.  Is that correct?

Lou.

On Wed, Nov 11, 2015 at 1:47 AM, Yi-Wen Liu  wrote:

> Hello,
>
> When I ran the project on DUCC, it finally reached "running" but then it
> showed "MonitorWarning", and canceled by driver.
> Here's the process info:
>
> ​And log:
>
> 10 Nov 2015 22:30:49,643  INFO HttpWorkerThread - T[33] 
> HttpWorkerThread.run()  Begin Processing Work Items - Thread Id:33
> 10 Nov 2015 22:30:49,644  INFO HttpWorkerThread - T[34] 
> HttpWorkerThread.run()  Begin Processing Work Items - Thread Id:34
> 10 Nov 2015 22:31:23,270 ERROR DuccHttpClient - T[34] run
> java.net.ConnectException: Connection refused
>   at java.net.PlainSocketImpl.socketConnect(Native Method)
>   at 
> java.net.AbstractPlainSocketImpl.doConnect(AbstractPlainSocketImpl.java:339)
>   at 
> java.net.AbstractPlainSocketImpl.connectToAddress(AbstractPlainSocketImpl.java:200)
>   at 
> java.net.AbstractPlainSocketImpl.connect(AbstractPlainSocketImpl.java:182)
>   at java.net.SocksSocketImpl.connect(SocksSocketImpl.java:392)
>   at java.net.Socket.connect(Socket.java:579)
>   at java.net.Socket.connect(Socket.java:528)
>   at java.net.Socket.(Socket.java:425)
>   at java.net.Socket.(Socket.java:280)
>   at 
> org.apache.commons.httpclient.protocol.DefaultProtocolSocketFactory.createSocket(DefaultProtocolSocketFactory.java:80)
>   at 
> org.apache.commons.httpclient.protocol.DefaultProtocolSocketFactory.createSocket(DefaultProtocolSocketFactory.java:122)
>   at 
> org.apache.commons.httpclient.HttpConnection.open(HttpConnection.java:707)
>   at 
> org.apache.commons.httpclient.MultiThreadedHttpConnectionManager$HttpConnectionAdapter.open(MultiThreadedHttpConnectionManager.java:1361)
>   at 
> org.apache.commons.httpclient.HttpMethodDirector.executeWithRetry(HttpMethodDirector.java:387)
>   at 
> org.apache.commons.httpclient.HttpMethodDirector.executeMethod(HttpMethodDirector.java:171)
>   at 
> org.apache.commons.httpclient.HttpClient.executeMethod(HttpClient.java:397)
>   at 
> org.apache.commons.httpclient.HttpClient.executeMethod(HttpClient.java:323)
>   at 
> org.apache.uima.ducc.transport.configuration.jp.DuccHttpClient.execute(DuccHttpClient.java:217)
>   at 
> org.apache.uima.ducc.transport.configuration.jp.HttpWorkerThread.run(HttpWorkerThread.java:168)
>   at 
> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)
>   at java.util.concurrent.FutureTask.run(FutureTask.java:262)
>   at 
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
>   at 
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
>   at 
> org.apache.uima.ducc.transport.configuration.jp.UimaServiceThreadFactory$1.run(UimaServiceThreadFactory.java:85)
>   at java.lang.Thread.run(Thread.java:745)
> 10 Nov 2015 22:31:23,276 ERROR DuccHttpClient - T[34] run
> java.net.ConnectException: Connection refused
>   at java.net.PlainSocketImpl.socketConnect(Native Method)
>   at 
> java.net.AbstractPlainSocketImpl.doConnect(AbstractPlainSocketImpl.java:339)
>   at 
> java.net.AbstractPlainSocketImpl.connectToAddress(AbstractPlainSocketImpl.java:200)
>   at 
> java.net.AbstractPlainSocketImpl.connect(AbstractPlainSocketImpl.java:182)
>   at java.net.SocksSocketImpl.connect(SocksSocketImpl.java:392)
>   at java.net.Socket.connect(Socket.java:579)
>   at java.net.Socket.connect(Socket.java:528)
>   at java.net.Socket.(Socket.java:425)
>
> I have no idea why the connection got refused, since when I submitted
> other test job it worked well.
>
> I tried to debug using all_in_one local/remote, but it didn't have error
> which I thought it was working, however there's no output...
>
> Could it be any privilege setting problem? Really hope somebody can help
> me out, I didn't find other keyword useful in log file, the connection
> refused error showed up 12 times in total.
>
> Thanks,
> Yi-Wen
>
>

Re: run fails but has output

[Lou DeGenaro]

Yi-Wen,

The DUCC Job first visits the Initialization state.  This is when the Job
Driver (JD) initializes your UIMA CollecitonReader (CR). The CR that you
specified in your DUCC Job submission is constructed and its getTotal()
method is invoked to determine the number of work items.  But I think your
Job made it past the Initialization state and into the Running state.
Looking at the log snippits that you posted, I see:

16 Nov 2015 23:34:38,353  INFO JobDriver - T[18] advanceJdState
current=Initializing request=Active result=Active

This means the JD has exited its Initializing state and is now Active and
is fetching work items.  Further evidence is the entry:

16 Nov 2015 23:34:39,066 ERROR CasManager - T[17] getMetaCas

What do the contents of the file ducc.log in your Job's log directory
show?  You should see "Initializing" followed by "Running".

It looks like the root cause of your problem is that your CR's getNext() is
trying to produce work items for the JD but is failing:

org.xml.sax.SAXParseException; Trying to serialize non-XML 1.0
character: , 0x0
at org.apache.uima.util.XMLSerializer$CharacterValidatingContentHand
ler.checkForInvalidXmlChars(XMLSerializer.java:254)

I recommend determining why your CR is broken, fixing, and re-submitting
your Job to DUCC.

Lou.

On Tue, Nov 17, 2015 at 2:46 AM, Yi-Wen Liu  wrote:

> Hello,
>
> I am still stuck in the following error of DUCC and job always got canceled
> by driver:
>
> 16 Nov 2015 23:34:38,353  INFO JobDriver - T[18] advanceJdState
> current=Initializing request=Active result=Active
> 16 Nov 2015 23:34:39,066 ERROR ProxyLogger - T[17] loggify
> org.xml.sax.SAXParseException; Trying to serialize non-XML 1.0
> character: , 0x0
> at
> org.apache.uima.util.XMLSerializer$CharacterValidatingContentHandler.checkForInvalidXmlChars(XMLSerializer.java:254)
>
> 16 Nov 2015 23:34:39,066 ERROR CasManager - T[17] getMetaCas
> org.apache.uima.ducc.container.common.classloader.ProxyException
> at
> org.apache.uima.ducc.container.jd.classload.ProxyJobDriverCollectionReader.getMetaCas(ProxyJobDriverCollectionReader.java:127)
> at
> org.apache.uima.ducc.container.jd.cas.CasManager.getMetaCas(CasManager.java:65)
> at
> org.apache.uima.ducc.container.jd.fsm.wi.ActionGet.engage(ActionGet.java:118)
> at
> org.apache.uima.ducc.container.common.fsm.Fsm.transition(Fsm.java:172)
>
> 16 Nov 2015 23:34:39,068 ERROR ActionGet - T[17] engage
> org.apache.uima.ducc.container.jd.JobDriverException
> at
> org.apache.uima.ducc.container.jd.cas.CasManager.getMetaCas(CasManager.java:74)
> at
> org.apache.uima.ducc.container.jd.fsm.wi.ActionGet.engage(ActionGet.java:118)
>
>
> However, when I looked into output directory, there were output (much
> omitted from below):
>
> 
>  mimeType="text" sofaString="
Dr.
> Nutritious

Medical Nutrition Therapy for
> Hyperlipidemia

Referral from: Julie Tester, RD, LD,
> CNSD
Phone contact: (555) 555-1212
Height: 144 cm
> Current Weight: 45 kg   Date of current weight: 02-29-2001   Admit Weight:
>  53 kg   BMI: 18 kg/m2
Diet: General
Daily Calorie needs
> (kcals): 1500 calories, assessed as HB + 20% for activity.
Daily
> Protein needs: 40 grams,  assessed as 1.0 g/kg.
Pt has been on a
> 3-day calorie count and has had an average intake of 1100 calories.  She
> was instructed to drink 2-3 cans of liquid supplement to help promote
> weight gain.  She agrees with the plan and has my number for further
> assessment. May want a Resting Metabolic Rate as well. She takes an aspirin
> a day for knee pain.
"/>
>  _id="8" documentID="testpatient_plaintext_1.txt"/>
>  begin="0" end="735" language="x-unspecified"/>
>  _id="15" _ref_sofa="1" begin="0" end="735" id="SIMPLE_SEGMENT"/>
>  _id="21" _ref_sofa="1" begin="2" end="16" sentenceNumber="0"/>
> ..
> ..
> ..
>  codingScheme="SNOMED" code="103733002" oid="103733002#SNOMED" score="0.0"
> disambiguated="false" cui="C0087111" tui="T061"/>
> 
> 2975
> 2985
> 2965
> 2995
> 
>  codingScheme="SNOMED" code="243121000" oid="243121000#SNOMED" score="0.0"
> disambiguated="false" cui="C0418981" tui="T061"/>
> 
> 2924
> 
>  codingScheme="RXNORM" code="1191" oid="1191#RXNORM" score="0.0"
> disambiguated="false" cui="C0004057" tui="T121"/>
>  codingScheme="RXNORM" code="1191" oid="1191#RXNORM" score="0.0"
> disambiguated="false" cui="C0004057" tui="T109"/>
> 
> 2880
> 2870
> 
> 
>
>
> I am not sure what's going on here, the job wasn't in running stage but has
> output...
> It always got canceled by driver and showed run fails after initializing,
> so I didn't expect there would have any output file.
> And I also doubt whether the content in output file is correctany
> suggestion is appreciated!
>
> Thanks,
> Yi-Wen
>

Re: run fails but has output

[Lou DeGenaro]

Hi Yi-Wen,

Does your CR function correctly outside of DUCC?  Have you tired debugging
using DUCC's All-In-One feature as suggested previously by Eddie?

Lou.

On Tue, Nov 17, 2015 at 3:54 PM, Yi-Wen Liu  wrote:

> Hi,
>
> I tried to remove redundant white spaces but it doesn't work, and I didn't
> add any other special characters in CR.
>
> Is that error caused by invalid space and characters in Xml, or it's more
> like syntax error of CR Xml?
>
>
>
> Thanks,
> Yi-Wen
>
> > On Nov 17, 2015, at 5:37 AM, Lou DeGenaro 
> wrote:
> >
> > Yi-Wen,
> >
> > The DUCC Job first visits the Initialization state.  This is when the Job
> > Driver (JD) initializes your UIMA CollecitonReader (CR). The CR that you
> > specified in your DUCC Job submission is constructed and its getTotal()
> > method is invoked to determine the number of work items.  But I think
> your
> > Job made it past the Initialization state and into the Running state.
> > Looking at the log snippits that you posted, I see:
> >
> > 16 Nov 2015 23:34:38,353  INFO JobDriver - T[18] advanceJdState
> > current=Initializing request=Active result=Active
> >
> > This means the JD has exited its Initializing state and is now Active and
> > is fetching work items.  Further evidence is the entry:
> >
> > 16 Nov 2015 23:34:39,066 ERROR CasManager - T[17] getMetaCas
> >
> > What do the contents of the file ducc.log in your Job's log directory
> > show?  You should see "Initializing" followed by "Running".
> >
> > It looks like the root cause of your problem is that your CR's getNext()
> is
> > trying to produce work items for the JD but is failing:
> >
> > org.xml.sax.SAXParseException; Trying to serialize non-XML 1.0
> > character: , 0x0
> >at
> org.apache.uima.util.XMLSerializer$CharacterValidatingContentHand
> > ler.checkForInvalidXmlChars(XMLSerializer.java:254)
> >
> > I recommend determining why your CR is broken, fixing, and re-submitting
> > your Job to DUCC.
> >
> > Lou.
> >
> >> On Tue, Nov 17, 2015 at 2:46 AM, Yi-Wen Liu  wrote:
> >>
> >> Hello,
> >>
> >> I am still stuck in the following error of DUCC and job always got
> canceled
> >> by driver:
> >>
> >> 16 Nov 2015 23:34:38,353  INFO JobDriver - T[18] advanceJdState
> >> current=Initializing request=Active result=Active
> >> 16 Nov 2015 23:34:39,066 ERROR ProxyLogger - T[17] loggify
> >> org.xml.sax.SAXParseException; Trying to serialize non-XML 1.0
> >> character: , 0x0
> >>at
> >>
> org.apache.uima.util.XMLSerializer$CharacterValidatingContentHandler.checkForInvalidXmlChars(XMLSerializer.java:254)
> >>
> >> 16 Nov 2015 23:34:39,066 ERROR CasManager - T[17] getMetaCas
> >> org.apache.uima.ducc.container.common.classloader.ProxyException
> >>at
> >>
> org.apache.uima.ducc.container.jd.classload.ProxyJobDriverCollectionReader.getMetaCas(ProxyJobDriverCollectionReader.java:127)
> >>at
> >>
> org.apache.uima.ducc.container.jd.cas.CasManager.getMetaCas(CasManager.java:65)
> >>at
> >>
> org.apache.uima.ducc.container.jd.fsm.wi.ActionGet.engage(ActionGet.java:118)
> >>at
> >> org.apache.uima.ducc.container.common.fsm.Fsm.transition(Fsm.java:172)
> >>
> >> 16 Nov 2015 23:34:39,068 ERROR ActionGet - T[17] engage
> >> org.apache.uima.ducc.container.jd.JobDriverException
> >>at
> >>
> org.apache.uima.ducc.container.jd.cas.CasManager.getMetaCas(CasManager.java:74)
> >>at
> >>
> org.apache.uima.ducc.container.jd.fsm.wi.ActionGet.engage(ActionGet.java:118)
> >>
> >>
> >> However, when I looked into output directory, there were output (much
> >> omitted from below):
> >>
> >> 
> >> >> mimeType="text" sofaString="
Dr.
> >> Nutritious

Medical Nutrition Therapy for
> >> Hyperlipidemia

Referral from: Julie Tester, RD, LD,
> >> CNSD
Phone contact: (555) 555-1212
Height: 144 cm
> >> Current Weight: 45 kg   Date of current weight: 02-29-2001   Admit
> Weight:
> >> 53 kg   BMI: 18 kg/m2
Diet: General
Daily Calorie
> needs
> >> (kcals): 1500 calories, assessed as HB + 20% for
> activity.
Daily
> >> Protein needs: 40 grams,  assessed as 1.0 g/kg.
Pt has been on
> a
> >> 3-day calorie count and has had an average intake of 1100 calories.  She
> >> was instruct

Re: run fails but has output

[Lou DeGenaro]

Yi-Wen,

With respect to the ducc.log, I still saw that the DUCC Job was Running as
far as the JD was concerned (as evidenced by the JD's log file).  There may
be some latency in reflecting this fact back to the Orchestrator daemon
from the JD.  Note that the Orchestrator writes the ducc.log file.
The ducc.log
not showing Running could be the subject of a Jira, but is not of concern
with respect to getting your Job to actually run.

I'm not sure I understand your statements regarding All-In-One testing.
You ran your Job successfully without any error using All-In-One, or that
had some issue too?  Did you run All-In-One locally or remotely?

Lou.

On Tue, Nov 17, 2015 at 5:18 PM, Yi-Wen Liu  wrote:

> Hi,
>
> The ducc.log didn't show running:
> 2015.11.16 23:29:26 Mon Received
> 2015.11.16 23:29:26 Mon WaitingForDriver
> 2015.11.16 23:29:54 Mon WaitingForResources
> 2015.11.16 23:30:04 Mon Assigned
> 2015.11.16 23:30:14 Mon Initializing
> 2015.11.16 23:34:48 Mon Completing
> 2015.11.16 23:35:18 Mon Completed
> 2015.11.16 23:35:18 Mon completion type: CanceledByDriver
> 2015.11.16 23:35:18 Mon rationale: unspecified
>
> And I didn't find error in all_in_one local:
>
> [1] Loading model:
>
> [1] ...
>
> [1] Loading model:
>
> [1] .
>
> [1] Loading configuration.
>
> [1] Loading feature templates.
>
> [1] Loading lexica.
>
> [1] Loading model:
>
> [1] 
>
> [1] Loading model:
>
> [1] .
>
> [1] 17 Nov 2015 13:49:33  INFO Chunker - Chunker model file:
> org/apache/ctakes/chunker/models/chunker-model.zip
>
> [1] 17 Nov 2015 13:49:35  INFO JdbcConnectionResourceImpl - Connection
> established to:
> jdbc:hsqldb:res:org/apache/ctakes/dictionary/lookup/umls2011ab/umls
>
> [1] 17 Nov 2015 13:49:35  INFO JdbcConnectionResourceImpl - Connection
> established to:
> jdbc:hsqldb:res:org/apache/ctakes/dictionary/lookup/rxnorm-hsqldb/umls
>
> [1] 17 Nov 2015 13:49:35  INFO JdbcConnectionResourceImpl - Connection
> established to:
> jdbc:hsqldb:res:org/apache/ctakes/dictionary/lookup/orange_book_hsqldb/umls
>
> [1] 17 Nov 2015 13:49:35  INFO UmlsDictionaryLookupAnnotator - Parsing
> descriptor:
>
> /Users/ducc/src/apache-ctakes-3.2.2/resources/org/apache/ctakes/dictionary/lookup/LookupDesc_Db.xml
>
> [1] 17 Nov 2015 13:49:35  INFO FirstTokenPermLookupInitializerImpl -
> Exclusion tagset loaded: [dt, to, rp, ls, pos, md, vbd, vbg, vb, ex, vbp,
> vbn, pdt, vbz, wp, wrb, in, wps, pp$, prp$, wdt, prp, pp, cc, cd]
>
> [1] 17 Nov 2015 13:49:35  INFO FirstTokenPermLookupInitializerImpl -
> Exclusion tagset loaded: [to, dt, rp, ex, vbp, ls, vbn, pdt, wp, vbz, wrb,
> in, pos, wps, md, wdt, pp$, vbd, vb, vbg, pp, cc, cd]
>
> [1] 17 Nov 2015 13:49:35  INFO UmlsDictionaryLookupAnnotator - Using
> ctakes.umlsaddr: https://uts-ws.nlm.nih.gov/restful/isValidUMLSUser:
> yiwenliu
>
> [1] 17 Nov 2015 13:49:39  INFO LvgCmdApiResourceImpl - Loading NLM Norm and
> Lvg with config file =
>
> /Users/ducc/src/apache-ctakes-3.2.2/resources/org/apache/ctakes/lvg/data/config/lvg.properties
>
> [1] 17 Nov 2015 13:49:39  INFO LvgCmdApiResourceImpl -   config file
> absolute path =
>
> /Users/ducc/src/apache-ctakes-3.2.2/resources/org/apache/ctakes/lvg/data/config/lvg.properties
>
> [1] 17 Nov 2015 13:49:39  INFO LvgCmdApiResourceImpl - cwd =
> /Users/ducc/src
>
> [1] 17 Nov 2015 13:49:39  INFO LvgCmdApiResourceImpl - cd
> /Users/ducc/src/apache-ctakes-3.2.2/resources/org/apache/ctakes/lvg/
>
> [1] 17 Nov 2015 13:49:40  INFO LvgCmdApiResourceImpl - cd /Users/ducc/src
>
> It looks like it suddenly canceled by driver and no more output on screen.
> Another log file shows connection refused so I checked driver logs, and
> found the three errors in first email.
>
> Is that error org.xml.sax.SAXParseException; Trying to serialize non-XML
> 1.0 character caused by invalid space and characters in Xml, or it's more
> like syntax error of CR Xml?
>
> Thanks,
> Yi-Wen
>
> On Tue, Nov 17, 2015 at 1:36 PM, Lou DeGenaro 
> wrote:
>
> > Hi Yi-Wen,
> >
> > Does your CR function correctly outside of DUCC?  Have you tired
> debugging
> > using DUCC's All-In-One feature as suggested previously by Eddie?
> >
> > Lou.
> >
> > On Tue, Nov 17, 2015 at 3:54 PM, Yi-Wen Liu  wrote:
> >
> > > Hi,
> > >
> > > I tried to remove redundant white spaces but it doesn't work, and I
> > didn't
> > > add any other special characters in CR.
> > >
> > > Is that error caused by invalid space and characters in Xml, or it's
> more
> > > like syntax error of CR Xml?

Re: run fails but has output

[Lou DeGenaro]

Yi-Wen,

Oh, I see now you ran all-in-one local.  But that had errors or ran to
completion successfully?

Lou.

On Tue, Nov 17, 2015 at 5:29 PM, Lou DeGenaro 
wrote:

> Yi-Wen,
>
> With respect to the ducc.log, I still saw that the DUCC Job was Running
> as far as the JD was concerned (as evidenced by the JD's log file).
> There may be some latency in reflecting this fact back to the Orchestrator
> daemon from the JD.  Note that the Orchestrator writes the ducc.log
> file.  The ducc.log not showing Running could be the subject of a Jira,
> but is not of concern with respect to getting your Job to actually run.
>
> I'm not sure I understand your statements regarding All-In-One testing.
> You ran your Job successfully without any error using All-In-One, or that
> had some issue too?  Did you run All-In-One locally or remotely?
>
> Lou.
>
> On Tue, Nov 17, 2015 at 5:18 PM, Yi-Wen Liu  wrote:
>
>> Hi,
>>
>> The ducc.log didn't show running:
>> 2015.11.16 23:29:26 Mon Received
>> 2015.11.16 23:29:26 Mon WaitingForDriver
>> 2015.11.16 23:29:54 Mon WaitingForResources
>> 2015.11.16 23:30:04 Mon Assigned
>> 2015.11.16 23:30:14 Mon Initializing
>> 2015.11.16 23:34:48 Mon Completing
>> 2015.11.16 23:35:18 Mon Completed
>> 2015.11.16 23:35:18 Mon completion type: CanceledByDriver
>> 2015.11.16 23:35:18 Mon rationale: unspecified
>>
>> And I didn't find error in all_in_one local:
>>
>> [1] Loading model:
>>
>> [1] ...
>>
>> [1] Loading model:
>>
>> [1] .
>>
>> [1] Loading configuration.
>>
>> [1] Loading feature templates.
>>
>> [1] Loading lexica.
>>
>> [1] Loading model:
>>
>> [1] 
>>
>> [1] Loading model:
>>
>> [1] .
>>
>> [1] 17 Nov 2015 13:49:33  INFO Chunker - Chunker model file:
>> org/apache/ctakes/chunker/models/chunker-model.zip
>>
>> [1] 17 Nov 2015 13:49:35  INFO JdbcConnectionResourceImpl - Connection
>> established to:
>> jdbc:hsqldb:res:org/apache/ctakes/dictionary/lookup/umls2011ab/umls
>>
>> [1] 17 Nov 2015 13:49:35  INFO JdbcConnectionResourceImpl - Connection
>> established to:
>> jdbc:hsqldb:res:org/apache/ctakes/dictionary/lookup/rxnorm-hsqldb/umls
>>
>> [1] 17 Nov 2015 13:49:35  INFO JdbcConnectionResourceImpl - Connection
>> established to:
>>
>> jdbc:hsqldb:res:org/apache/ctakes/dictionary/lookup/orange_book_hsqldb/umls
>>
>> [1] 17 Nov 2015 13:49:35  INFO UmlsDictionaryLookupAnnotator - Parsing
>> descriptor:
>>
>> /Users/ducc/src/apache-ctakes-3.2.2/resources/org/apache/ctakes/dictionary/lookup/LookupDesc_Db.xml
>>
>> [1] 17 Nov 2015 13:49:35  INFO FirstTokenPermLookupInitializerImpl -
>> Exclusion tagset loaded: [dt, to, rp, ls, pos, md, vbd, vbg, vb, ex, vbp,
>> vbn, pdt, vbz, wp, wrb, in, wps, pp$, prp$, wdt, prp, pp, cc, cd]
>>
>> [1] 17 Nov 2015 13:49:35  INFO FirstTokenPermLookupInitializerImpl -
>> Exclusion tagset loaded: [to, dt, rp, ex, vbp, ls, vbn, pdt, wp, vbz, wrb,
>> in, pos, wps, md, wdt, pp$, vbd, vb, vbg, pp, cc, cd]
>>
>> [1] 17 Nov 2015 13:49:35  INFO UmlsDictionaryLookupAnnotator - Using
>> ctakes.umlsaddr: https://uts-ws.nlm.nih.gov/restful/isValidUMLSUser:
>> yiwenliu
>>
>> [1] 17 Nov 2015 13:49:39  INFO LvgCmdApiResourceImpl - Loading NLM Norm
>> and
>> Lvg with config file =
>>
>> /Users/ducc/src/apache-ctakes-3.2.2/resources/org/apache/ctakes/lvg/data/config/lvg.properties
>>
>> [1] 17 Nov 2015 13:49:39  INFO LvgCmdApiResourceImpl -   config file
>> absolute path =
>>
>> /Users/ducc/src/apache-ctakes-3.2.2/resources/org/apache/ctakes/lvg/data/config/lvg.properties
>>
>> [1] 17 Nov 2015 13:49:39  INFO LvgCmdApiResourceImpl - cwd =
>> /Users/ducc/src
>>
>> [1] 17 Nov 2015 13:49:39  INFO LvgCmdApiResourceImpl - cd
>> /Users/ducc/src/apache-ctakes-3.2.2/resources/org/apache/ctakes/lvg/
>>
>> [1] 17 Nov 2015 13:49:40  INFO LvgCmdApiResourceImpl - cd /Users/ducc/src
>>
>> It looks like it suddenly canceled by driver and no more output on screen.
>> Another log file shows connection refused so I checked driver logs, and
>> found the three errors in first email.
>>
>> Is that error org.xml.sax.SAXParseException; Trying to serialize non-XML
>> 1.0 character caused by invalid space and characters in Xml, or it's more
>> like syntax error of CR Xml?
>>
>> Thanks,
>> Yi-Wen
>>
>> On Tue, Nov 17, 2015 at 1:36 PM, Lou DeGenaro 
>> wrote

Re: some short questions about UIMA DUCC

[Lou DeGenaro]

Yi-Wen,

Privileged ducc_ling is for situations where sharing of resources and data
privacy are desired.  For example, user "ducc" would use ducc_ling to
assume the identity of the job submitter in order to gain access to read
and write.  This is useful on large compute clusters with many users. When
testing on my small simulated cluster I do not use a privileged ducc_ling.

Your second question is harder to answer without more information.  Was
there any contention for resources (CPU,disk)?  Look at the Work Items tab
for the longest and shortest jobs and see if you notice any pattern.  Were
work items slow on a particular node?  Were delivery times longer?  Were
process times longer for a few or all the work items?  Was there a lot of
pre-emption (two or more jobs running at once)?  Was there an equal amount
of resource (e.g. number of processes) allocated to each job?

Lou.

On Tue, Nov 24, 2015 at 5:35 PM, Yi-Wen Liu  wrote:

> Hello,
>
> I have some small questions about DUCC, most of them are not technical,
> hope somebody can help me out, thanks!
>
> From https://cwiki.apache.org/confluence/display/UIMA/DUCC, it includes a
> step "Privileged ducc_ling".
> But if I ignore this step, DUCC still works well, I am wondering what is
> this step especially for?
>
> The second question is, I submitted same input files many times, and the
> completed time were very different, range from 2 min to 7 min.
> While I was running DUCC I didn't let the computer busy, and one job
> running at a time.
> Is there a reason why it sometimes finished early sometimes it took such a
> long time to complete the job?
>
> Thanks,
> Yi-Wen
>

Re: some short questions about UIMA DUCC

[Lou DeGenaro]

ducc_submit --help will give you a list of possibilities.  Of interest may
be --environment, --driver_jvm_args and --process_jvm_args among others.

Lou.

On Tue, Nov 24, 2015 at 8:05 PM, Yi-Wen Liu  wrote:

> Hello,
>
> I see, thanks for your reply!
>
> I just recall another question, is that possible to specify configuration
> file for my project running on ducc, in the job file?
>
> Thanks,
> Yi-Wen
>
> On Tue, Nov 24, 2015 at 3:15 PM, Lou DeGenaro 
> wrote:
>
> > Yi-Wen,
> >
> > Privileged ducc_ling is for situations where sharing of resources and
> data
> > privacy are desired.  For example, user "ducc" would use ducc_ling to
> > assume the identity of the job submitter in order to gain access to read
> > and write.  This is useful on large compute clusters with many users.
> When
> > testing on my small simulated cluster I do not use a privileged
> ducc_ling.
> >
> > Your second question is harder to answer without more information.  Was
> > there any contention for resources (CPU,disk)?  Look at the Work Items
> tab
> > for the longest and shortest jobs and see if you notice any pattern.
> Were
> > work items slow on a particular node?  Were delivery times longer?  Were
> > process times longer for a few or all the work items?  Was there a lot of
> > pre-emption (two or more jobs running at once)?  Was there an equal
> amount
> > of resource (e.g. number of processes) allocated to each job?
> >
> > Lou.
> >
> > On Tue, Nov 24, 2015 at 5:35 PM, Yi-Wen Liu  wrote:
> >
> > > Hello,
> > >
> > > I have some small questions about DUCC, most of them are not technical,
> > > hope somebody can help me out, thanks!
> > >
> > > From https://cwiki.apache.org/confluence/display/UIMA/DUCC, it
> includes
> > a
> > > step "Privileged ducc_ling".
> > > But if I ignore this step, DUCC still works well, I am wondering what
> is
> > > this step especially for?
> > >
> > > The second question is, I submitted same input files many times, and
> the
> > > completed time were very different, range from 2 min to 7 min.
> > > While I was running DUCC I didn't let the computer busy, and one job
> > > running at a time.
> > > Is there a reason why it sometimes finished early sometimes it took
> such
> > a
> > > long time to complete the job?
> > >
> > > Thanks,
> > > Yi-Wen
> > >
> >
>

Re: submit job programically through http request

[Lou DeGenaro]

There is nothing magical about DUCC's CLI/API, though it is quite
convenient. If one wanted to send/receive HTTP directly to the appropriate
DUCC daemon one could.  Fetch the DUCC's CLI package source code to
discover how it is done.

Lou.

On Sun, Nov 29, 2015 at 6:19 PM, Eddie Epstein  wrote:

> On Sat, Nov 28, 2015 at 7:28 PM, Yi-Wen Liu  wrote:
>
> > Hi,
> >
> > Yes I've looked at the document.
> > Sorry I am not sure I am understanding, so because of security identity
> > issue, it is not allowed to simply send http request to the ducc server
> by
> > the user?
> >
>
> Yes, I believe this is correct.
>
> Eddie
>

Re: lower preprocessing time

[Lou DeGenaro]

Yi-Wen,

The latency you are experiencing is by-design for a large-ish computing
cluster.  The normal life-cycle for a Job is:

Received WaitingForDriver WaitingForResources
Assigned
Initializing
Running
Completing
Completed

There are some knobs you can turn to tune for your situation.

1. DUCC intra-daemon communications - states affected: All

DUCC is implemented as a small collection of daemons that communicate with
each other at discrete publishing intervals.  The publishing intervals are
configured in $DUCC_HOME/resources/ducc.properties.  The default interval
values are on the order of 15-60 seconds.  At the cost of more chatter
between daemons on the network, you can try lowering some of these values.

These times are the current default ones and are specified in milliseconds:

ducc.jd.state.publish.rate=15000
ducc.orchestrator.state.publish.rate=1
ducc.pm.state.publish.rate=15000

I have not tried this myself, but perhaps try lowering them to:

ducc.jd.state.publish.rate=2000
ducc.orchestrator.state.publish.rate=1000
ducc.pm.state.publish.rate=1000

2. DUCC scheduling - state affected: WaitingForResources

The DUCC scheduler does not do continuous resource management, but rather
calculates a desired allocation at discrete intervals.  After each
scheduling cycle, the scheduler publishes its layout for the other daemons
to implement.  By default, the scheduler is doing this calculation and
publication whenever it receives an orchestrator.state publication:

ducc.rm.state.publish.ratio = 1

This seems fine as is.

3. DUCC deployment of Job - states affected: WaitingForDriver, Initializing

Once a Job is accepted, the Job Driver [your CollectionReader] and one or
more Job Processes [your AnlaysisEngine] must be launched.

The partial sequence of states here are:

WaitingForDriver: The Job Driver is launched, and not until it reports that
is is ready to produce work items will the next state (WaitingForResources)
occur
...
Initializing: A Job Process is launched, and not until it has completed
initialization of all threads will it ask the Job Driver for the first work
item
Running: The first work item has been dispatched

Minimizing the time for your CR to initialize will help make the transition
from WaitingForDriver to WaitingForResources faster.
Minimizing the time for your AE to initialize will help make the transition
from Initializing to Running faster.

Hope this helps.

Lou.

On Sun, Nov 29, 2015 at 11:25 PM, Yi-Wen Liu  wrote:

> Hi,
>
> Thanks for the reply, and yes, I only have a single work item.
>
> Thanks,
> Yi-Wen
>
> On Sun, Nov 29, 2015 at 7:45 PM, Eddie Epstein 
> wrote:
>
> > Hi,
> >
> > Yes, there are some site.ducc.property entries that will speed up the
> > timing. Will respond with those tomorrow.
> > Are you often running jobs with only a single work item?
> >
> > Eddie
> >
> > On Sat, Nov 28, 2015 at 7:23 PM, Yi-Wen Liu  wrote:
> >
> > > Hi,
> > >
> > > I am using ducc to process text files(cTAKES), and one of my input is
> > quite
> > > short, about 10 lines.
> > > But it takes more than two minutes to process it, as follows:
> > > After submitting,
> > > 00:00-00:08 > no status
> > > 00:09-00:30 > waiting for driver
> > > 00:31-01:00 > waiting for resources
> > > 01:01-02:00 > initializing
> > > 02:01-02:30 > completing
> > > 02:31 > completed
> > >
> > > Is there any way to lower the preprocessing time?(Time to wait for
> > driver,
> > > resources, initializing...)
> > >
> > > I am wondering why it takes so long before completing, and have tried
> > > different parameter values, for example lower initialization time,
> lower
> > > resources needed, but didn't have much improvement.
> > >
> > > Here's parameters I am using now: process_memory_size 2
> > > process_jvm_args -Xmx4g
> > > driver_jvm_args -Xmx4g
> > > process_thread_count 2
> > > process_per_item_time_max 5
> > > process_deployments_max 999
> > > environment AE_INIT_TIME=5 AE_INIT_RANGE=5 INIT_ERROR=0
> > >
> > > Any suggestion is appreciated.
> > >
> > > Thanks,
> > > Yi-Wen
> > >
> >
>

Re: submit job programically through http request

[Lou DeGenaro]

In org.apache.uima.ducc.transport.dispatcher.DuccEventDispatcher.
dispatchAndWaitForDuccReply DUCC uses Camel to send a Job request and
receive the reply (and thus HTTP is not directly used).

Lou.

On Mon, Nov 30, 2015 at 6:11 AM, Lou DeGenaro 
wrote:

> There is nothing magical about DUCC's CLI/API, though it is quite
> convenient. If one wanted to send/receive HTTP directly to the appropriate
> DUCC daemon one could.  Fetch the DUCC's CLI package source code to
> discover how it is done.
>
> Lou.
>
> On Sun, Nov 29, 2015 at 6:19 PM, Eddie Epstein 
> wrote:
>
>> On Sat, Nov 28, 2015 at 7:28 PM, Yi-Wen Liu  wrote:
>>
>> > Hi,
>> >
>> > Yes I've looked at the document.
>> > Sorry I am not sure I am understanding, so because of security identity
>> > issue, it is not allowed to simply send http request to the ducc server
>> by
>> > the user?
>> >
>>
>> Yes, I believe this is correct.
>>
>> Eddie
>>
>
>

Re: submit job programically through http request

[Lou DeGenaro]

You're right!  My mistake.  I was fooled by my poor use of eclipse
reference resolution.  BaseHttpDispatcher is used, not DuccEventDispatcher.

Lou.

On Mon, Nov 30, 2015 at 9:52 AM, Jim Challenger  wrote:

> Lou, the CLI actually uses  DuccEventHttpDispatcherCl which derives from
> BaseHttpDispatcher which does use the java.net HTTP stuff, not Camel.  Jim
>
> On 11/30/15 8:22 AM, Lou DeGenaro wrote:
>
>> In org.apache.uima.ducc.transport.dispatcher.DuccEventDispatcher.
>> dispatchAndWaitForDuccReply DUCC uses Camel to send a Job request and
>> receive the reply (and thus HTTP is not directly used).
>>
>> Lou.
>>
>> On Mon, Nov 30, 2015 at 6:11 AM, Lou DeGenaro 
>> wrote:
>>
>> There is nothing magical about DUCC's CLI/API, though it is quite
>>> convenient. If one wanted to send/receive HTTP directly to the
>>> appropriate
>>> DUCC daemon one could.  Fetch the DUCC's CLI package source code to
>>> discover how it is done.
>>>
>>> Lou.
>>>
>>> On Sun, Nov 29, 2015 at 6:19 PM, Eddie Epstein 
>>> wrote:
>>>
>>> On Sat, Nov 28, 2015 at 7:28 PM, Yi-Wen Liu  wrote:
>>>>
>>>> Hi,
>>>>>
>>>>> Yes I've looked at the document.
>>>>> Sorry I am not sure I am understanding, so because of security identity
>>>>> issue, it is not allowed to simply send http request to the ducc server
>>>>>
>>>> by
>>>>
>>>>> the user?
>>>>>
>>>>> Yes, I believe this is correct.
>>>>
>>>> Eddie
>>>>
>>>>
>>>
>

Re: lower preprocessing time

[Lou DeGenaro]

I have now tried the suggestion of my previous append.

My first recommendation is that one should not change these values
(especially for a production system) away from the defaults, as some of my
co-committers have reminded me offline.

On my *test* system I did modify site.ducc.properites and the system ran
1.job just fine.  I did not examine resource consumption (CPU), though I
sure it had to be higher to support the increased communications and
scheduling overhead.  And remember that 1.job is a "fake" job - the work
items only sleep so there is no competition for CPU.  Also, on the System
Daemons page, the ResourceManager showed as "down" every once-in-a-while
(even though it was really up) because its minimum publish rate is 5
seconds.

My second recommendation is to do one of the following instead:

1. submit Jobs with more than 1 work item
2. re-imagine your Job as a Service
3. Use all-in-one local

Lou.



On Mon, Nov 30, 2015 at 9:21 AM, Lou DeGenaro 
wrote:

> Yi-Wen,
>
> The latency you are experiencing is by-design for a large-ish computing
> cluster.  The normal life-cycle for a Job is:
>
> Received WaitingForDriver WaitingForResources
> Assigned
> Initializing
> Running
> Completing
> Completed
>
> There are some knobs you can turn to tune for your situation.
>
> 1. DUCC intra-daemon communications - states affected: All
>
> DUCC is implemented as a small collection of daemons that communicate
> with each other at discrete publishing intervals.  The publishing intervals
> are configured in $DUCC_HOME/resources/ducc.properties.  The default
> interval values are on the order of 15-60 seconds.  At the cost of more
> chatter between daemons on the network, you can try lowering some of these
> values.
>
> These times are the current default ones and are specified in milliseconds:
>
> ducc.jd.state.publish.rate=15000
> ducc.orchestrator.state.publish.rate=1
> ducc.pm.state.publish.rate=15000
>
> I have not tried this myself, but perhaps try lowering them to:
>
> ducc.jd.state.publish.rate=2000
> ducc.orchestrator.state.publish.rate=1000
> ducc.pm.state.publish.rate=1000
>
> 2. DUCC scheduling - state affected: WaitingForResources
>
> The DUCC scheduler does not do continuous resource management, but rather
> calculates a desired allocation at discrete intervals.  After each
> scheduling cycle, the scheduler publishes its layout for the other daemons
> to implement.  By default, the scheduler is doing this calculation and
> publication whenever it receives an orchestrator.state publication:
>
> ducc.rm.state.publish.ratio = 1
>
> This seems fine as is.
>
> 3. DUCC deployment of Job - states affected: WaitingForDriver,
> Initializing
>
> Once a Job is accepted, the Job Driver [your CollectionReader] and one or
> more Job Processes [your AnlaysisEngine] must be launched.
>
> The partial sequence of states here are:
>
> WaitingForDriver: The Job Driver is launched, and not until it reports
> that is is ready to produce work items will the next state
> (WaitingForResources) occur
> ...
> Initializing: A Job Process is launched, and not until it has completed
> initialization of all threads will it ask the Job Driver for the first work
> item
> Running: The first work item has been dispatched
>
> Minimizing the time for your CR to initialize will help make the
> transition from WaitingForDriver to WaitingForResources faster.
> Minimizing the time for your AE to initialize will help make the
> transition from Initializing to Running faster.
>
> Hope this helps.
>
> Lou.
>
> On Sun, Nov 29, 2015 at 11:25 PM, Yi-Wen Liu  wrote:
>
>> Hi,
>>
>> Thanks for the reply, and yes, I only have a single work item.
>>
>> Thanks,
>> Yi-Wen
>>
>> On Sun, Nov 29, 2015 at 7:45 PM, Eddie Epstein 
>> wrote:
>>
>> > Hi,
>> >
>> > Yes, there are some site.ducc.property entries that will speed up the
>> > timing. Will respond with those tomorrow.
>> > Are you often running jobs with only a single work item?
>> >
>> > Eddie
>> >
>> > On Sat, Nov 28, 2015 at 7:23 PM, Yi-Wen Liu  wrote:
>> >
>> > > Hi,
>> > >
>> > > I am using ducc to process text files(cTAKES), and one of my input is
>> > quite
>> > > short, about 10 lines.
>> > > But it takes more than two minutes to process it, as follows:
>> > > After submitting,
>> > > 00:00-00:08 > no status
>> > > 00:09-00:30 > waiting for driver
>> > > 00:31-01:00 > waiting for resources
>> > > 01:01-02:00 > initializing
>> > > 02:01-02:30 >

Re: [jira] [Commented] (UIMA-4563) Traffic monitoring using Google Analytics

[Lou DeGenaro]

To initially get to the DUCC demo, one must start from here http://uima.
apache.org/ which has the privacy statement link, then click on Demo Page
or Demo Live.  I don't think we need to add privacy statement links on the
demo pages themselves.  The site has already clearly stated its policy.

I'm happy to ask for a legal opinion if there is a place to ask for such
within the Apache community.  I would otherwise resist hacking the demo
pages themselves.

Lou.

On Tue, Dec 22, 2015 at 2:57 PM, Marshall Schor (JIRA) 
wrote:

>
> [
> https://issues.apache.org/jira/browse/UIMA-4563?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15068643#comment-15068643
> ]
>
> Marshall Schor commented on UIMA-4563:
> --
>
> It seems that the page where Google Analytics (GA) was added is missing
> the privacy link.  I think it needs to be added?
>
> Also I think our privacy page is slightly defective for GA - I think it
> needs to have actual links to the relevant GA pages?
>
>
>
> > Traffic monitoring using Google Analytics
> > -
> >
> > Key: UIMA-4563
> > URL: https://issues.apache.org/jira/browse/UIMA-4563
> > Project: UIMA
> >  Issue Type: Improvement
> >  Components: Website
> >Reporter: Richard Eckart de Castilho
> >Assignee: Richard Eckart de Castilho
> >
> > Apparently, the ASF main page uses Google Analytics now as well, so I
> think we can also do that - they have privacy statement that we can simply
> reuse.
> > http://www.apache.org/foundation/policies/privacy.html
>
>
>
> --
> This message was sent by Atlassian JIRA
> (v6.3.4#6332)
>

Re: [jira] [Commented] (UIMA-4563) Traffic monitoring using Google Analytics

[Lou DeGenaro]

Oh, I see you added this to the Jira.  I'll respond in kind.

Lou.

On Tue, Dec 22, 2015 at 3:14 PM, Lou DeGenaro 
wrote:

> To initially get to the DUCC demo, one must start from here http://uima.
> apache.org/ which has the privacy statement link, then click on Demo Page
> or Demo Live.  I don't think we need to add privacy statement links on the
> demo pages themselves.  The site has already clearly stated its policy.
>
> I'm happy to ask for a legal opinion if there is a place to ask for such
> within the Apache community.  I would otherwise resist hacking the demo
> pages themselves.
>
> Lou.
>
> On Tue, Dec 22, 2015 at 2:57 PM, Marshall Schor (JIRA) <
> dev@uima.apache.org> wrote:
>
>>
>> [
>> https://issues.apache.org/jira/browse/UIMA-4563?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15068643#comment-15068643
>> ]
>>
>> Marshall Schor commented on UIMA-4563:
>> --
>>
>> It seems that the page where Google Analytics (GA) was added is missing
>> the privacy link.  I think it needs to be added?
>>
>> Also I think our privacy page is slightly defective for GA - I think it
>> needs to have actual links to the relevant GA pages?
>>
>>
>>
>> > Traffic monitoring using Google Analytics
>> > -
>> >
>> > Key: UIMA-4563
>> > URL: https://issues.apache.org/jira/browse/UIMA-4563
>> > Project: UIMA
>> >  Issue Type: Improvement
>> >  Components: Website
>> >Reporter: Richard Eckart de Castilho
>> >Assignee: Richard Eckart de Castilho
>> >
>> > Apparently, the ASF main page uses Google Analytics now as well, so I
>> think we can also do that - they have privacy statement that we can simply
>> reuse.
>> > http://www.apache.org/foundation/policies/privacy.html
>>
>>
>>
>> --
>> This message was sent by Atlassian JIRA
>> (v6.3.4#6332)
>>
>
>

Re: uima ducc wait_for_completion

[Lou DeGenaro]

Yi-Wen,

When submitting a DUCC Job, wait for completion should prevent the return
until all your work items have been processed.

What exactly do you mean by "main process"?

Lou.

On Sat, Feb 20, 2016 at 10:30 PM, Yi-Wen Liu  wrote:

> Hello,
>
> I am using ducc in a web app, and I need to know when does ducc finish its
> work, ie, generate the output file.
> I use the two properties in my job file:
>
> cancel_on_interrupt
> wait_for_completion
>
> However, when I submitted job file, it did show the status on terminal, but
> it finished earlier than the main process, I have to wait for few minutes
> for the output file to be generated.
>
> Is there any way to get notified when the main process finishes without
> looking at the web server page?
>
> Any suggestion is appreciated!
>
>
> Thanks,
> Yi-Wen
>

Re: Quarterly board report items?

[Lou DeGenaro]

According to GA for Apache UIMA site from inception of data collection (17
Dec 2015) to date:

Sessions = 15,536
Avg. duration = 2:39

Lou.


On Fri, Mar 4, 2016 at 1:17 PM, Marshall Schor  wrote:

> Hi,
>
> The board report is due Wed Mar 9th (this coming Wednesday).
>
> I'll put in the normal items about releases, etc., and mention some work
> toward
> looking at signing Jars, and looking into SCRCLR stuff, and other general
> progress info.
>
> Anything else?
>
> -Marshall
>

Re: Moving UIMA DUCC out of the sandbox?

[Lou DeGenaro]

Will you be listening to Pomp and Circumstance during the graduation?

:-)

Lou.

On Wed, Apr 13, 2016 at 2:46 PM, Jaroslaw Cwiklik  wrote:

> Thanks to all for suggestions and comments. I will graduate DUCC from the
> sandbox before the next release of DUCC (2.1.0). I'll do svn move before
> RC1 is made.
> Jerry
>
> On Wed, Apr 13, 2016 at 1:30 PM, Marshall Schor  wrote:
>
> > Hi,
> >
> > I'm +1 for moving DUCC from the Sandbox - it certainly has had a lot of
> > time and
> > energy invested in multiple releases, and has been getting quite a bit of
> > use.
> > I think the community would appreciate the validation of it being a
> > regular project.
> >
> > Re: formal voting - my view is that voting serves to catalyze consensus
> > building, and it's of course also there for legal reasons (for releases).
> > In
> > this case, it seems pretty obvious to me that there's consensus already
> > around
> > moving it from the Sandbox to be a regular project; I think this is
> > essentially
> > mostly an SVN move, plus some publicity when it happens (tweeting :-) ,
> > website
> > updating, news).
> >
> > So, it's OK with me to proceed to move it in SVN.
> >
> > -Marshall
> >
> > On 4/11/2016 11:54 AM, Peter Klügl wrote:
> > > I think a formal vote with a special mail header and a 72h delay would
> > > be a good thing. If somebody has reasons against moving it, it is more
> > > likely that the vote is not missed and that there is enough time to
> > react.
> > >
> > > Best,
> > >
> > > Peter
> > >
> > > Am 11.04.2016 um 17:32 schrieb Richard Eckart de Castilho:
> > >> Thanks Jerry,
> > >>
> > >> DUCC is not an incubator project - it is a sub-project of Apache UIMA.
> > >>
> > >> My understanding that for the purposes of the ASF, DUCC is already an
> > >> integral part of the Apache UIMA project. That is reflected in the
> > >> fact that DUCC versions do not include the "incubator" marker that
> > >> ASF incubator projects have.
> > >>
> > >> I believe moving the project from the sandbox spot in SVN to the
> > >> "regular" location is entirely our own decision - and that is all
> > >> that needs to be done.
> > >>
> > >> On the website, DUCC is not even listed as part of the UIMA Sandbox
> > >> (https://uima.apache.org/sandbox.html) but it is rather advertised
> > >> as a proper part of UIMA under the "Scaleout Frameworks".
> > >>
> > >> We can do a formal vote, but actually I think that based on the +1's
> we
> > >> have already and the fact that this is not an official ASF action,
> > >> you can simply move it.
> > >>
> > >> @Marshall: what do you think?
> > >>
> > >> Cheers,
> > >>
> > >> -- Richard
> > >>
> > >>> On 11.04.2016, at 17:26, Jaroslaw Cwiklik  wrote:
> > >>>
> > >>> Richard, from what I read to graduate DUCC to UIMA subproject we
> need a
> > >>> formal vote:
> > >>>
> > >>> Graduation Approval Vote
> > >>>
> > >>> To graduate as a subproject, the Mentors
> > >>> <
> >
> http://incubator.apache.org/incubation/Roles_and_Responsibilities.html#Mentor
> > >
> > >>> should
> > >>> start a VOTE  thread
> on
> > >>> the general
> > >>> list
> > >>> <
> >
> http://incubator.apache.org/guides/lists.html#general+at+incubator.apache.org
> > >
> > >>> proposing
> > >>> that the IPMC
> > >>> <
> >
> http://incubator.apache.org/incubation/Roles_and_Responsibilities.html#Incubator+Project+Management+Committee+%28PMC%29
> > >
> > >>> signs
> > >>> off the graduation of the podling as a subproject. This VOTEs should
> > only
> > >>> be started once the project has VOTEd to accept the subproject.
> > >>>
> > >>>
> > >>> Perhaps we should wait until DUCC 2.1.0 is released. Looks like we
> are
> > a
> > >>> few weeks away from the release candidate. After the release I can
> > formally
> > >>> put DUCC graduation to a vote.
> > >>>
> > >>> Jerry
> > >>>
> > >>> On Mon, Apr 11, 2016 at 10:42 AM, Jaroslaw Cwiklik  >
> > wrote:
> > >>>
> >  +1
> > 
> >  On Mon, Apr 11, 2016 at 3:24 AM, Peter Klügl <
> > peter.klu...@averbis.com>
> >  wrote:
> > 
> > > +1
> > >
> > > Peter
> > >
> > > Am 10.04.2016 um 01:11 schrieb Richard Eckart de Castilho:
> > >> Hi,
> > >>
> > >> how about moving UIMA DUCC out of the sandbox and promoting it to
> > the
> > > same level as uimaFIT and RUTA?
> > >> Cheers,
> > >>
> > >> -- Richard
> > >
> >
> >
>

Re: [VOTE] Release UIMA-AS 2.8.1 RC3

[Lou DeGenaro]

- Downloaded from bin - OK
- Checked README, RELEASE_NOTES, LICENSE and NOTICE - OK but...

Noticed that SLF4J API Module (http://www.slf4j.org)
org.slf4j:slf4j-api:jar:1.7.10 is not the same version as found in the
AMQ lib dir

- Started Broker using startBroker.sh - OK
- Deployed Deploy_MeetingDetectorTAE.xml and ran a few test CASes - OK but...

Tried to run test case from wrong dir (not too good at following
instructions, it seems) and got unexpected results.
My bad, but would be nice if my initial error were made more obvious somehow.

[x] +1 OK to release (non-binding)


Lou.

Re: [VOTE] Release UIMA-AS 2.8.1 RC3

[Lou DeGenaro]

Why do we feel the need to edit the AMQ distribution rather than include it
and its Notice and License information in its entirety?  If we think a
lighter weight AMQ distribution is desirable, any chance of get the AMQ
folks to provide such instead of us hacking?

On Tue, Apr 26, 2016 at 5:13 PM, Richard Eckart de Castilho 
wrote:

> I agree with Marshall. The NOTICE/LICENSE files should reflect what
> is actually inside a JAR/ZIP and non-applicable parts should - if feasible
> -
> be removed.
>
> That said, the LICENSE/NOTICE files e.g. from the Spring distribution are
> also
> usually shooting beyond the goal and make references to third-party code
> that
> may or may not be in a particular artifact... and all artifacts contain the
> same notices/licenses.
>
> In the case of Spring, I find it not particularly practicable to figure out
> what is relevant and what not.
>
> But here, specific parts seem to be clearly assignable to specific
> non-packaged JARs,
> so it appears to be practicable to be more accurate.
>
> -- Richard
>
> > On 26.04.2016, at 23:03, Marshall Schor  wrote:
> >
> > Here's a summary (please correct if wrong):
> >
> > There are two "optional" JARs not distributed with UIMA-AS have license
> and
> > (partial) notice info in the uima-as LICENSE/NOTICE files.
> >
> > One of the value propositions that lets others make use of our
> technology is the
> > reputation we maintain about our always somewhat imperfect attempts at
> having
> > accurate license and notice files.  I would prefer that we strive to
> keep our
> > reputation where it is by removing the license and partial notice for
> these
> > JARs, and perhaps adding some documentation (if needed) specifying what
> JARs can
> > be optionally downloaded (from ActiveMQ distribution) for providing
> additional
> > functionality, not provided out of the box by the UIMA-AS binary
> distribution.
> >
> > Having said that, if the others on the PMC feel that this flaw (having
> extra
> > licenses and extra (partial) notices not needed is OK for releasing, I
> won't
> > stand in the way.
> >
> > I'll do a bit more testing, and then if nothing more is found, vote -0 to
> > indicate this.
> >
> > http://www.apache.org/foundation/voting.html
> >
> > -Marshall
> >
> >
> > On 4/26/2016 11:05 AM, Jaroslaw Cwiklik wrote:
> >> Thanks Marshall. Just to provide more context for the problems found
> >>
> >> The JmDNS seems to be part of auto discovery of network of brokers via
> >> unicast instead of hard coded URLs.  This is not part of standard
> uima-as
> >> configuration we distribute. When such functionality is needed users may
> >> download their own copy of AMQ and use that. Of course there is an
> issue of
> >> having this jar documented in LICENSE and NOTICE but not present in the
> >> distribution.
> >>
> >> The second one jasypt is providing encryption and decryption of user
> >> credentials per: http://activemq.apache.org/encrypted-passwords.html. I
> >> think the lack of this jar can also be dealt with the same way as above.
> >>
> >> Given the above I will await your vote. One way or the other I need your
> >> vote to proceed. Seems like quality of the distribution mandates RC3
> vote
> >> down.
> >>
> >> Jerry
> >>
> >> On Mon, Apr 25, 2016 at 5:56 PM, Marshall Schor  wrote:
> >>
> >>> Although others may be voting +1 to release, just to be clear, I'm
> >>> currently -1
> >>> until the license / notice issues mentioned above are resolved.
> >>>
> >>> -Marshall
>
>

Re: [jira] [Created] (UIMA-4911) Promote ducc from sandbox to the regular project

[Lou DeGenaro]

I notice that on the wiki page
https://cwiki.apache.org/confluence/display/UIMA/Index there is a section
entitled "Sandbox" comprising two bullet items: uimaFIT and DUCC[demo].
This should probably be changed.

On Wed, Apr 27, 2016 at 9:27 AM, Jerry Cwiklik (JIRA) 
wrote:

> Jerry Cwiklik created UIMA-4911:
> ---
>
>  Summary: Promote ducc from sandbox to the regular project
>  Key: UIMA-4911
>  URL: https://issues.apache.org/jira/browse/UIMA-4911
>  Project: UIMA
>   Issue Type: Task
>   Components: DUCC
> Reporter: Jerry Cwiklik
> Assignee: Jerry Cwiklik
>  Fix For: 2.1.0-Ducc
>
>
> Ducc has been in the uima sandbox since its inception. It has a number of
> releases behind it and is being actively developed and used. After a vote
> we've got a consensus to go ahead with the svn move.
> Before the next DUCC release candidate move the source to make it a
> regular uima project like uima sdk and uimaFIT.
>
>
>
> --
> This message was sent by Atlassian JIRA
> (v6.3.4#6332)
>

Re: [VOTE] Release UIMA-AS 2.8.1 RC6

[Lou DeGenaro]

- Downloaded from bin - OK
- Checked README, RELEASE_NOTES, LICENSE and NOTICE - OK - slf4j issue fixed
- Set JAVA_HOME to use 1.7.0
- Set UIMA_HOME
- Started Broker using startBroker.sh - OK
- ran OK:
adjustExamplePath.sh
   ./validateDescriptor.sh
../examples/descriptors/tutorial/ex4/MeetingDetectorTAE.xml




On Thu, May 12, 2016 at 10:26 AM, Jaroslaw Cwiklik 
wrote:

> Hi,
>
> the UIMA-AS 2.8.1 release candidate 6 is ready for voting.
>
> This version contains the following fixes for problems found in RC5:
>
> - Fixed LICENSE file - license text was missing for some of the
> dependencies.
>
> The list of changes in Jira:
>
> https://issues.apache.org/jira/issues/?jql=fixVersion%20%3D%202.8.1AS%20AND%20project%20%3D%20UIMA
>
> The source and binary zip/tars and the Eclipse update site are staged to
> https://dist.apache.org/repos/dist/dev/uima/uima-as/2.8.1/rc6/
>
> The Maven artifacts are here:
> https://repository.apache.org/content/repositories/orgapacheuima-1093
>
> The SVN tags are here:
> http://svn.apache.org/repos/asf/uima/uima-as/tags/uima-as-2.8.1/
>
> and for the Eclipse Update Site:
> https://dist.apache.org/repos/dist/dev/uima/eclipse-update-site/uima-as/
>
> See http://uima.apache.org/testing-builds.html for suggestions on how to
> test
> release candidates.
>
> Please vote on the release:
>
> [ ] +1 OK to release
> [ ] 0   Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks
>

Re: [VOTE] Release UIMA-AS 2.8.1 RC6

[Lou DeGenaro]

Sorry, previous was sent prematurely.

- Downloaded from bin - OK
- Checked README, RELEASE_NOTES, LICENSE and NOTICE - OK - slf4j issue fixed
- Set JAVA_HOME to use 1.7.0
- Set UIMA_HOME
- Started Broker using startBroker.sh - OK
- ran OK:
   adjustExamplePath.sh
   validateDescriptor.sh
../examples/descriptors/tutorial/ex4/MeetingDetectorTAE.xml
   runRemoteAsyncAE.sh tcp://localhost:61616 MeetingDetectorTaeQueue
-d Deploy_MeetingDetectorTAE.xml -c
$UIMA_HOME/examples/descriptors/collection_reader/FileSystemCollectionReader.xml

[X] +1 OK to release

Lou.

​

Re: [VOTE] Release UIMA-DUCC 2.1.0 RC1

[Lou DeGenaro]

- fetched uima-ducc-2.1.0-bin.tar.gz

and installed as potential new apache ducc demo
- WebServer DUCC Book link is broken

[X] -1 Not OK to release

Lou.





On Thu, Jul 14, 2016 at 2:00 PM, Jaroslaw Cwiklik 
wrote:

> Hi,
>
> the UIMA-DUCC 2.1.0 release candidate 1 is ready for voting.
>
> This version contains the following new features and fixes:
>
> - DUCC framework runs on Java 7 or Java 8, each supporting applications
> using either JRE
> - Ubuntu and RHEL 7 support
> - cgroup enhancements
>- uses standard cgroup organization
>- supports cgroup swappiness setting, restricting any swapping if
> desired
> - DUCC state and history storage moved from flat files to Cassandra DB,
> reducing storage size 5x
> - ships with latest UIMA-AS v2.8.1
> - ships with recent AMQ v5.13.2
> - DUCC's UIMA-AS services support failover: and ssl: connectors
> - many ducc webpage improvements
> - clear user display of ducc classes and relation to machines
> - robust handling of dynamic changes to ducc class and nodepool definitions
> - full support of nodepools with different quantum
> - DUCC broker access restricted to user ducc
> - eliminate need for user home directories located on a shared filesystem
> - built in Job error handler programmable per job
> - migration utility for DUCC updates
> - change to vary_off behavior to facilitate cluster management
> - horizontal stacking of service instance allocations
> - java-viaducc improvements including separation of stdout from stderr
> responses - an alert banner is displayed on ducc-mon pages if any daemons
> are down
>
>
> The list of all changes in Jira:
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.1.0-Ducc%22%20ORDER%20BY%20key%20ASC
>
> The source and binary zip/tars and the Eclipse update site are staged to
> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.1.0/RC1/
>
> The Maven artifacts are here:
> https://repository.apache.org/content/repositories/orgapacheuima-1099
>
> The SVN tags are here:
> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.1.0/
>
> See http://uima.apache.org/testing-builds.html for suggestions on how to
> test
> release candidates.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ] 0   Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks.
>

Re: [VOTE] Release UIMA-DUCC 2.1.0 RC2

[Lou DeGenaro]

* fetched tar.gz from link
* employed approximate instructions on wiki
[https://cwiki.apache.org/confluence/display/UIMA/DUCC#DUCC-t2] to
deploy demo (currently http://uima-ducc-vm2.apache.org:42133/jobs.jsp
which in the midst of migrating
http://uima-ducc-demo.apache.org:42133/jobs.jsp)
* spot check WebServer links
* spot check README, LICENCE, NOTICE, RELEASE_NOTES.html

Not found in LICENSE file:

> jetty-all-9.2.13.v20150730.jar
> log4j-1.2.17.jar (but .16 is?)
> commons-logging-1.2.jar (but 1.2.0 is?)
> slf4j-api-1.7.13.jar
> slf4j-log4j12-1.7.13.jar
> guava-14.0.1.jar (but 16.0 is?)
> jna-4.2.2.jar (but 4.0.0 is?)
> javax.annotation-api-1.2.jar

> jgrowl

[X] -1 Not OK to release

Lou.


On Mon, Jul 18, 2016 at 4:31 PM, Jaroslaw Cwiklik 
wrote:

> Hi,
>
> the UIMA-DUCC 2.1.0 release candidate 2 is ready for voting.
>
> This version contains the following changes for issues found in RC1:
>
> - Updated README
> - Updated binary LICENSE
> - Fixes missing DUCC documentation in binary distribution
> - Fixes via-ducc incorrect handling of -jar option
>
> The list of all changes in Jira:
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.1.0-Ducc%22%20ORDER%20BY%20key%20ASC
>
> The source and binary zip/tars are staged here:
> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.1.0/RC2/
>
> The Maven artifacts are here:
> https://repository.apache.org/content/repositories/orgapacheuima-1100
>
> The SVN tags are here:
> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.1.0/
>
> See http://uima.apache.org/testing-builds.html for suggestions on how to
> test
> release candidates.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ] 0   Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks.
>

Re: [VOTE] Release UIMA-DUCC 2.1.0 RC3

[Lou DeGenaro]

When I follow this link

The source and binary zip/tars are staged here:
https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.1.0/RC3/

I get the response

Forbidden

You don't have permission to access
/repos/dist/dev/uima/uima-ducc/2.1.0/RC3/ on this server.


Lou.



On Wed, Jul 20, 2016 at 2:43 PM, Jaroslaw Cwiklik 
wrote:

> Hi,
>
> the UIMA-DUCC 2.1.0 release candidate 3 is ready for voting.
>
> This version contains the following changes for issues found in RC2:
>
> - Updated RELEASE_NOTES
> - Updated binary LICENSE
>
> The list of all changes in Jira:
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.1.0-Ducc%22%20ORDER%20BY%20key%20ASC
>
> The source and binary zip/tars are staged here:
> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.1.0/RC3/
>
> The Maven artifacts are here:
> https://repository.apache.org/content/repositories/orgapacheuima-1101
>
> The SVN tags are here:
> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.1.0/
>
> See http://uima.apache.org/testing-builds.html for suggestions on how to
> test
> release candidates.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ] 0   Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks.
>

Re: [VOTE] Release UIMA-DUCC 2.1.0 RC3

[Lou DeGenaro]

Ok, so now it works.  Dunno why!?

Lou.

On Thu, Jul 21, 2016 at 6:16 AM, Lou DeGenaro 
wrote:

> When I follow this link
>
> The source and binary zip/tars are staged here:
> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.1.0/RC3/
>
> I get the response
>
> Forbidden
>
> You don't have permission to access
> /repos/dist/dev/uima/uima-ducc/2.1.0/RC3/ on this server.
>
>
> Lou.
>
>
>
> On Wed, Jul 20, 2016 at 2:43 PM, Jaroslaw Cwiklik 
> wrote:
>
>> Hi,
>>
>> the UIMA-DUCC 2.1.0 release candidate 3 is ready for voting.
>>
>> This version contains the following changes for issues found in RC2:
>>
>> - Updated RELEASE_NOTES
>> - Updated binary LICENSE
>>
>> The list of all changes in Jira:
>>
>> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.1.0-Ducc%22%20ORDER%20BY%20key%20ASC
>>
>> The source and binary zip/tars are staged here:
>> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.1.0/RC3/
>>
>> The Maven artifacts are here:
>> https://repository.apache.org/content/repositories/orgapacheuima-1101
>>
>> The SVN tags are here:
>> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.1.0/
>>
>> See http://uima.apache.org/testing-builds.html for suggestions on how to
>> test
>> release candidates.
>>
>> Please vote on release:
>>
>> [ ] +1 OK to release
>> [ ] 0   Don't care
>> [ ] -1 Not OK to release, because ...
>>
>> Thanks.
>>
>
>

Re: [VOTE] Release UIMA-DUCC 2.1.0 RC3

[Lou DeGenaro]

* fetched tar.gz from link
* employed approximate instructions on wiki
[https://cwiki.apache.org/confluence/display/UIMA/DUCC#DUCC-t2] to
deploy demo (currently http://uima-ducc-vm2.apache.org:42133/jobs.jsp
which in the midst of migrating
http://uima-ducc-demo.apache.org:42133/jobs.jsp)
* spot check WebServer links
* spot check README, LICENCE, NOTICE, RELEASE_NOTES.html

java-viaducc improvements including separation of stdout from
stderr respoonses 

"respoonses" is spelled wrong in RELEASE_NOTES.


[X] +1 OK to release

Lou.


On Wed, Jul 20, 2016 at 2:43 PM, Jaroslaw Cwiklik 
wrote:

> Hi,
>
> the UIMA-DUCC 2.1.0 release candidate 3 is ready for voting.
>
> This version contains the following changes for issues found in RC2:
>
> - Updated RELEASE_NOTES
> - Updated binary LICENSE
>
> The list of all changes in Jira:
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.1.0-Ducc%22%20ORDER%20BY%20key%20ASC
>
> The source and binary zip/tars are staged here:
> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.1.0/RC3/
>
> The Maven artifacts are here:
> https://repository.apache.org/content/repositories/orgapacheuima-1101
>
> The SVN tags are here:
> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.1.0/
>
> See http://uima.apache.org/testing-builds.html for suggestions on how to
> test
> release candidates.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ] 0   Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks.
>

Re: [VOTE] Release UIMA-DUCC 2.1.0 RC4

[Lou DeGenaro]

- Installed from tar.gz (to uima-ducc-demo.apache.org:42133) - OK
- Installed demo, which exercises Jobs, Services, and Reservations - OK
- Spot checked WebServer links, including DuccBook - OK
- Spot checked README, RELEASE_NOTES, LICENSE and NOTICE - OK

[x] +1 OK to release

Lou.

On Thu, Jul 28, 2016 at 4:50 PM, Jaroslaw Cwiklik 
wrote:

> Hi,
>
> the UIMA-DUCC 2.1.0 release candidate 4 is ready for voting.
>
> This version contains the following changes for issues found in RC3:
> - Missing issuesFixed in source
> - Fixed LICENSE and NOTICE issues
> - Fixed poms to use consitent naming across projects
> - Updated docs to document little-endian support and improved migration
> instructions
>
> The list of all changes in Jira:
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%222.1.0-Ducc%22%20ORDER%20BY%20key%20ASC
>
> The source and binary zip/tars and the Eclipse update site are staged to
> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.1.0/RC4/
>
> The Maven artifacts are here:
> https://repository.apache.org/content/repositories/orgapacheuima-1103
>
> The SVN tags are here:
> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.1.0/
>
> See http://uima.apache.org/testing-builds.html for suggestions on how to
> test
> release candidates.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ] 0   Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks.
>

Re: [VOTE] Release UIMA-AS 2.9.0 RC1

[Lou DeGenaro]

- OK: on Red Hat Linux downloaded & unzipped uima-as-2.9.0-bin.zip

- OK: followed README instructions to
  > setup environment variables
  > adjust examples paths
  > start broker
  > deploy & utilize async service (MeetingDetector)
  > Completed 8 documents; 17165 characters
  > Time Elapsed : 2553 ms
- OK: README & Release Notes
- OK: License & Notice
- OK: cursory inspection of Jira Report
- OK: cursory inspection of uima_async_scaleout.html

+1 Approve the release

Lou.

On Mon, Nov 28, 2016 at 1:53 PM, Jaroslaw Cwiklik 
wrote:

> - Signatures (md5, sha1) - OK
>
> - Built from source - OK
>
> - jira-report - OK - the list is not sorted on a Key (JIRA#)
>
> - README & Release Notes- OK
>
> - License & Notice - OK
>
> - UIMA-AS Eclipse plugins - OK
>
> - UIMA-AS Eclipse example runtime configurations - OK
>
> - Example scripts from /bin: startBroker, deployAsyncService,
> runRemoteAsyncAE - OK
>
>
> +1 Approve the release
>
> - jerry
>
> On Wed, Nov 23, 2016 at 2:01 PM, Burn Lewis  wrote:
>
> > But what is the purpose of this test?   Currently it expects the major
> > version numbers to match, and prints a misleading error message if they
> > don't.  Are we sure that UIMA-AS 2.9.0 will work with any V2 UIMA?  Or
> > should the test be for the version of core UIMA that UIMA-AS includes,
> and
> > has been tested with?  Perhaps that would be too restrictive since it may
> > work with older or newer versions, so should we just generate a warning
> if
> > only the major versions match?
> >
> > Burn
> >
> >
> > On Tue, Nov 22, 2016 at 4:50 PM, Jaroslaw Cwiklik 
> > wrote:
> >
> > > I did find a "small" bug in the code that logs a message when versions
> > > don't match.
> > > Its only effecting logging in this particular case where I chose v
> 3.0.0
> > as
> > > the next
> > > version for UIMA-AS.
> > >
> > > I think the RC1 should not be taken down for this reason. In RC1 both
> > uimaj
> > > and
> > > uima-as are at 2.9.0 and auto generated code (UimaVersion &
> > UimaAsVersion)
> > > are
> > > correct.
> > >
> > > Ultimately its up to you testers to decide if this is critical bug
> > > deserving -1.
> > >
> > >
> > > Jerry
> > >
> > > On Tue, Nov 22, 2016 at 4:21 PM, Marshall Schor  wrote:
> > >
> > > > Ah - OK.  So you're saying, this RC is (possibly) good, and we should
> > go
> > > > ahead
> > > > and test it, etc., and you'll fix the "next" release RC later, right?
> > > >
> > > > -Marshall
> > > >
> > > >
> > > > On 11/22/2016 3:57 PM, Jaroslaw Cwiklik wrote:
> > > > > Marshall, this is caused by me choosing next version to be
> > > 3.0.0-SNAPSHOT
> > > > > when doing
> > > > >
> > > > > mvn release:prepare -DautoVersionSubmodules
> > > > >
> > > > > The UIMA-AS code does this check:
> > > > >
> > > > >   if (UimaAsVersion.getMajorVersion() !=
> > > > UimaVersion.getMajorVersion())
> > > > > {
> > > > >
> > > > > I will choose 2.9.1 as the next version when doing RC2 :)
> > > > >
> > > > > Jerry
> > > > >
> > > > > On Tue, Nov 22, 2016 at 3:40 PM, Marshall Schor 
> > wrote:
> > > > >
> > > > >> hmmm, the latest Jenkins build is showing 4 test failures?
> > > > >>
> > > > >> -Marshall
> > > > >>
> > > > >>
> > > >
> > > >
> > >
> >
>

Re: [jira] [Created] (UIMA-5227) Customized user logging fails in DUCC Job Driver

[Lou DeGenaro]

The JD claims to employ the user's classpath to the build the user's error
handler in the initialize() method of
org.apache.uima.ducc.container.jd.classload.ProxyJobDriverErrorHandler.

Does -Djava.util.logging.config.class get put into the Job specification's
user "classpath"?


On Wed, Dec 21, 2016 at 8:07 AM, Burn Lewis (JIRA) 
wrote:

> Burn Lewis created UIMA-5227:
> 
>
>  Summary: Customized user logging fails in DUCC Job Driver
>  Key: UIMA-5227
>  URL: https://issues.apache.org/jira/browse/UIMA-5227
>  Project: UIMA
>   Issue Type: Bug
>   Components: DUCC
> Reporter: Burn Lewis
> Priority: Minor
>  Fix For: 2.2.0-Ducc
>
>
> When a user specifies a custom logging configurator via
> -Djava.util.logging.config.class it fails in the JD as its ToLog class
> creates a logger using the DUCC classpath which does not have the users
> class.
>
>
>
> --
> This message was sent by Atlassian JIRA
> (v6.3.4#6332)
>

Re: [VOTE] Release UIMA-DUCC 2.2.0 RC1

[Lou DeGenaro]

No, this admin command never worked and was never documented in 2.1.0.
Having it appear in 2.1.0 was an escape from quality control - harmless,
but slightly embarrassing.  So it was removed from 2.2.0 build, but kept in
source with the hope that it may be re-instated in a future release.

On Thu, Dec 22, 2016 at 10:03 AM, Marshall Schor  wrote:

> I noticed this release moved a command, q_process, from admin to
> admin.future
> (in source-release) and deleted it from the binary distribution.  Will any
> users
> notice? If not, then this is fine.  If yes, then some mention of it in
> release
> notes may be useful.
>
> -Marshall
>

Re: [VOTE] Release UIMA-DUCC 2.2.0 RC1

[Lou DeGenaro]

Yep, looks like bin/tools are not documented.

On Thu, Dec 22, 2016 at 10:14 AM, Marshall Schor  wrote:

> Maybe missing documentation?
>
> I see some new tooling for ducc is provided, in the folder bin/tools.  At
> least
> one of these (ducc_status) and maybe others were not present in the
> previous
> release. I searched for documentation about this, but didn't see any
> (maybe I
> missed it?).
>
> -Marshall
>

Re: [VOTE] Release UIMA-DUCC 2.2.0 RC1

[Lou DeGenaro]

Sorry for being terse earlier.  These need to be documented, as you point
out - thx!  Jira 5232 has been opened against 2.2.0-Ducc to address this
shortcoming.

On Fri, Dec 23, 2016 at 10:00 AM, Marshall Schor  wrote:

> If the DUCC team feel these need to be documented, please do so in a next
> release candidate.  Otherwise, please say why these don't need to be
> documented :-)
>
> -Marshall
>
>
> On 12/22/2016 11:06 AM, Lou DeGenaro wrote:
> > Yep, looks like bin/tools are not documented.
> >
> > On Thu, Dec 22, 2016 at 10:14 AM, Marshall Schor  wrote:
> >
> >> Maybe missing documentation?
> >>
> >> I see some new tooling for ducc is provided, in the folder bin/tools.
> At
> >> least
> >> one of these (ducc_status) and maybe others were not present in the
> >> previous
> >> release. I searched for documentation about this, but didn't see any
> >> (maybe I
> >> missed it?).
> >>
> >> -Marshall
> >>
>
>

Re: Load on uima-ducc-vm2

[Lou DeGenaro]

Thanks!

I found 3 renegade process that were likely left over from a previous DUCC
lifetime.  I killed them and top looks better now.

Lou.

On Tue, Jan 17, 2017 at 3:43 PM, Pono Takamori  wrote:

> Greetings from Infra,
> The load on the uima-ducc-vm2.apache.org seems to have been about max
> since 05:00 Dec. 23rd.  There are some outstanding java processes
> which seem to be eating resources.  Just checking in that this is
> expected behavior, maybe a bug, or possibly you need more resources.
> Let me know how I can help!
>
> Cheers,
> -Pono
>

Re: [VOTE] Release UIMA-DUCC 2.2.0 RC2

[Lou DeGenaro]

I contend that MessagesTest and the corresponding internationalization that
I started during DUCC's gestation period years ago should be removed or
uniformly employed, most especially in the UI.  This is otherwise invisible
to the enduser and administrator.  Not a blocker, and non-trivial to
remove.

If there is no objection, I will open a Jira to remove this a future
release as it adds no value in its present state.

Lou.

On Thu, Jan 26, 2017 at 9:30 AM, Marshall Schor  wrote:

> I've finished a partial review of DUCC 2.2.0:
>
> signatures OK
>
> build from source (on Windows, exclude latex) OK.
>
> issues fixed - one issue open 5157  not a blocker - forgot to get closed.
>
> compare svn and source-release - issue with "issuesFixed". - not a blocker
>
> License Notice: OK
>
> I'm voting +1 to release, after others confirm install and operational
> functionality :-).
>
> -Marshall
>
>
>
>
> On 1/26/2017 9:17 AM, Jaroslaw Cwiklik wrote:
> > The issuesFixed was added to the svn sometime ago to deal with changes
> > plugin 100 issues limitation. I was not able to figure out how to deal
> with
> > that and after consultation with you I've delivered this folder to the
> svn
> > as a short term fix. We aim to be more agile now :), and post releases
> more
> > often. I just removed this folder from the svn. The issuesFixes is auto
> > generated by mvn during a build and should not be in the svn.
> >
> > Not sure what to say about MessagesTest. Perhaps others can chime in.
> >
> > Indeed I forgot to modify TOC in the Release Notes. Will fix that either
> in
> > the next RC or for the next release.
> >
> > -jerry
> >
> > On Wed, Jan 25, 2017 at 5:08 PM, Marshall Schor  wrote:
> >
> >> The Release Notes file added 2 new h2. level topics:  Migration from a
> >> Prior
> >> Release, and Limitations.
> >> The table of contents is missing the last section.
> >>
> >> not a blocker...
> >>
> >> -Marshall
> >>
>
>

Re: [VOTE] Release UIMA-DUCC 2.2.0 RC3

[Lou DeGenaro]

* downloaded uima-ducc-2.2.0-bin.tar.gz

* employed wiki instructions to replace uima-ducc-demo
https://cwiki.apache.org/confluence/display/UIMA/DUCC#DUCC-t2
* demo running as expected since Friday
* checked daemon logs and saw no execptions
* discovered System->Broker page incomplete with Preference -> Table Style
= Scroll
 - not a regression
 - missing data available with Preference -> Table Style = Classic
 - not a blocker to release
* discovered DUCC Book missing option "head = { or, pm, rm, sm, ws, db,
broker }" for start_ducc and stop_ducc
 - info is available when --help is specified
 - not a blocker to release
* spot checked INSTALL, LICENCE, NOTICE, README, RELEASE_NOTES.html

[X] +1 OK to release

Lou.

On Mon, Feb 6, 2017 at 9:57 AM, Burn Lewis  wrote:

> + downloaded & checked signatures
> + installed binary on a multi-node cluster and ran a large experiment with
> jobs, services & APs
> + also ran a large set of test jobs & services on many simulated nodes
> (unprivileged)
> + built from source and checked signatures
> + ran that build with simulated nodes
> + spot checked install, readme, release notes, issues fixed, licenses &
> notices
> noted that the copyright dates for UIMA & UIMA-AS are older than in
> 2.1.0
>
> [+1]
>
> Burn
>
> On Fri, Feb 3, 2017 at 10:29 AM, Jaroslaw Cwiklik 
> wrote:
>
> > Hi,
> >
> > the UIMA-DUCC 2.2.0 release candidate 3 is ready for voting.
> >
> > Issues fixed since RC2:
> > - Fixed Agent bug preventing some JPs from stopping
> > - Fixed duplicate daemon detector to allow virtual agents to start
> > - Fixed DD based JPs to send FailedInitialization on init failure
> > - Fixed WS to show correct values in PgIn
> > - Fixed WS status for Broker and Database
> > - Fixed OR to propagate extended reason to the WS
> > - Downgraded JNA from 4.2.2 to 4.0.0
> >
> >
> > The list of changes in Jira:
> > https://issues.apache.org/jira/issues/?jql=fixVersion%
> > 20%3D%202.2.0-Ducc%20AND%20project%20%3D%20UIMA
> >
> > The source and binary zip/tars are staged to
> > https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.2.0/RC3/
> >
> > The Maven artifacts are here:
> > https://repository.apache.org/content/repositories/orgapacheuima-1122
> >
> > The SVN tags are here:
> > http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.2.0/
> >
> > See http://uima.apache.org/testing-builds.html for suggestions on how to
> > test release candidates.
> >
> > Please vote on release:
> >
> > [ ] +1 OK to release
> > [ ] 0   Don't care
> > [ ] -1 Not OK to release, because ...
> >
> > Thanks
> > - Jerry Cwiklik
> >
>

Re: UIMA Website - replacing the "forum" view of our mailing list?

[Lou DeGenaro]

The LHS seems inconsistent between, for example, between home page and
mailing list page in terms of order and content.

On home page order is: General, community, scaleout...
On mailing list page order is: General, scaleout, components...

Under Community, HipChat appears on home page but not on mailing list page.

Lou.

On Mon, Mar 20, 2017 at 10:19 AM, Marshall Schor  wrote:

> I updated the mailing list web page (not the left hand nav bar), to remove
> gmane, and add the pony mail links.
>
> http://uima.apache.org/mail-lists.html
>
> look in the table at the right-most column, where it says "Forum view".
>
> -Marshall
>
>
> On 3/19/2017 5:20 PM, Marshall Schor wrote:
> > ok, I'm updating the site for alpha02 release, and I'll update the forum
> view to
> > pony mail (lists.apache.org).
> >
> > -Marshall
> >
> > On 3/19/2017 4:22 PM, Richard Eckart de Castilho wrote:
> >> PonyMail or MarkMail would be options. I would tend towards PonyMail
> >> since it is provided by Apache itself.
> >>
> >> -- Richard
> >>
> >>> On 19.03.2017, at 21:01, Marshall Schor  wrote:
> >>>
> >>> Hi -
> >>>
> >>> I see from history that Richard removed the left-hand links to gmane -
> because
> >>> that site has been shut down.
> >>>
> >>> Is there anything to replace this with that's reasonable?
> >>> e.g.
> >>>
> >>> https://lists.apache.org/list.html?u...@uima.apache.org
> >>>
> >>> -Marshall
> >
>
>

Re: UIMA Website - replacing the "forum" view of our mailing list?

[Lou DeGenaro]

Never mind.  Seems that clearing my Firefox cache fixed the problem, so
this was user error...kind of.

Lou.

On Mon, Mar 20, 2017 at 1:55 PM, Marshall Schor  wrote:

> thanks for noticing, will investigate...
>
> -M
>
>
> On 3/20/2017 11:14 AM, Lou DeGenaro wrote:
> > The LHS seems inconsistent between, for example, between home page and
> > mailing list page in terms of order and content.
> >
> > On home page order is: General, community, scaleout...
> > On mailing list page order is: General, scaleout, components...
> >
> > Under Community, HipChat appears on home page but not on mailing list
> page.
> >
> > Lou.
> >
> > On Mon, Mar 20, 2017 at 10:19 AM, Marshall Schor  wrote:
> >
> >> I updated the mailing list web page (not the left hand nav bar), to
> remove
> >> gmane, and add the pony mail links.
> >>
> >> http://uima.apache.org/mail-lists.html
> >>
> >> look in the table at the right-most column, where it says "Forum view".
> >>
> >> -Marshall
> >>
> >>
> >> On 3/19/2017 5:20 PM, Marshall Schor wrote:
> >>> ok, I'm updating the site for alpha02 release, and I'll update the
> forum
> >> view to
> >>> pony mail (lists.apache.org).
> >>>
> >>> -Marshall
> >>>
> >>> On 3/19/2017 4:22 PM, Richard Eckart de Castilho wrote:
> >>>> PonyMail or MarkMail would be options. I would tend towards PonyMail
> >>>> since it is provided by Apache itself.
> >>>>
> >>>> -- Richard
> >>>>
> >>>>> On 19.03.2017, at 21:01, Marshall Schor  wrote:
> >>>>>
> >>>>> Hi -
> >>>>>
> >>>>> I see from history that Richard removed the left-hand links to gmane
> -
> >> because
> >>>>> that site has been shut down.
> >>>>>
> >>>>> Is there anything to replace this with that's reasonable?
> >>>>> e.g.
> >>>>>
> >>>>> https://lists.apache.org/list.html?u...@uima.apache.org
> >>>>>
> >>>>> -Marshall
> >>
>
>

Re: [VOTE] Release UIMA-DUCC 2.2.1 RC1

[Lou DeGenaro]

I vote -1 for 3 reasons:

1. ducc_ling does not work for sim-cluster
2. WS has NPE for some non-existent job requests than can be easily fixed
3. Jobs with service dependencies are not working under Apache DUCC demo

Lou.

On Thu, Jul 27, 2017 at 3:45 PM, Jaroslaw Cwiklik 
wrote:

> Hi,
>
> the UIMA-DUCC 2.2.1 release candidate 1 is ready for voting.
>
> This version contains the following fixes:
>
> -The userid of a privileged DUCC installation does not have to be "ducc"
> -ducc-mon login can be used on systems where users do not have password
> login
> -The DUCC head-node daemons may be moved to another host without breaking
> working applications
> -The deployment descriptor for a UIMA-AS service can be loaded from the
> classpath
> -Interactive applications run correctly with viaducc (fixed lost inputs)
> -Files created by DUCC jobs inherit the permissions of the launching
> shell's umask.
> -DUCC performance breakdown for scaled synchronous pipelines is now correct
>
> The list of changes in Jira:
> https://issues.apache.org/jira/issues/?jql=fixVersion%
> 20%3D%202.2.1-Ducc%20AND%20project%20%3D%20UIMA
>
> The source and binary zip/tars are staged to
> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.2.1/RC1/
>
> The Maven artifacts are here:
> https://repository.apache.org/content/repositories/orgapacheuima-1147
>
> The SVN tags are here:
> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.2.1/
>
> See http://uima.apache.org/testing-builds.html for suggestions on how to
> test release candidates.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ] 0   Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks
> - Jerry Cwiklik
>

Re: [VOTE] Release UIMA-DUCC 2.2.1 RC2

[Lou DeGenaro]

> Downloaded uima-ducc-2.2.1-bin.tar.gz

: OK
> Installed and started Apache DUCC demo
http://uima-ducc-demo.apache.org:42133/jobs.jsp :
   - Jobs, Services, Reservations running : OK
   - WS
  - Jobs + Details: OK
  - Reservations + Details : OK
  - Services + Details : OK
  - System Administration, Broker, Classes, Daemons, Machines : OK
  * System DuccBook : MISSING!
  - System Viz : OK
> Spot check LICENSE file : OK
   - note typo:
  (webserver>navigation
   - not inconsistency:
  (Webserver) vs. (webserver)
> Check README file : OK
   - should it also say JAVA_HOME should be set to Java 7 install location
for mvn building?
> Spot check NOTICE file: OK
> Check ReleaseNotes.html : OK, except
   - could not easily find: *ducc_update* script which can be copied from
the UIMA Downloads page

  (maybe it's just me - has to resort to searching the page for
"ducc_update").

Vote on RC2: -1

Lou.





On Tue, Aug 1, 2017 at 7:57 PM, Jaroslaw Cwiklik  wrote:

> Hi,
>
> the UIMA-DUCC 2.2.1 release candidate 2 is ready for voting.
>
> Fixed the following issues found in RC1:
>
> - DUCC now fails to handle fake ids when running a simulated cluster
> - DUCC Web Server (WS) NPE handling JobSpecificationData requests
>
> The list of changes in Jira:
> https://issues.apache.org/jira/issues/?jql=fixVersion%
> 20%3D%202.2.1-Ducc%20AND%20project%20%3D%20UIMA
>
> The source and binary zip/tars are staged to
> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.2.1/RC2/
>
> The Maven artifacts are here:
> https://repository.apache.org/content/repositories/orgapacheuima-1148
>
> The SVN tags are here:
> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.2.1/
>
> See http://uima.apache.org/testing-builds.html for suggestions on how to
> test release candidates.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ] 0   Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks
> - Jerry Cwiklik
>

Re: [VOTE] Release UIMA-DUCC 2.2.1 RC3

[Lou DeGenaro]

> Downloaded uima-ducc-2.2.1-bin.tar.gz

: OK

> Installed and started Apache DUCC demo http://uima-ducc-demo.apache.
org:42133/jobs.jsp :
   - Jobs, Services, Reservations running : OK
   - WS
  - Jobs + Details: OK
  - Reservations + Details : OK
  - Services + Details : OK
  - System Administration, Broker, Classes, Daemons, Machines : OK
  * System DuccBook : OK
  - System Viz : OK

> Spot check LICENSE file : OK
   - note typo:
  (webserver>navigation

> Check README file : OK
   - should it also say JAVA_HOME should be set to Java 7 install location
for mvn building?

> Spot check NOTICE file: OK

> Check ReleaseNotes.html : OK

> Checked DUCC Book for ducc_update information: OK

Vote on RC3 = +1

Lou.


On Wed, Aug 2, 2017 at 6:53 PM, Jaroslaw Cwiklik  wrote:

> Hi,
>
> the UIMA-DUCC 2.2.1 release candidate 3 is ready for voting.
>
> Fixed the following issues found in RC2:
>
> - Missing docs from binary tarball
> - README missing info about JAVA_HOME when running mvn build
> - Typos in LICENSE
> - The ducc_update script not documented in DuccBook
> - RELEASE_NOTES.html section about ducc_update should change
>
> The list of changes in Jira:
> https://issues.apache.org/jira/issues/?jql=fixVersion%
> 20%3D%202.2.1-Ducc%20AND%20project%20%3D%20UIMA
>
> The source and binary zip/tars are staged to
> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.2.1/RC3/
>
> The Maven artifacts are here:
> https://repository.apache.org/content/repositories/orgapacheuima-1149
>
> The SVN tags are here:
> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.2.1/
>
> See http://uima.apache.org/testing-builds.html for suggestions on how to
> test release candidates.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ] 0   Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks
> - Jerry Cwiklik
>

Re: How about moving to GitHub?

[Lou DeGenaro]

I applaud your taking the initiative towards improvement.  I did have
experience perhaps 1.5 years ago using GitHub for another non-Apache
project.  My view was negative, though I could not now recite the precise
issues encountered. My best recollection is that the happy paths worked
fine, but when mistakes were made or problems were encountered recovery was
often difficult and frustrating.  If I am the only objector then I vote 0,
but my gut votes -1.

With respect to speculation, my sense is that switching to GitHub will not
open the floodgates to contributions.

Lou.

On Sat, Aug 26, 2017 at 6:54 AM, Richard Eckart de Castilho 
wrote:

> Hi all,
>
> several Apache projects have moved parts of their development to GitHub,
> some
> more (no more Jira), some less (issues in Jira, code on GitHub).
>
> I have used GitHub more and more lately and find its processes quite
> convenient.
> This mostly revolves around the concept of "pull requests". People submit
> "pull-requests" instead of patches. Based on these, GitHub allows to do:
>
> * convenient code reviews
> * leave comments (line-based as well as general)
> * approve/request changes
> * discuss the pull request as it evolves
> * even exporting a pull request as a patch is easy
>
> Pull requests are essentially branches and it is possible for
> interested developers to:
>
> * check out these branches
> * collaborate on them
> * finally merge them into the master (trunk) branch when ready.
>
> I have also made some experience setting up Jenkins in such a way that
> it can automatically check pull requests:
>
> * an admin developer leaves a comment on a pull request such as
>   "Jenkins, test this please"
> * Jenkins sees this comment and triggers a build of that particular
>   contribution
> * Jenkins updates a "build status" that is visible directly in the
>   discussion thread of the pull request on GitHub. Depending on the
>   result of the build, it will give a red or green light.
> * That is very useful to do an integration test of a contribution after
>   an initial code screening - no need to locally integrate a patch,
>   locally test it, etc. The initial screening can happen by looking at
>   a side-by-side div on the GitHub website - again no local effort.
>
> Finally, GitHub allows to enforce certain processes if desired. E.g.
> one can *optionally* enforce that:
>
> * limit access to specific branches (e.g. master/trunk or maintenance
> branches)
> * no changes can go into master (trunk) without being reviewed first as
>   pull requests (a very strong 4 eyes principle, has ups and downs)
> * pull requests must have passed a Jenkins check before they can be merged
>
> All in all, I believe our project would benefit from moving to GitHub.
> After an initial transition and adaption phase, our processes should become
> smoother, easier and as a result more interactive. The main hurdle for
> us developers would in my view be to learn how do to source code management
> using a distributed approach. The way one has to think about code
> management
> in git (or probably any other DSCM tool) is different from a centralized
> approach like SVN. To take full advantage, branches move into the center
> of attention and branching/merging becomes a much more common operation
> than in SVN. Fortunately, git supports such operations very well.
>
> The hurdle for contributions would also be lower (more and more people are
> familiar with GitHub processes, and already have GitHub accounts.
> Juggling with patch files is (at least for me) a great annoyance).
>
> Of course, there is also the draw back that GitHub may shut down
> at some point (like Google Code did) or start charging or introduce
> terms-of-service that are incompatible with our goals. But due to the
> strong interest of various Apache projects on using the GitHub
> infrastructure,
> INFRA has been working on various integrations e.g. mirroring GitHub
> repositories (not sure how they handle it when Apache projects use GitHub
> issues though). Also at present GitHub appears to be very
> community-oriented
> when it comes to the terms-of-service, e.g. making text updates available
> to
> the community early, soliciting and incorporating community feedback, etc.
>
> On the bottom line, my personal judgement at this point in time is, that
> the
> benefits of smoother process and easier community involvement seriously
> outweigh such risks and that many of the risks have been mitigated by INFRA
> anyway.
>
> What do you think?
>
> Best,
>
> -- Richard
>
>

Re: How about moving to GitHub?

[Lou DeGenaro]

Looking back at e-mails from 2015 on this topic, we used Git & Gerrit.
Even merge conflicts were sometimes a nightmare.  It was just 2 of us, and
we got stuck more than once needing expert help.

With respect to the DUCC part of UIMA, things seems fine to me the way they
are using Jira.  But then again, we have very few patches (if any).  My
vote 0 is unchanged, so far.

Lou.

On Mon, Aug 28, 2017 at 6:10 AM, Peter Klügl 
wrote:

> +1 from me
>
>
> At work, we use mercurial which is probably quite similar to git. I had
> some hard to time adapt and learn at the beginning, but now I would not
> want to miss it. I find it much simpler to develop stuff in parallel.
>
> The last weeks, I gained some experience with GitHub and git
> (dkpro-core). It took me a bit to get "productive", but after a short
> while it was OK. For me, it's important to find the right tooling.
>
> A switch to GitHub will not open the floodgates, but it could lower the
> hurdle for contributions, and even some more contributions would really
> be good. Just speaking for ruta: the first release was in March 2013 and
> I am still the only active developer. That's not a desirable situation.
>
> Peter
>
>
> Am 26.08.2017 um 14:47 schrieb Lou DeGenaro:
> > I applaud your taking the initiative towards improvement.  I did have
> > experience perhaps 1.5 years ago using GitHub for another non-Apache
> > project.  My view was negative, though I could not now recite the precise
> > issues encountered. My best recollection is that the happy paths worked
> > fine, but when mistakes were made or problems were encountered recovery
> was
> > often difficult and frustrating.  If I am the only objector then I vote
> 0,
> > but my gut votes -1.
> >
> > With respect to speculation, my sense is that switching to GitHub will
> not
> > open the floodgates to contributions.
> >
> > Lou.
> >
> > On Sat, Aug 26, 2017 at 6:54 AM, Richard Eckart de Castilho <
> r...@apache.org>
> > wrote:
> >
> >> Hi all,
> >>
> >> several Apache projects have moved parts of their development to GitHub,
> >> some
> >> more (no more Jira), some less (issues in Jira, code on GitHub).
> >>
> >> I have used GitHub more and more lately and find its processes quite
> >> convenient.
> >> This mostly revolves around the concept of "pull requests". People
> submit
> >> "pull-requests" instead of patches. Based on these, GitHub allows to do:
> >>
> >> * convenient code reviews
> >> * leave comments (line-based as well as general)
> >> * approve/request changes
> >> * discuss the pull request as it evolves
> >> * even exporting a pull request as a patch is easy
> >>
> >> Pull requests are essentially branches and it is possible for
> >> interested developers to:
> >>
> >> * check out these branches
> >> * collaborate on them
> >> * finally merge them into the master (trunk) branch when ready.
> >>
> >> I have also made some experience setting up Jenkins in such a way that
> >> it can automatically check pull requests:
> >>
> >> * an admin developer leaves a comment on a pull request such as
> >>   "Jenkins, test this please"
> >> * Jenkins sees this comment and triggers a build of that particular
> >>   contribution
> >> * Jenkins updates a "build status" that is visible directly in the
> >>   discussion thread of the pull request on GitHub. Depending on the
> >>   result of the build, it will give a red or green light.
> >> * That is very useful to do an integration test of a contribution after
> >>   an initial code screening - no need to locally integrate a patch,
> >>   locally test it, etc. The initial screening can happen by looking at
> >>   a side-by-side div on the GitHub website - again no local effort.
> >>
> >> Finally, GitHub allows to enforce certain processes if desired. E.g.
> >> one can *optionally* enforce that:
> >>
> >> * limit access to specific branches (e.g. master/trunk or maintenance
> >> branches)
> >> * no changes can go into master (trunk) without being reviewed first as
> >>   pull requests (a very strong 4 eyes principle, has ups and downs)
> >> * pull requests must have passed a Jenkins check before they can be
> merged
> >>
> >> All in all, I believe our project would benefit from moving to GitHub.
> >> After an initial transition and adaption phase, our processes should
> 

Re: [VOTE] Release UIMA-DUCC 2.2.1 RC5

[Lou DeGenaro]

Downloaded uima-ducc-2.2.1-bin.tar.gz https://dist.apache.org/
repos/dist/dev/uima/uima-ducc/ : OK
Installed as UIMA DUCC demo http://uima-ducc-demo.apache.org:42133/jobs.jsp
following wiki instructions
https://cwiki.apache.org/confluence/display/UIMA/DUCC#DUCC-t2 : OK
Inspected ducc-mon web pages:
   - Jobs, Services, Reservations running : OK
   - WS
  - Jobs + Details: OK
  - Reservations + Details : OK
  - Services + Details : OK
  - System Administration, Broker, Classes, Daemons, Machines : OK
  * System DuccBook : OK
  - System Viz : OK

Spot check LICENSE file : OK
  However, file calls out jna-4.2.2 when in fact jna-4.0.0 is used.

Check README file : OK

Spot check NOTICE file: OK

Check ReleaseNotes.html : OK

Vote on RC5: +1

Lou.

On Tue, Aug 22, 2017 at 3:31 PM, Burn Lewis  wrote:

> - checked binary & source signatures (md5 & sha1)
> - installed binary on a 20-node cluster and ran some large jobs & services
> & APs
> - built from source and installed on a 5-node cluster and ran a job, AP &
> service
> - checked docs (duccbook & javadocs)
> - spot checked readme license notices etc.
>
> [x] +1 OK to release
>
> ~Burn
>
> On Tue, Aug 22, 2017 at 9:15 AM, Jaroslaw Cwiklik 
> wrote:
>
> > - Built from source - OK
> > - Spot checked signatures - OK
> > - issues fixed - OK
> > - Checked README, RELEASE_NOTES,LICENSE and NOTICE - OK
> > - Ran example jobs (1.job + 1.dd.job) . Checked GC stats in agent log -
> OK
> > - Spot checked duccmon tabs & links - OK
> > - Documentation (ducc book, javadoc api) - OK
> >
> > [x] +1 OK to release
> >
> > Jerry
> >
> > On Mon, Aug 21, 2017 at 4:03 PM, Jaroslaw Cwiklik 
> > wrote:
> >
> > > Hi,
> > >
> > > the UIMA-DUCC 2.2.1 release candidate 5 is ready for voting.
> > >
> > > Fixes the following issues found in RC4:
> > >
> > > - GC stats not available sometimes from remote JP
> > >
> > > The list of changes in Jira:
> > > https://issues.apache.org/jira/issues/?jql=fixVersion%20%3D%
> > > 202.2.1-Ducc%20AND%20project%20%3D%20UIMA
> > >
> > > The source and binary zip/tars are staged to
> > > https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.2.1/RC5
> > >
> > > The Maven artifacts are here:
> > > https://repository.apache.org/content/repositories/orgapacheuima-1151
> > >
> > > The SVN tags are here:
> > > http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.2.1/
> > >
> > > See http://uima.apache.org/testing-builds.html for suggestions on how
> to
> > > test release candidates.
> > >
> > > Please vote on release:
> > >
> > > [ ] +1 OK to release
> > > [ ] 0   Don't care
> > > [ ] -1 Not OK to release, because ...
> > >
> > > Thanks
> > > - Jerry Cwiklik
> > >
> >
>

Re: board report input, please

[Lou DeGenaro]

+1

Lou.

On Tue, Sep 12, 2017 at 10:27 AM, Marshall Schor  wrote:

> here's the report:
>
> Board report for Apache UIMA, for September 2017.
>
> Apache UIMA's mission: the creation and maintenance of open-source
> software related to the analysis of unstructured data, guided by the
> UIMA Oasis Standard.
>
> Dates:
> 30 Aug 2017  (new)last release - Apache UIMA DUCC 2.2.1
> 03 May 2016  (no change)  last PMC addition
> 18 Apr 2017  (no change)  last Committer addition
>
> Releases:
>   30 Aug 2017 Apache UIMA DUCC 2.2.1
>   29 Aug 2017 Apache Java SDK 2.10.1
>   24 Jul 2017 Apache UIMA Ruta 2.6.1
>
> Activity:
>   Most major projects (Java SDK, DUCC, uimaFIT, RUTA) are
>   actively being worked on.  The Java SDK version 3 had another
>   release candidate for a beta version.  Testing of version 3
>   with other parts of UIMA is being done, as part of the
>   beta version release candidate.
>
>   Mailing list activity remains moderate, with
>   people from the extended community helping out by answering
>   questions.
>
> Community:
>
>   The community continues to be moderately active.
>
> Issues:
> No Board level issues at this time.
>
> -Marshall Schor
>
> On 8/31/2017 4:00 PM, Marshall Schor wrote:
>
> > Hi everyone,
> >
> > Please append any items for the board report.
> >
> > I already know about and can report on Ruta 2.6.1, Java SDK 2.10.1, and
> DUCC
> > 2.2.1 :-) (we've been busy).
> >
> > -Marshall
> >
> >
>
>

Re: UIMA-AS Client API Change Discussion

[Lou DeGenaro]

No need to deprecate.  The legacy signatures can contain stubs that invoke
submit with the corresponding PO.

Lou.

On Tue, Dec 12, 2017 at 10:33 AM, Marshall Schor  wrote:

> Re: 2nd & 3rd proposal, the client would supply an instance of
> "ProcessingOptions".  I assume the user would create this using new
> ProcessingOptions() and then call the setters it wants.  Or perhaps there
> would
> be a short-hand positional-arg constructor, as well.
>
> Overall, it seems to me that it would be less disruptive to the API
> stability,
> and easier for users to learn and use, to just add an overloaded call
> (proposal
> #1), and not deprecate anything.   I think users would appreciate not
> having to
> update older code.
>
> -Marshall
>
> On 12/11/2017 5:42 PM, Jaroslaw Cwiklik wrote:
> > Hi, I need to extend UIMA-AS client API as defined by
> > interface UimaAsynchronousEngine.java
> >
> > The new feature I've been working on, should allow a client to send a CAS
> > to a specific instance of a service. Such targeting is optional and could
> > be useful to test if a service is viable (processing can be done).
> >
> > To support targeting, a service uses a JMS selector which expects a
> message
> > property that uniquely identifies the service. By default, the selector
> > uses  IP:PID as a unique identifier, but a service deployer can define a
> > custom identifier using -D property on the  service command line.
> >
> > An application client must be able to invoke a UIMA-AS client method
> which
> > takes a CAS, unique id of a service to target, and other properties. The
> > targeting is again optional and used for special use cases.
> >
> > Current UIMA-AS client uses two styles of calls: asynchronous and
> > synchronous
> >
> > sendCAS(CAS cas) - this is an asynchronous style
> >
> > and there are two APIs for synchronous style:
> >
> > sendAndReceive(CAS cas)
> > sendAndReceiveCAS(CAS cas, List
> > componentMetricsList)
> >
> > To support targeting, I can overload the above creating two new methods
> >
> > sendCAS(CAS cas, String serviceTargetId)
> > sendAndReceiveCAS(CAS cas, List
> > componentMetricsList, String serviceTargetId )
> >
> > Another suggestion from Lou Degenaro is to create a new method
> >
> > submit(CAS cas, ProcessingOptions pos);
> >
> > where
> >
> > ProcessingOptions() {
> >   void setAsynchronous();  // if this is not called the default is
> > synchronous
> >   void setMetrics(List list);  //null
> for
> > asynchronous style
> >   void setTargetInstance(String id);  // if not called ,targeting is not
> > done
> > }
> >
> > Burn Lewis commented on this proposal with:
> >
> > "Has the advantage of being extendable without having to add another send
> > method.
> >
> > Could be a map with a specified set of legal keys but a special class is
> > good too ... especially if had extra constructors, e.g. for a targeted
> > async request:
> >
> > submit ( cas, new ProcessingOptions(true, null, "127.0.0.1:12345") )"
> >
> > If submit() is a preferred choice should the sendCAS() and
> sendAndReceive()
> > be deprecated to encourage new code to use more flexible API? Or should
> we
> > just document that submit() is only
> > for targeting services?
> >
> > Are there any suggestions perhaps which should be considered?
> >
> > Jerry
> >
>
>

Re: [VOTE] UIMA-DUCC 2.2.2 RC1

[Lou DeGenaro]

Installed on Apache-DUCC.  Get the following in the job submission script:

id:5 location:26530@uima-ducc-vm2
ERROR: java.lang.NoSuchMethodException: fromJson

java.lang.NoSuchMethodException: fromJson
at
org.apache.uima.ducc.transport.dispatcher.ClassManager.invoke(ClassManager.java:275)
at
org.apache.uima.ducc.transport.dispatcher.DuccEventHttpDispatcherCl.fromJson(DuccEventHttpDispatcherCl.java:110)
at
org.apache.uima.ducc.transport.dispatcher.DuccEventHttpDispatcherCl.dispatchJson(DuccEventHttpDispatcherCl.java:119)
at
org.apache.uima.ducc.cli.DuccMonitor.runInternal(DuccMonitor.java:361)
at org.apache.uima.ducc.cli.DuccMonitor.run(DuccMonitor.java:578)
at
org.apache.uima.ducc.cli.MonitorListener.run(MonitorListener.java:87)
at java.lang.Thread.run(Thread.java:745)

[X ] -1 Not OK to release

Lou.

On Tue, Feb 20, 2018 at 10:45 AM, Jerry Cwiklik  wrote:

> - Spot checked signatures - OK
>
> - Compared DUCC source with SVN tag - OK
>
> - Built from source - OK
>
> - Issues Fixed report - OK
>
> - Spot checked duccbook - OK
>
> - Spot checked README, RELEASE_NOTES, NOTICE, LICENSE - OK
>
> - Spot checked links in duccmon - OK
>
> - Spot checked javadocs API - OK
>
> - Ran 1.job, 1.dd.job, 1.error.job, deployed test service - OK
>
>
> [X] +1 OK to release
>
> Jerry
>
>
> On 2/19/2018 1:22 PM, Jerry Cwiklik wrote:
>
>> Hi,
>>
>> the UIMA-DUCC 2.2.2 release candidate 1 is ready for voting.
>>
>> This version contains the following fixes:
>> -Upgraded UIMA (2.10.2), UIMA-AS (2.10.2), ActiveMQ (5.15.2), Spring
>> Framework (4.3.9), Camel (2.19.3), XStream (1.4.10)
>> -Requires java 8 for builds and runtime
>> -Fixed rogue process detector to prevent truncation of user names
>> -Reduced size of publications between the PM and Agents
>> -Improved Agent's detection of defunct processes
>> -Improved logging of system events
>> -ducc_watcher improvements
>> -Job & managed reservation specs now saved in DB
>> -DUCC Web Server (WS) queries DB too often
>> -DUCC Web Server (WS) should use Resource Manager (RM) calculated value
>> for machine's memory available
>>
>>
>> The list of changes in Jira:
>> https://issues.apache.org/jira/issues/?jql=fixVersion%20%3D%
>> 202.2.2-Ducc%20AND%20project%20%3D%20UIMA
>>
>> The source and binary zip/tars are staged to
>> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.2.2/RC1/
>>
>> The Maven artifacts are here:
>> https://repository.apache.org/content/repositories/orgapacheuima-1186
>>
>> The SVN tags are here:
>> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.2.2/
>>
>> See http://uima.apache.org/testing-builds.html for suggestions on how to
>> test release candidates.
>>
>> Please vote on release:
>>
>> [ ] +1 OK to release
>> [ ] 0   Don't care
>> [ ] -1 Not OK to release, because ...
>>
>> Thanks
>>
>>
> --
> Jerry Cwiklik
> Apache UIMA
>
>

Re: [VOTE] UIMA-DUCC 2.2.2 RC2

[Lou DeGenaro]

Installed on Apache DUCC http://uima-ducc-demo.apache.org:42133/jobs.jsp
which looks better that RC1!  Hope vote later today or tomorrow.

Lou.

On Thu, Feb 22, 2018 at 10:55 AM, Jerry Cwiklik  wrote:

> Hi,
>
> the UIMA-DUCC 2.2.2 release candidate 2 is ready for voting.
>
> This version contains the following fixes since RC1:
>
> - modified service/jp wrapper to log errors from process()
> - modified to remove calls to XMLUtils
> - modified to fix the cause of error preventing launching jobs with
> wait_for_completion flag
>
> The list of changes in Jira:
> https://issues.apache.org/jira/issues/?jql=fixVersion%20%3D%
> 202.2.2-Ducc%20AND%20project%20%3D%20UIMA
>
> The source and binary zip/tars are staged to
> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.2.2/RC2/
>
> The Maven artifacts are here:
> https://repository.apache.org/content/repositories/orgapacheuima-1188
>
> The SVN tags are here:
> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.2.2/
>
> See http://uima.apache.org/testing-builds.html for suggestions on how to
> test release candidates.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ] 0   Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks
>
> --
> Jerry Cwiklik
> Apache UIMA
>
>

Re: [VOTE] UIMA-DUCC 2.2.2 RC2

[Lou DeGenaro]

- Downloaded bin -OK
- Installed on Apache DUCC - OK
- Spot checked various ducc-mon web pages - OK
- Spot checked duccbook - OK, however
  > There is no mention in the Glossary of  Managed Reservations (aka
Arbitrary Processes) - should be added for next release - not a blocker.
  > I propose:
  > \item[Managed Reservation] A DUCC managed reservation comprises an
arbitrary process that is
  > deployed on the computing cluster within a {\em share} assigned by the
DUCC scheduler.
- Spot checked LICENSE - OK, however
  > I could not find geronimo-servlet_3.0_spec-1.0.jar? - if so, should be
removed - not a blocker.
- Spot checked README - OK
- Spot checked RELEASE_NOTES - OK, however
  > note that "wrapt" is misspelled - not a blocker.

[X] +1 OK to release

Lou.

On Thu, Feb 22, 2018 at 10:55 AM, Jerry Cwiklik  wrote:

> Hi,
>
> the UIMA-DUCC 2.2.2 release candidate 2 is ready for voting.
>
> This version contains the following fixes since RC1:
>
> - modified service/jp wrapper to log errors from process()
> - modified to remove calls to XMLUtils
> - modified to fix the cause of error preventing launching jobs with
> wait_for_completion flag
>
> The list of changes in Jira:
> https://issues.apache.org/jira/issues/?jql=fixVersion%20%3D%
> 202.2.2-Ducc%20AND%20project%20%3D%20UIMA
>
> The source and binary zip/tars are staged to
> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.2.2/RC2/
>
> The Maven artifacts are here:
> https://repository.apache.org/content/repositories/orgapacheuima-1188
>
> The SVN tags are here:
> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.2.2/
>
> See http://uima.apache.org/testing-builds.html for suggestions on how to
> test release candidates.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ] 0   Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks
>
> --
> Jerry Cwiklik
> Apache UIMA
>
>

Re: [VOTE] UIMA-DUCC 2.2.2 RC3

[Lou DeGenaro]

 - downloaded uima-ducc-2.2.2-bin.tar.gz

- spot checked issues fixed report - OK
- spot checked README - OK
- spot checked RELEASE_NOTES - OK ("verbose" fix not mentioned, no problem)
- spot checked NOTICE - OK
- spot checked LICENSE - OK

- installed on apache-DUCC - OK
- spot checked ducc-mon links, including DuccBook and its contents - OK
- checked WS log for "verbose" messages and saw none

[X] +1 OK to release

Lou.




On Thu, Mar 1, 2018 at 3:04 PM, Jerry Cwiklik  wrote:

> Hi,
>
> the UIMA-DUCC 2.2.2 release candidate 3 is ready for voting.
>
> This version contains the following fixes since RC2:
>
> - DUCC Web Server (WS) logs queries to DB too verbosely
> - DUCC may restart failing custom services forever
> - check_ducc & start_ducc problem when Banner is enabled in sshd_config
> - fixed typo in README
>
> The list of changes in Jira:
> https://issues.apache.org/jira/issues/?jql=fixVersion%20%3D%
> 202.2.2-Ducc%20AND%20project%20%3D%20UIMA
>
> The source and binary zip/tars are staged to
> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.2.2/RC3
>
> The Maven artifacts are here:
> https://repository.apache.org/content/repositories/orgapacheuima-1189
>
> The SVN tags are here:
> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.2.2/
>
> See http://uima.apache.org/testing-builds.html for suggestions on how to
> test release candidates.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ] 0   Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks
>
> Jerry Cwiklik
> Apache UIMA
>

Re: [VOTE] UIMA-DUCC 2.2.2 RC4

[Lou DeGenaro]

> Downloaded uima-ducc-2.2.2-bin.tar.gz

: OK

> Installed and started Apache DUCC demo http://uima-ducc-demo.apache.o
rg:42133/jobs.jsp :
   - Jobs, Services, Reservations running : OK
   - WS
  - Jobs + Details: OK
  - Reservations + Details : OK
  - Services + Details : OK
  - System Administration, Broker, Classes, Daemons, Machines : OK
  * System DuccBook : OK
  - System Viz : OK

> Spot check LICENSE file : OK
   - note typo:
  (webserver>navigation

> Check README file : OK

> Spot check NOTICE file: OK
   - line 225 Apache Tomcat is used?

> Check ReleaseNotes.html : OK

> Spot checked DUCC Book: OK
   - 19.4-6 are "To be filled in."
   - system-events.log not documented?

Vote on RC4 = +1

Lou.

On Wed, Mar 14, 2018 at 3:01 PM, Jerry Cwiklik  wrote:

> Hi,
> the UIMA-DUCC 2.2.2 release candidate 4 is ready for voting.
>
> This version contains the following fixes since RC3:
>
> - Fixed Agent to support long user names (up to 32 chars) when collecting
> node processes
>
> The list of changes in Jira:
> https://issues.apache.org/jira/issues/?jql=fixVersion%20%3D%
> 202.2.2-Ducc%20AND%20project%20%3D%20UIMA
>
> The source and binary zip/tars are staged to
> https://dist.apache.org/repos/dist/dev/uima/uima-ducc/2.2.2/RC4
>
> The Maven artifacts are here:
> https://repository.apache.org/content/repositories/orgapacheuima-1190
>
> The SVN tags are here:
> http://svn.apache.org/repos/asf/uima/uima-ducc/tags/uima-ducc-2.2.2/
>
> See http://uima.apache.org/testing-builds.html for suggestions on how to
> test release candidates.
>
> Please vote on release:
>
> [ ] +1 OK to release
> [ ] 0   Don't care
> [ ] -1 Not OK to release, because ...
>
> Thanks
>
> Jerry Cwiklik
> Apache UIMA
>
> --
> Jerry Cwiklik
> Apache UIMA
>
>