Hi Sajjad
This sounds like it could be this issue:
https://jira.duraspace.org/browse/FCREPO-998
This is fixed for the next release - in the mean time could you check your
logging level - if this is set to DEBUG then you will see this error.
Steve
> -Original Message-
> From: Sajjad Anw
line
6. Purge Object
7. Run same query:
http://www.w3.org/2001/sw/DataAccess/rf1/result";>
test with a
newline
Cheers,
Jonathan
On Mon, Jan 30, 2012 at 12:47 PM, Stephen Bayliss
wrote:
Hi Jonathan
Hi Jonathan
It would be interesting to see what happens if the object is purged - do all
(both) of the label triples disappear? And if you then re-ingest it, do you
then have the correct number of labels?
It does look like there may have been a failure to remove an old version.
Which version
I've raised https://jira.duraspace.org/browse/FCREPO-1057 based on the
thread so far, any comments please add to the issue.
Thanks
Steve
> -Original Message-
> From: Stephen Bayliss [mailto:stephen.bayl...@acuityunlimited.net]
> Sent: 20 December 2011 07:55
> To:
Hi Asger
This looks great, thanks - would you like me to go ahead and merge this into
master?
Regards
Steve
-Original Message-
From: Stephen Bayliss [mailto:stephen.bayl...@acuityunlimited.net]
Sent: 19 December 2011 15:40
To: 'Support and info exchange list for Fedora
Hi Leandro
Take a look in your $FEDORA_HOME/install/install.properties file - that will
contain all the options you selected when installing.
Steve
-Original Message-
From: Leandro Castro [mailto:castroleand...@gmail.com]
Sent: 06 January 2012 21:42
To: Support and info exchange list f
Hi João
Looking at
https://github.com/fcrepo/fcrepo/blob/v3.4.2/fcrepo-security/fcrepo-security
-pdp/src/pdp_home/conf/config-attribute-finder.xml the config is there in
3.4.2 to reveal that property using the Fedora RI property URI for it, ie:
info:fedora/fedora-system:def/model#ownerId - but you
Hi João
This would appear to be the case - the resouce owner ID attribute doesn't
appear to be available in FeSL directly.
I've reported this as a bug as this object property is available in the
legacy XACML engine (FCREPO-1043) and therefore should also be available in
FeSL. I've also reported
And to add it's also valid to have URIs as objects of the triples in
addition to literals.
I'd recommend that you create the properties in your own namespace rather
than using properties in one of the Fedora namespaces.
Steve
> -Original Message-
> From: aj...@virginia.edu [mailto:aj...@
Hi Richard
>From memory I don't believe it is possible to pass datastreams by value (and
in fact as the CMA service calls are HTTP GET operations, it wouldn't be
ideal to pass through contents as URL parameters).
You should be able to use XSLT to dereference the URL and extract the values
you n
lidator understand about how some format
> uris can be descendants of each other?
>
>
>
> This is just some thoughts on the issue. I do feel that the
> current design, where you can specify a number of mimetypes
> in the content model, and the object is valid if at least o
various allowed charsets, and one without charset, and you should be safe.
Regards
On 12/15/2011 01:03 PM, Stephen Bayliss wrote:
As far as I can tell, ECM validation of a datastream's MIMETYPE is strict -
the entire MIMETYPE property contents have to match that declared in the
content model.
nship target gives 404 if target does not exist
The 1041 branch should now use a separate error message for the case of
allValuesFrom to an non-exisiting object. Unit tests have verified this.
Please try it out.
Regards
On 12/15/2011 10:21 PM, Stephen Bayliss wrote:
Hi Asger
I've verified
11 04:11 PM, Stephen Bayliss wrote:
<https://jira.duraspace.org/browse/FCREPO-1041>
https://jira.duraspace.org/browse/FCREPO-1041 raised for this.
-Original Message-
From: Stephen Bayliss [mailto:stephen.bayl...@acuityunlimited.net]
Sent: 13 December 2011 14:58
To: 'Sup
As far as I can tell, ECM validation of a datastream's MIMETYPE is strict -
the entire MIMETYPE property contents have to match that declared in the
content model.
What about the case where one might want to specify the MIMETYPE of a
datastream in the CModel, but not the character set? If I sp
On 12/13/2011 04:11 PM, Stephen Bayliss wrote:
<https://jira.duraspace.org/browse/FCREPO-1041>
https://jira.duraspace.org/browse/FCREPO-1041 raised for this.
-Original Message-
From: Stephen Bayliss [mailto:stephen.bayl...@acuityunlimited.net]
Sent: 13 December 2011 14:58
To: &
<https://jira.duraspace.org/browse/FCREPO-1041>
https://jira.duraspace.org/browse/FCREPO-1041 raised for this.
-Original Message-
From: Stephen Bayliss [mailto:stephen.bayl...@acuityunlimited.net]
Sent: 13 December 2011 14:58
To: 'Support and info exchange list for F
just work!
Regards
Steve
-Original Message-
From: Asger Askov Blekinge [mailto:a...@statsbiblioteket.dk]
Sent: 13 December 2011 14:21
To: fedora-commons-users@lists.sourceforge.net
Subject: Re: [fcrepo-user] ECM - validating the "type" of an object
On 12/13/2011 01:51 PM, Stephen Bayl
empty page, or the xml? Using the http return codes might not be in
error.
Should validate returon 2xx when the validation fails?
Regards
On 12/13/2011 01:51 PM, Stephen Bayliss wrote:
In the ONTOLOGY datastream of a content model object, one can specify
- the cardinality of a relationship
- th
In the ONTOLOGY datastream of a content model object, one can specify
- the cardinality of a relationship
- the type (ie type from hasModel relationship) of the target
I am finding that if the target object does not exist, then the validate
method is reporting a 404 error. That's the same error
In ECM, the type of an object is defined effectively from its CModel. One
can validate the type of the target of a relationship, the typing being
specified by the hasModel relationship.
Is it possible to do any additional type validation; eg if I want to enforce
that all objects "belonging" to t
ATALINA_HOME/bin/shutdown.sh doesn't succeed; I have to use
> > kill -15 to remove the process.
> >
> >
> > Janna Wemekamp
> >
> >
> > On 15/09/2011 3:19 AM, Stephen Bayliss wrote:
> >> When starting Fedora with all logging set to DEBUG, I'
y requests from the target server to the
> source server would be allowed in, but I suppose the XACML could be
> tightened to just apply to certain objects or operations, if that were
> necessary.
>
> -- Scott
>
> On 12/02/11, Stephen Bayliss wrote:
> > > However,
o authenticate, and therefore cannot
> obtain the datastreams. Is that right?
>
> ---
> A. Soroka
> Online Library Environment
> the University of Virginia Library
>
>
>
>
> On Dec 2, 2011, at 8:24 AM, Stephen Bayliss wrote:
>
> > Hi Richard
> >
&
onment
the University of Virginia Library
On Dec 2, 2011, at 8:24 AM, Stephen Bayliss wrote:
> Hi Richard
>
> I think your non-developer eye may have spotted the issue here.
>
> The doCommit(...) method is where the object is being saved on the target
> server, and that'
ent
has credentials for both source and destination).
Richard
-Original Message-
From: Stephen Bayliss [mailto:stephen.bayl...@acuityunlimited.net]
Sent: 01 December 2011 6:15 PM
To: 'Support and info exchange list for Fedora users.'
Subject: Re: [fcrepo-user] Ingest Question
Hi Richard
This looks like the error/stack trace from the Java client (apologies if I
am mistaken) - if that is the case can you find the relevant section of the
receiving server's log file and see if it provides any more detail.
Thanks
Steve
-Original Message-
From: Richard Green [mailt
> -Original Message-
> From: Stephen Bayliss [mailto:stephen.bayl...@acuityunlimited.net]
> Sent: 09 November 2011 14:53
> To: 'Support and info exchange list for Fedora users.'
> Subject: Re: [fcrepo-user] fedora-users.xml being
> persistently cached whenserver re
ctly prompt me again for the password, and only accepted
> the new one. This was without me touching the SESSIONS.ser file.
>
> So it appears that if my browser already has a session
> established under which it has authenticated once, the old
> password still works through t
I'm using Fedora with the embedded Tomcat, FeSL AuthN and AuthZ.
I modified my fedora-users.xml (changed admin password), but after a server
restart it won't accept the new password and is still accepting the old one.
I removed $CATALINA_HOME/work/Catalina/localhost/fedora/SESSIONS.ser and now
Hi Alistair
Muradora is no longer being maintained, I believe the last release was a 1.4
beta which was the first release that worked with Fedora 3.x.
The authentication and authorisation components of Muradora have been
migrated into the core Fedora codebase (starting with Fedora 3.3) and are
kn
Hi Ben
As far as I'm aware the basic OAI-PMH provider doesn't provide set
information (though I haven't looked at the source code recently).
You might want to look at the PROAI-based OAI-PMH provider at
https://wiki.duraspace.org/display/FCSVCS/OAI+Provider+Service+1.2.1
This does provide set in
Hi Ben
You should see the startup scripts in the external Tomcat's bin directory.
Steve
> -Original Message-
> From: Benjamin Ryan [mailto:b.r...@leeds.ac.uk]
> Sent: 04 October 2011 14:28
> To: 'fedora-commons-users@lists.sourceforge.net'
> Subject: [fcrepo-user] Starting Fedora when u
y_mindmapping
which is the last one in the RELS-EXT relationships. Curious why only one
returned. Also what would be a general query to return all objects that
have:
isInCategory demo:category_mindmapping
thanks,
Alistair
--
mov eax,1
mov ebx,0
int 80
On 6 Sep 2011, at 10:59, St
When starting Fedora with all logging set to DEBUG, I'm getting the
following exception (resource index is enabled). It's fine if I set it to
WARN. Tomcat and Fedora logs below.
I wonder if anyone replicate this?
Steve
Tomcat log:
**
** FEDORA STARTUP ERROR **
Hi Tobias
You need to use the fully-qualified name; namespace-prefixed names won't
work correctly in the relationships methods.
So your predicate
fedora:hasDerivation
Should instead be
info:fedora/fedora-system:def/relations-external#hasDerivation
Steve
> -Original Message-
> From: T
Just a thought, but if the stack overflow is triggering this, have you tried
modifying the memory parameters for the JVM in which the client is running?
The set-env script seems to specify -Xms64m -Xmx96m
Possibly these values are no longer appropriate for the java admin client;
or increasing t
Hi Alistair
A couple of points that may help (and I think the examples on that page
won't work):
- The relationship to a content model object is not an rdf:type relationship
(in fact it is a property)
-- so I think instead $object
for the part of the where
clause dealing with content models
- T
Hi Shuo
Are you sure Fedora is not starting? The script starts up Tomcat, and will
terminate once Tomcat has started. After the script has terminated do you
see anything if you go to http://server:port/fedora/objects (eg
http://localhost:8080/fedora/objects) ?
Perhaps you could monitor the Fedo
Hi Ben
In your first rule, under actions, you have:
http://www.w3.org2001/XMLSchema#string";>
This appears to be missing a "/"; ie it should be
"http://www.w3.org/2001/XMLSchema#string";
Regards
Steve
> -Original Message-
> From: Benjamin Ryan [mailto:b.r...@leeds.ac.uk]
> Sent:
Hi Olivier
The source code is on github - https://github.com/fcrepo/mptstore - there is
a tag for that version.
Regards
Steve
-Original Message-
From: Olivier Meurice [mailto:olivier.meur...@iriscorporate.com]
Sent: 18 August 2011 09:05
To: 'fedora-commons-users@lists.sourceforge.net'
One thing I do notice is that you seem to be using "myfedora" rather than
the default "fedora" for the Fedora webapp - is it possible that the clients
that are being refused connections are trying to connect to
http://localhost:8080/fedora rather than http://localhost:8080/myfedora?
You say you we
Hi Alistair
Fedora's built-in search service ("Basic Search") only indexes DC - it's not
intended as a search service for a production repository. For this most
people use an external search service, such as SOLR - GSearch provides a
framework for using Fedora with SOLR.
An alternative is the Re
Out of curiosity, is this whitespace in the XML (in the XACML policy), or
whitespace in the attribute values returned by an attribute finder? (if
it's in the XML I appreciate the issue also; I've now got into the habit of
never having linebreaks in element content for instance; and if generating
w
Hi Dave
The technique is described in this thread:
http://fedora-commons.1317035.n2.nabble.com/fcrepo-user-Gsearch-and-fedora-d
atastreams-td5413479.html
It is possible you will hit a problem with authentication, for which there
is a patch which I believe isn't in 2.2; see
https://github.com/fc
Hi Alistair
I don't think there is an easy way to do what you want (there's no facility
in Fedora for modifying object identifiers - as they are intended as
persistent identifiers). I think you would need to update the FOXML for
each object to do what you want.
Regards
Steve
> -Original Mes
">
>
>
>rdf:about="info:fedora/demo:usercollection">
> true
> testowner
>
>
>
>
>
>
>
> Alistair
>
> --
> mov eax,1
> mo
Hi Alistair
Are you able to paste in the RDF/XML for RELS-EXT that that is failing?
Steve
> -Original Message-
> From: Alistair Young [mailto:alistair.yo...@uhi.ac.uk]
> Sent: 30 July 2011 10:25
> To: fedora-commons-users@lists.sourceforge.net
> Subject: [fcrepo-user] API-M second RELS-
I don't believe this is possible in ITQL - I'm less sure about SPARQL but I
can't see anything useful there.
Is it not possible for you to number your results outside of your query
results - eg in whatever is processing those results? For instance if your
results were formatted in XML you could
> > belongs to an organisation:
> > xxx
> >
> > So if the user belongs to orgid 42 then the newly created object is
> > has to have the line:
> > 42
> > in the RELS-EXT.
> >
> >
> > Now another use case.
> > A user with an
Hi Tomasz
At a high level I think what you are saying fits within XACML - though I'm
not sure how much will be present within the current FeSL implementation.
It is possible to base policies on the content that is being added - for
instance in the modifyDatastream method it is possible to restri
Hi Linton
There doesn't look to be anything wrong with the information you have
supplied.
Maybe you could also supply your Fedora installation properties file from
$FEDORA_HOME/install/iinstall.properties.
Were there any errors reported by the installation process?
Also, what do you see in
Hi Alistair
There's no distinction as such in terms of basic Fedora objects between data
objects and collections. The Fedora relationships ontology
(http://www.fedora.info/definitions/1/0/fedora-relsext-ontology.rdfs) and
the content models you implement are usually used to describe the
distincti
It's a supported configuration. Best to use the same version of Mulgara
that's currently packaged with Fedora. The configuration is described here:
https://wiki.duraspace.org/display/FCR30/Resource+Index#ResourceIndex-Remote
MulgaraDatastoreConfiguration
> -Original Message-
> From: Sco
Hello Tomasz
I am guessing you are using Fedora 3.4.2 from the log.
Resource index query-based attributes are a new feature in Fedora 3.5; which
would explain the error I think.
Regards
Steve
> -Original Message-
> From: Tomasz Cielecki [mailto:tom...@ostebaronen.dk]
> Sent: 13 July 20
I've replicated this, it does look like a bug.
In the REST API code, the ownerId query parameter for new, empty objects is
ignored and instead the value of the user making the request is used.
Instead it should only use the value of the user making the request if the
value is not supplied as a qu
" MustBePresent="false"
> DataType="http://www.w3.org/2001/XMLSchema#string"/>
>
>
>
>
>
> This results in the same thing as the first approach. No
> objects appear when I search for Notes...
>
> I have remembere
Hi Laney
What happens when you attempt to access Fedora using the REST API endpoints
(see https://wiki.duraspace.org/display/FCR30/REST+API)? In particular
could you try
https://wiki.duraspace.org/display/FCR30/REST+API#RESTAPI-getDatastream the
getDatastream method as this is a Management metho
POLICY datastream is not working
>
>
> 2011/6/15 Stephen Bayliss :
> > In the non-FeSL XACML implementation, POLICY datastreams in objects
> > implicitly apply to the object containing the datastream. In other
> > words in a POLICY datastream you can only use a
at 7:48 PM, Stephen Bayliss
wrote:
Hi Jason
That doesn't look right.
Do you have a FOXML file that you ingested to create this object?
Regards
Steve
-Original Message-
From: Jason Darwin [mailto:jcdar...@gmail.com]
Sent: 15 June 2011 11:38
To: Support and info exchange
crepo-user] A couple of questions: character encoding
andfiletype extensions
Hi Stephen,
Yes, had a look and I find
Cheers,
Jason
On Wed, Jun 15, 2011 at 9:25 PM, Stephen Bayliss
wrote:
Hi Jason
Sounds like there could be an encoding issue in the display.
To make sure it's in
ocreateFeSLpolicies.
>
>
> Hi Stephen,
>
> Where should I look if I want to take a shot at allowing
> access to other subject attributes in FeSL?
>
> I see there is a lot of related code in fcrepo-security, but
> not sure where to start.
>
> On Wed, J
however in the label fields in web admin display and in the web page display in
the REST interface (e.g.
http://localhost:8080/fedora/objects/test:1/datastreams) the text is appearing
as "Taku wh%u0101nau".
Jason
On Wed, Jun 15, 2011 at 7:46 PM, Stephen Bayliss
wrote:


Regarding the character encoding, FOXML should be using UTF-8 for these.
Where are you seeing the unicode codepoints (ie how are you accessing the
label) - does there seem to be any problem in the FOXML itself?
Steve
-Original Message-
From: Jason Darwin [mailto:jcdar...@gmail.co
In the non-FeSL XACML implementation, POLICY datastreams in objects
implicitly apply to the object containing the datastream. In other words in
a POLICY datastream you can only use a XACML policy for the containing
object.
These object-specific policies are only evaluated when accessing that
part
t;> xmlns:k2rdf="http://kemibrug.dk/k2/rdf#";
> >> xmlns:dc="http://purl.org/dc/elements/1.1/";
> >> xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/";>
> >>
> >> >> rdf:resource="info:fedora/or
Hi Swithun
I've added you.
Probably best to put the changes in a branch initially. You might want to
take a look at Scott's quick-start guide to using Git:
https://wiki.duraspace.org/display/FCREPO/Git+Quick+Start+Guide
Regards
Steve
> -Original Message-
> From: Swithun Crowe [mailto:
Hello Swithun
This sounds like some useful contributions.
A short while ago we moved Chris' EZService to github so that folks could
collaborate.
It's here - https://github.com/AcuityUnlimited/FedoraEZService
There's one modification over the original EZService, which is to allow
multiple CModel
Hi Tomasz
Basing policies directly on XML content (and restricting access to XML
content) is part of the XACML 2.0 spec as part of the Hierarchical Resource
Profile -
http://docs.oasis-open.org/xacml/2.0/access_control-xacml-2.0-hier-profile-s
pec-os.pdf
However this is not implemented in FeSL (i
ty of Virginia Library
>
>
>
>
> On May 25, 2011, at 4:56 PM, Scott Prater wrote:
>
> > This seems like a good use for the Fedora JMS message producer, too:
> > set up a listener for Fedora API-A and API-M events, then
> hand those
> > events of
Hi Damian
I've raised https://jira.duraspace.org/browse/FCREPO-949 for this.
Thanks for the patch.
Regards
Steve
> -Original Message-
> From: Damian Kaliszan [mailto:dam...@man.poznan.pl]
> Sent: 25 May 2011 14:29
> To: Support and info exchange list for Fedora users.
> Subject: [fcrep
Hi
As part of the PIRUS2 project [1] we are examining approaches to collecting
usage statistics from Fedora repositories.
We are interested in hearing of existing approaches to this topic. Do you
collect usage information? If so, how?
Some methods that I can think of are:
1) Fedora behind Apa
Hi Damian
Take a look at GetDissemination.methodParam(String key, String value)
- I think that's what you'll need to add query parameters to the method.
Regards
Steve
> -Original Message-
> From: Damian Kaliszan [mailto:dam...@man.poznan.pl]
> Sent: 24 May 2011 12:47
> To: fedora-comm
the logs, just an internal server error. I noticed with 3.4.2 this
can indicate the JVM ran out memory (not surprising if the export is
still being collected into a ByteArrayOutputStream, I guess).
Scott
On 05/18/2011 11:53 AM, Stephen Bayliss wrote:
Hi Scott
Thanks for that feedback.
l alternative paths with my
> particular project,
> so it isn't too big an issue to *me* I just have to do a little
> more coding in a middle-tier. Don't know about other folks, of course.
>
> -Scott
>
> On 05/18/2011 01:08 AM, Stephen Bayliss wrote:
> &
Looking at those lines of code it looks like in theory there would be a
problem there. Once this is confirmed we should probably add a test case to
the large datastreams test suite. And it is likely to cause a problem with
datastreams smaller than 2GB (2^31-1 as maximum array index) due to the
ar
> I also see one way in which the names of the installation properties
> might be misleading- "xacml.enabled=True" means that an authz
> framework is enabled, but it's the older xacml implementation rather
> than fesl. If I remember correctly, it's an error to have both of
> those properties set
Hi Swithun
Do you have the date/time that you grabbed the source code for the
3.5-SNAPSHOT build that you are using?.
DBXML isn't required if you are using a build from master. Instead a
file-backed memory-based policy index is used (these are the files in the
pdp/policy-db/ folder). DBXML an
Hi Benoit
The procedure should be similar to the upgrade procedure for a new version.
I would suggest you use the same version of Fedora on the new server. You
will need to ensure you use the same low-level storage configuration (ie
akubra-fs or legacy).
Set up and configure your new server
Hello Swithun
> One question is how I would go about triggering a reevaluation of an
> external policy. The script fedora-reload-policies.sh doesn't seem to do
> this. Even viewing the updated content of the external datastream in
> Fedora doesn't make a difference.
What you may be hitting her
Hi Swithun
Obviously the URL needs escaping in the METS document to make it valid XML.
But it sounds like this is passing through unescaped and therefore ingesting
managed content datastreams that contain query parameters then doesn't work
from what you are saying?
It would be useful to verify i
Hi Harry
Unfortunately the tutorial and resouces are out of date and missing chunks.
I would suggest that you first familiarise yourself with the documentation
at https://wiki.duraspace.org/display/FCR30/Fedora+Digital+Objects
Then you could try the following:
1) Use the Fedora admin client (ht
Hi Swithun
Unfortunately FeSL isn't that well documented at present, I'm hoping to add
some additional documentation as part of the next release.
In FeSL, all XACML policies are stored as datastreams in Fedora objects. So
you can either have separate Fedora objects for each policy (with FESLPOLI
utility available for the switch to Akubra?
Our current next step is to try putting Apache httpd in front of Tomcat and
see if its improved socket handling would help.
Rob
-Original Message-----
From: Stephen Bayliss [mailto:stephen.bayl...@acuityunlimited.net]
Sent: Thursday, April 28, 2
I just hit a similar (but unrelated) issue, the underlying cause of which
was a PrintWriter on the HTTP response not being flushed - I got partial
content, but it was intermittent.
So I wonder if this could be related?
Are you using the legacy low-level storage implementation or Akubra?
(And, if
85 matches
Mail list logo
