Jenkins slow after a few days and throwing exceptions all the time

[Knuplesch , Jürgen]

Hello,

 

I recently updated to Jenkind 1.470.

 

We installed the new CVS-Client.

Sadly the update command is not working because of a bug in Jenkins, so we 
disabled this.

 

CVS became slow too, so I disabled rlog in my jobs. That helped CVS but not 
Jenkins.

 

We use a lot of CVS!

 

Jenkins absorbed all the 2GB I gave him.

 

After a few days Jenkins becomes extremely slow and throws this:

 

hudson.util.IOException2: Failed to parse D:\Programme\Jenkins\.\jobs\CI_DOPE_01

02_Compile\builds\2012-06-29_15-33-05\changelog.xml

   at hudson.scm.CVSChangeLogSet.parse(CVSChangeLogSet.java:115)

at hudson.scm.CVSChangeLogParser.parse(CVSChangeLogParser.java:43)

at hudson.scm.CVSChangeLogParser.parse(CVSChangeLogParser.java:38)

at hudson.model.AbstractBuild.calcChangeSet(AbstractBuild.java:832)

at hudson.model.AbstractBuild.getChangeSet(AbstractBuild.java:806)

at hudson.model.AbstractBuild.hasParticipant(AbstractBuild.java:358)

at hudson.model.AbstractProject.hasParticipant(AbstractProject.java:1413

)

at hudson.model.User.getProjects(User.java:409)

at hudson.scm.MailAddressResolverImpl.findMailAddressFor(MailAddressReso

lverImpl.java:21)

at hudson.tasks.MailAddressResolver.resolve(MailAddressResolver.java:100

)

at hudson.tasks.Mailer$UserProperty.getAddress(Mailer.java:530)

at sun.reflect.GeneratedMethodAccessor220.invoke(Unknown Source)

at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAcces

sorImpl.java:25)

at java.lang.reflect.Method.invoke(Method.java:597)

at org.kohsuke.stapler.export.MethodProperty.getValue(MethodProperty.jav

a:66)

at org.kohsuke.stapler.export.Property.writeTo(Property.java:113)

at org.kohsuke.stapler.export.Model.writeNestedObjectTo(Model.java:179)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:231)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:181)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:136)

at org.kohsuke.stapler.export.Property.writeTo(Property.java:113)

at org.kohsuke.stapler.export.Model.writeNestedObjectTo(Model.java:179)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:231)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:136)

at org.kohsuke.stapler.export.Property.writeTo(Property.java:113)

at org.kohsuke.stapler.export.Model.writeNestedObjectTo(Model.java:179)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:231)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:181)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:136)

at org.kohsuke.stapler.export.Property.writeTo(Property.java:113)

at org.kohsuke.stapler.export.Model.writeNestedObjectTo(Model.java:179)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:231)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:181)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:136)

at org.kohsuke.stapler.export.Property.writeTo(Property.java:113)

at org.kohsuke.stapler.export.Model.writeNestedObjectTo(Model.java:179)

at org.kohsuke.stapler.export.Model.writeNestedObjectTo(Model.java:176)

at org.kohsuke.stapler.export.Model.writeNestedObjectTo(Model.java:176)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:231)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:181)

at org.kohsuke.stapler.export.Property.writeValue(Property.java:136)

at org.kohsuke.stapler.export.Property.writeTo(Property.java:113)

at org.kohsuke.stapler.export.Model.writeNestedObjectTo(Model.java:179)

at org.kohsuke.stapler.export.Model.writeNestedObjectTo(Model.java:176)

at org.kohsuke.stapler.export.Model.writeNestedObjectTo(Model.java:176)

at org.kohsuke.stapler.export.Model.writeNestedObjectTo(Model.java:176)

at org.kohsuke.stapler.export.Model.writeTo(Model.java:152)

at org.kohsuke.stapler.ResponseImpl.serveExposedBean(ResponseImpl.java:2

04)

at hudson.model.Api.doXml(Api.java:89)

at sun.reflect.GeneratedMethodAccessor564.invoke(Unknown Source)

at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAcces

sorImpl.java:25)

at java.lang.reflect.Method.invoke(Method.java:597)

at org.kohsuke.stapler.Function$InstanceFunction.invoke(Function.java:28

8)

 

 

All developers start to grumble at me )-:...

 

Please help! Any ideas?

 

Jürgen

-- 

Jürgen Knupleschwww.icongmbh.de  

icon Systemhaus GmbHTel. +49 711 806098-275

Sophienstraße 40juergen.knuple...@icongmbh.de

D-70178 Stuttg

Re: Choose build slave according to params

[krishna chaitanya kurnala]

Hi Michele

You can use Labels to just do that. Just group a bunch of Slaves with a
common labels, like test/production etc and use the Label in "Restrct where
this project can be run" in your Jenkins Job.

Thanks,
Krishna Kurnala


On Tue, Jul 17, 2012 at 11:32 PM, Michele Sacchetti <
michele.sacche...@gmail.com> wrote:

> Hi all
>
> is there a way/plugin to be able to force build node depending on a build
> parameter?
>
> I have production/staging/test profiles and I'd like to have different
> slaves in different sub-network to deploy them but I do not want to create
> 3 different job for this.
>
> thanks in advance
>

Choose build slave according to params

[Michele Sacchetti]

Hi all

is there a way/plugin to be able to force build node depending on a build 
parameter?

I have production/staging/test profiles and I'd like to have different 
slaves in different sub-network to deploy them but I do not want to create 
3 different job for this.

thanks in advance

How do I ignore mDNS?

[xipmox]

Hi all,

something in our environment seems to be producing mDNS broadcasts that 
jmdns is not coping with.
Unfortunately I am unlikely to be able to convince the offenders to fix 
their broken configuration.
Is it possible to make jenkins ignore mDNS? I don't need (or want) it to 
advertise by that mechanism.
I'd like to preserve normal DNS resolution, however.

I tried starting with -Dnet.mdns.interface=127.0.0.1 but jenkins didn't 
seem too happy about that.

I am running 1.474 via winstone (debian packages, amd64 architecture).

Any suggestions gratefully received.

Re: Source code confidentiality on EC2

[Ben Caradoc-Davies]

On 18/07/12 13:45, Jan Goyvaerts wrote:

Personally I was hoping for something with encrypted file systems only
the customers can mount.


Any system that can read your source code to build it must also have 
access to the plaintext. An administrator of the virtualisation cluster 
just has to wait until the filesystem is mounted, copy the contents, and 
snapshot the VM. If they have the VM, they have the keys already.


In practice, you are not safe from your own disgruntled in-house 
sysadmin, so it is all a matter of risk management. In management think 
your source code is valuable enough to keep secret, they should pay for 
secure private compute capacity for building.


Jenkins can probably help you by keeping your limited but secure private 
compute capacity for compile-only jobs, and shift all your unit and 
integration tests to EC2 nodes as stripped binaries.


--
Ben Caradoc-Davies 
Software Engineer
CSIRO Earth Science and Resource Engineering
Australian Resources Research Centre

Re: All Matrix Jobs missing after upgrade to 1.473 !!

[xipmox]

On Tuesday, July 3, 2012 11:35:25 PM UTC+10, Chris Withers wrote:
>
>
> When is 1.475 going to be released? 
>
> Chris 
>
>
>
I see trunk has been pointed to 1.476 and I don't see 1.475 out (18 July) 
except as a RC (11 July).
Are there issues with 1.475 that mean it won't be released?

Re: Source code confidentiality on EC2

[Jan Goyvaerts]

It's not that we're in defence or anyhting of that kind. But these kind of
questions will be the first the managers will ask here.

Personally I was hoping for something with encrypted file systems only the
customers can mount.

On Wed, Jul 18, 2012 at 5:26 AM, Ben Caradoc-Davies <
ben.caradoc-dav...@csiro.au> wrote:

> On 17/07/12 17:12, Jan Goyvaerts wrote:
>
>> Are there safe solutions to these concerns on EC2 ? Or any other
>> provider for that matter.
>>
>
> No.
>
> "Don’t put anything in the cloud you wouldn’t want a competitor, your
> government, or another government to see."
> http://www.netop.com/**solutions/report-privacy-and-**
> confidentiality-in-cloud-**computing.htm
>
> "Companies should not underestimate the level of corporate espionage –
> often backed by governments and their intelligence agencies – now taking
> place across the world. That is the message of security specialists
> following the revelation by Jonathan Evans, the head of the UK's
> intelligence service, MI5, that one company suffered an estimated £800m
> loss as a result of the theft of its intellectual property."
> http://www.computing.co.uk/**ctg/news/2187123/corporate-**
> espionage-an-industrial-scale-**targeting-uk
>
> Less risky options are your own private cloud, or a cloud operated by a
> provider in your own country where you have both contractual agreements to
> protect confidentiality and the protection of the law. Any provider with
> any services physically located in the US fails this test because the
> Patriot Act allows US intelligence agencies warrantless access to your
> data, and forbids your provider from telling you about it.
>
> Kind regards,
>
> --
> Ben Caradoc-Davies 
> Software Engineer
> CSIRO Earth Science and Resource Engineering
> Australian Resources Research Centre
>
>
>

Re: DOUBT regarding people link on jenkins

[xipmox]

On Wednesday, July 18, 2012 9:07:40 AM UTC+10, Holger Levsen wrote:
>
> Hi, 
>
> On Mittwoch, 4. Juli 2012, Sami Tikka wrote: 
> > Usually it is desireable if everyone who commits to a software 
> repository 
> > also has access to the Jenkins that builds from the repository. 
>
> thats not true for every software you build, eg as a company you might 
> very 
> well build eg the linux kernel but you still dont want to give access to 
> your 
> jenkins instance to everyone ;-) 
>
> And thus those (outside) committers are not as interesting as internal 
> ones. 
>
>
I agree. It is actually counterproductive not to distinguish committers and 
authenticated users;
when I first saw those names in the people listing** in our jenkins 
instance I thought there had been a security issue of some kind.

**Which is not really explained anywhere that I could find; if there is a 
good explanation please point to it so I can make it more visible.

Re: Jenkins, Gerrit-trigger and git submodules

[mooyah]

I would suggest the following:

- One jenkins job for each submodule, allow the jobs for each module to run 
the unit tests for each module to ensure the gerrit change is good to go
- Another jenkins job for each supermodule. This jenkins job would poll for 
every commit in the supermodule and build them as they happen.

I see a big flaw with this strategy that each submodule gerrit change would 
need to be merged into the respective submodule before integration testing 
could occur with the supermodules. I'm not sure what other possible 
solutions could be. I will be looking into implementing something like this 
soon (in the next quarter), so will try to update you about my progress.

-Joel

On Friday, July 6, 2012 4:47:02 AM UTC-6, Karsten Dambekalns wrote:
>
> Hi Rens. 
>
> On 02.07.2012, at 20:31, Rens Admiraal  wrote: 
> > I'm searching for a while now for a solution to use continues 
> integration in a good way. I'm using a (base) repository which contains the 
> base structure for my application, and furthermore a lot of git submodules 
> pointing to repositories containing the packages used in my project. 
>
> It seems we are the only ones doing that… or not? Consider this mail a 
> "ping" for this thread, and a plea for feedback - even if it's only "we 
> need more details to provide hints on that one" :) 
>
> Regards, 
> Karsten

Re: Source code confidentiality on EC2

[Ben Caradoc-Davies]

On 17/07/12 17:12, Jan Goyvaerts wrote:

Are there safe solutions to these concerns on EC2 ? Or any other
provider for that matter.


No.

"Don’t put anything in the cloud you wouldn’t want a competitor, your 
government, or another government to see."

http://www.netop.com/solutions/report-privacy-and-confidentiality-in-cloud-computing.htm

"Companies should not underestimate the level of corporate espionage – 
often backed by governments and their intelligence agencies – now taking 
place across the world. That is the message of security specialists 
following the revelation by Jonathan Evans, the head of the UK's 
intelligence service, MI5, that one company suffered an estimated £800m 
loss as a result of the theft of its intellectual property."

http://www.computing.co.uk/ctg/news/2187123/corporate-espionage-an-industrial-scale-targeting-uk

Less risky options are your own private cloud, or a cloud operated by a 
provider in your own country where you have both contractual agreements 
to protect confidentiality and the protection of the law. Any provider 
with any services physically located in the US fails this test because 
the Patriot Act allows US intelligence agencies warrantless access to 
your data, and forbids your provider from telling you about it.


Kind regards,

--
Ben Caradoc-Davies 
Software Engineer
CSIRO Earth Science and Resource Engineering
Australian Resources Research Centre

Re: Need a way to pass parameter to slave builds in a matrix job

[Owen B. Mehegan]

I discovered the Jenkins Dynamic Parameter plugin 
(https://wiki.jenkins-ci.org/display/JENKINS/Jenkins+Dynamic+Parameter+Plug-in),
 
which will do what I want. I wrote a little Groovy code to generate the 
version number in our preferred format, so that gets pre-populated as a 
parameter to the build when it is run. Then it's available for all the 
slaves, and my build code does the right thing thereafter.

On Tuesday, July 17, 2012 2:17:34 PM UTC-7, Owen B. Mehegan wrote:
>
> Historically, we have used Jenkins to build and package about 35 projects, 
> all on Debian Lenny 32 bit. We are now in the process of migrating to 
> Ubuntu Precise 64 bit systems, and we have a need to build some of our 
> projects on both of these architectures during the transition phase. A 
> perfect use case for matrix jobs, right? Seems like it, but I've run hard 
> into a limitation of this type of build, and no combination of plugins or 
> other hacks has given me a solution.
>
> Our build jobs compile and run tests, and if those pass, they run a rake 
> task to increment the version number (using our internal format) and then 
> generate a package. The version number is timestamp based, so when we run a 
> matrix build on two different nodes, we end up with two slightly different 
> version numbers for e.g. the 32 and 64 bit packages. This in turn causes 
> problems for our deploy system, which wants to ingest one version number, 
> and then grab the appropriate package for the appropriate hosts.
>
> What I need is a way to increment the version number in one place, and 
> then distribute that to all the nodes that run the build steps of the 
> matrix job. I can even handle generating the version number independently 
> of an SCM checkout of our code - a shell script option would be fine. Or if 
> I could trigger an external job ONCE (not for every slave that the matrix 
> runs on), which would generate a version number and offer it as an artifact 
> for my build to ingest, that would work too. What I don't want to do is 
> create a top level job that has to be run to generate a number, then pass 
> that on to whichever of the 35 downstream jobs actually need it.
>
> I've looked at the Env Inject plugin, I've looked at touchstone builds, 
> and a few other things. Nothing seems to do what I want, and it's really 
> blocking progress on this migration for us. If anyone has suggestions for 
> what I could try, I'm all ears! Thanks :)
>

Re: DOUBT regarding people link on jenkins

[Holger Levsen]

Hi,

On Mittwoch, 4. Juli 2012, Sami Tikka wrote:
> Usually it is desireable if everyone who commits to a software repository
> also has access to the Jenkins that builds from the repository.

thats not true for every software you build, eg as a company you might very 
well build eg the linux kernel but you still dont want to give access to your 
jenkins instance to everyone ;-) 

And thus those (outside) committers are not as interesting as internal ones.


cheers,
Holger

Re: Jenkins as Windows Service; Mercurial.ini

[Jesse Glick]

On 07/17/2012 07:56 AM, Andreas Schilling wrote:

our jenkins master is set up as windows service using the windows local system 
account.
where are we supposed to put the mercurial.ini with the credentials for the 
https access?


hgrc(5) [1] gives a number of locations where Mercurial looks, or can be made 
to look, for config files.

You can also pass '--config section.key=value' as part of the Mercurial command, so defaults could be configured in the Mercurial installation your jobs use (see 
$jenkins/configure). Of course this is less secure than storing credentials in a local file location; for example, the values will probably appear in build logs.


By the way if you are trying to use client certificates, rather than just a 
username/password combo, and get it to work, please update JENKINS-7351 [2] 
accordingly.

[1] http://www.selenic.com/mercurial/hgrc.5.html
[2] https://issues.jenkins-ci.org/browse/JENKINS-7351

Re: Trigger action just before a build is aborted.

[Jesse Glick]

On 07/17/2012 05:02 AM, Michael wrote:

What I would like to happen is that a pre-abort action be triggered, for
example run jstack on all the java processes.


This would be a nice enhancement for the Build Timeout plugin. For example, using [1] it is possible (on JDK 6+) to obtain rich diagnostics on Java processes being run as 
part of the current build (hudson.util.ProcessTree), assuming the processes are still live enough to respond to this protocol (same as jstack) and not hung in JVM native 
code.


[1] 
http://docs.oracle.com/javase/6/docs/jdk/api/attach/spec/com/sun/tools/attach/VirtualMachine.html#loadAgent(java.lang.String,
 java.lang.String)

How to set hudson.scm.SubversionSCM.pollFromMaster ?

[orman]

https://wiki.jenkins-ci.org/display/JENKINS/Subversion+Plugin
https://wiki.jenkins-ci.org/display/JENKINS/Subversion+Plugin  says one can
Perform Polling from the Master if the
hudson.scm.SubversionSCM.pollFromMaster system property is set to true (for 
http://issues.jenkins-ci.org/browse/JENKINS-5413
http://issues.jenkins-ci.org/browse/JENKINS-5413 ) 

We put JENKINS_ARGS="-Dhudson.scm.SubversionSCM.pollFromMaster=true" at the
bottom of /etc/sysconfig/jenkins and we see the property in the cmd line
below. 

jenkins   1338 1 14 16:15 ?00:05:44
/usr/lib/jvm/java-1.6.0/bin/java -Dcom.sun.akuma.Daemon=daemonized
-Djava.awt.headless=true -Djavax.net.ssl.keyStoreType=jks
-Djavax.net.ssl.keyStore=/usr/java/default/jre/lib/security/cacerts
-Djavax.net.ssl.keyStorePassword=changeit -XX:MaxPermSize=1024m
-DJENKINS_HOME=/var/lib/jenkins -jar /usr/lib/jenkins/jenkins.war
--logfile=/var/log/jenkins/jenkins.log --webroot=/var/cache/jenkins/war
--daemon --httpPort=8082 --ajp13Port=8009 --debug=5 --handlerCountMax=100
--handlerCountMaxIdle=20 -Dhudson.scm.SubversionSCM.pollFromMaster=true 

Can't really tell if it's working. Is this correct ? 


--
View this message in context: 
http://jenkins.361315.n4.nabble.com/How-to-set-hudson-scm-SubversionSCM-pollFromMaster-tp4635040.html
Sent from the Jenkins users mailing list archive at Nabble.com.

Need a way to pass parameter to slave builds in a matrix job

[Owen B. Mehegan]

Historically, we have used Jenkins to build and package about 35 projects, 
all on Debian Lenny 32 bit. We are now in the process of migrating to 
Ubuntu Precise 64 bit systems, and we have a need to build some of our 
projects on both of these architectures during the transition phase. A 
perfect use case for matrix jobs, right? Seems like it, but I've run hard 
into a limitation of this type of build, and no combination of plugins or 
other hacks has given me a solution.

Our build jobs compile and run tests, and if those pass, they run a rake 
task to increment the version number (using our internal format) and then 
generate a package. The version number is timestamp based, so when we run a 
matrix build on two different nodes, we end up with two slightly different 
version numbers for e.g. the 32 and 64 bit packages. This in turn causes 
problems for our deploy system, which wants to ingest one version number, 
and then grab the appropriate package for the appropriate hosts.

What I need is a way to increment the version number in one place, and then 
distribute that to all the nodes that run the build steps of the matrix 
job. I can even handle generating the version number independently of an 
SCM checkout of our code - a shell script option would be fine. Or if I 
could trigger an external job ONCE (not for every slave that the matrix 
runs on), which would generate a version number and offer it as an artifact 
for my build to ingest, that would work too. What I don't want to do is 
create a top level job that has to be run to generate a number, then pass 
that on to whichever of the 35 downstream jobs actually need it.

I've looked at the Env Inject plugin, I've looked at touchstone builds, and 
a few other things. Nothing seems to do what I want, and it's really 
blocking progress on this migration for us. If anyone has suggestions for 
what I could try, I'm all ears! Thanks :)

Re: configure main view (Maven columns)

[Max Spring]

Duh, the Maven Info plugin was creating the additional columns.
https://wiki.jenkins-ci.org/display/JENKINS/Maven+Info+Plugin

-Mxa


On 07/16/2012 10:48 AM, Max Spring wrote:

All of a sudden I've got Maven-related columns in my main view:
Maven Dependencies, Maven Dependencies (within groupId), Maven Modules (Full), 
Maven Modules (Short), Maven Modules (Version).

In any custom view I can configure these columns.
How can I do the same for the main view?

Thanks!
-Max

Re: Highly secured Jenkins getting Java connection refused exceptions

[johnaparker]

Hi... Can anyone offer insight on this?

The bottom line is this:

While attempting to use the manage plugins function connections are 
being interfered with by something in our host based firewall rules or our 
hardware load balancer.

If I could get the default port usage during this type of traffic I can get 
our rules adjusted.

Please help.

John

>

Re: Conditinal build setup plugin not showing token for multiple

[Varghese Renny]

One more point to add, when i click ? symbol, it is not showing details,
instead going to intial part of page..


Any help would be thankful.



Thanks in Advance,
varghese

Re: Trigger action just before a build is aborted.

[krishna chaitanya kurnala]

Hi Micheal

Interesting to have something like that, I am currently using Build TimeOut
Plugin to would Abort/Fail the Build, it wont be able to Trigger any other
Action though. Please let me know as well, if you find something.

Thanks,
Krishna Kurnala


On Tue, Jul 17, 2012 at 2:02 AM, Michael wrote:

> *Use Case*
>
> A java process is causing the build to hang and as a result the build
> times out and the build is aborted.
>
> What I would like to happen is that a pre-abort action be triggered, for
> example run jstack on all the java processes. I understand that
> there could be a problem of this action itself hanging causing the
> build to hang and never terminate.
>
> Done a quick search on the news groups and could not find anything
> that fits the bill.
>
> Anyone any ideas?
>

Conditinal build setup plugin not showing token for multiple

[Varghese Renny]

Hi,
 i have installed conditional build setup plugin, but it doesn't show token 
for conditional multiple. For single it is showing the token and build 
setup..
Is something missing for me? Any more configuration i have to do for 
condtional multiple option to activate..
Please give me a solution

Thanks in Advance
varghese

Re: TestNG plugin - no results displayed

[krishna chaitanya kurnala]

Hi Nigel

I guess, I am facing the same issue, can you please let me know, how have
you fixed the un-synchronized clocks?

Thanks,
Krishna Chaitanya


On Tue, Jul 17, 2012 at 3:52 AM, Nig wrote:

> It was indeed the clocks out of sync between the Jenkins master and the
> file server.
>
> Many thanks Sami !
>
>
> On Tuesday, 17 July 2012 09:01:14 UTC+1, sti wrote:
>
> If you look at the message from Jenkins, it seems Jenkins is not happy
>> with the last-modified attribute of the xml file. If Jenkins sees the xml
>> file is older than the build, it thinks the file might be a left-over from
>> an earlier build. This is a common mistake if people assume build always
>> start with an empty workspace. It doesn't, unless you specifically
>> configure it to be so.
>>
>> Another possibility is the xml file has actually been checked into the
>> version control by someone.
>>
>> Yet another possibility is clocks aren't synchronized in your Jenkins
>> server, your file server and your version control server. Unsynchronized
>> clocks are an easy way to generate strange errors that are loads of fun to
>> debug!
>>
>> -- Sami
>>
>> Nig kirjoitti 16.7.2012 kello 12.27:
>>
>> > Hi,
>> >
>> > I've installed  testng-6.5.2 and version 0.32 of the pulgin but no
>> results are displayed.
>> >
>> > The following is logged in the Console Output:
>> > TestNG Reports Processing: START
>> > Looking for TestNG results report in workspace using pattern:
>> **/testng-results.xml
>> > testng-results.xml was last modified before this build started.
>> Ignoring it.
>> > Saving reports...
>> > Found matching files but did not find any TestNG results.
>> >
>> > The testng-results.xml file is attached.
>> >
>> > Am I doing something wrong ?
>> >
>> > Thanks !
>> >
>> > 
>>
>>

Re: Putting condition for build of same job

[Varghese Renny]

For build cause where i needed to set reason like poll scm..i am not able
to see the configuration for build cause?
Where i can configure build cause?






Thanks in Advance
varghese

Re: Checkout the code from two repository without cleaning local directory

[Varghese Renny]

How can i write using shell script like checking whether any updates on 
repository every time..(polling repository everytime) do update if any 
changes..replacing existing poll scm..?


Thanks in advance

Re: Disabling hover/mouseover menus

[Stephen Connolly]

Nope core feature

On 16 July 2012 21:48, Jan Seidel  wrote:

> *uhm* does it not require a plugin to have that feature?
>
>
> Am Mittwoch, 11. Juli 2012 16:49:19 UTC+2 schrieb rayn:
>
>> Is there any way to turn these off?  Quite annoying.
>
>

Re: Checkout the code from two repository without cleaning local directory

[Jan Seidel]

Hi Varghese,

only solution I know of is to do an update yourself from shell/command line.
It's no big deal. Not as nice as running the plugins but sometimes are you 
facing special situations which requires this kind of action.

Take care
Jan

Am Dienstag, 17. Juli 2012 14:00:40 UTC+2 schrieb Varghese Renny:
>
> Hi,
>I have one workspace which i am checking out from repository A. This 
> workspace contain modules folder which contains some modules.
>   I needed to checkout from another repository B to modules folder where 
> the commit is going.
> But when checkout is happening it is cleaning modules folder completely. 
> it doesn't perform any updation but replacing existing modules.
>
> My question is whether any possible way to perform svn updation only 
> without cleanout the workspace module folder.
>
>
> Thanks in Advance
> varghese
>
>

Checkout the code from two repository without cleaning local directory

[Varghese Renny]

Hi,
   I have one workspace which i am checking out from repository A. This 
workspace contain modules folder which contains some modules.
  I needed to checkout from another repository B to modules folder where 
the commit is going.
But when checkout is happening it is cleaning modules folder completely. it 
doesn't perform any updation but replacing existing modules.

My question is whether any possible way to perform svn updation only 
without cleanout the workspace module folder.


Thanks in Advance
varghese

Jenkins as Windows Service; Mercurial.ini

[Andreas Schilling]

hi everyone!

we're currrently setting up our new mercurial solution and are facing a 
problem getting the builds to run again.
our jenkins master is set up as windows service using the windows local 
system account.
where are we supposed to put the mercurial.ini with the credentials for 
the https access?
on the slave it works fine, but the slave service is run as another user, 
not the local system user. thus we simply put the .ini file into the user 
home as you can read in many tutorials and it worked.
we're not keen on simply changing the user our master runs on, because 
we're not sure what else it might affect.
has anyone ever straightened that out?
Btw, mercurial plugin version is 1.34. 

kind regards and thanks for any help in advance,
 
Andreas Schilling
CAE Processes & Data Management
---
Dipl. Inf. Andreas Schilling
Senior Software Architect

TWT GmbH
Science & Innovation
Bernhäuser Str. 40 - 42
73765 Neuhausen
 
Tel: +49 - 7158 - 17 15 - 673
E-Mail: andreas.schill...@twt-gmbh.de

www.twt-gmbh.de

Geschäftsführung: Dimitrios Vartziotis, Joachim Laicher (stv.)
Registergericht: Amtsgericht Stuttgart, HRB Nr. 212778
Umsatzsteuer: ID-Nr.: DE147841145

Re: TestNG plugin - no results displayed

[Nig]

It was indeed the clocks out of sync between the Jenkins master and the 
file server.

Many thanks Sami !

On Tuesday, 17 July 2012 09:01:14 UTC+1, sti wrote:

> If you look at the message from Jenkins, it seems Jenkins is not happy 
> with the last-modified attribute of the xml file. If Jenkins sees the xml 
> file is older than the build, it thinks the file might be a left-over from 
> an earlier build. This is a common mistake if people assume build always 
> start with an empty workspace. It doesn't, unless you specifically 
> configure it to be so. 
>
> Another possibility is the xml file has actually been checked into the 
> version control by someone. 
>
> Yet another possibility is clocks aren't synchronized in your Jenkins 
> server, your file server and your version control server. Unsynchronized 
> clocks are an easy way to generate strange errors that are loads of fun to 
> debug! 
>
> -- Sami 
>
> Nig kirjoitti 16.7.2012 kello 12.27: 
>
> > Hi, 
> > 
> > I've installed  testng-6.5.2 and version 0.32 of the pulgin but no 
> results are displayed. 
> >   
> > The following is logged in the Console Output: 
> > TestNG Reports Processing: START 
> > Looking for TestNG results report in workspace using pattern: 
> **/testng-results.xml 
> > testng-results.xml was last modified before this build started. Ignoring 
> it. 
> > Saving reports... 
> > Found matching files but did not find any TestNG results. 
> >   
> > The testng-results.xml file is attached. 
> >   
> > Am I doing something wrong ? 
> > 
> > Thanks ! 
> > 
> >  
>
>

Re: Custom login screen message

[Sorin Sbarnea]

The System Message does not appear to the login page and if you allow
access to jenkins only to logged users, they will see only the login
page.

Re: Deploying application cause error in tomcat

[Varghese Renny]

  Thanks for reply..

i will add the ant task i have used for this


 
   
 
   

it is deploying the my project, then server will startup after that it will 
show due to log4j error occur while converting date, it cannot load the 
class reference..Listener start fail also..

Any help regarding this would be thankful..

Thanks in Advance
varghese
 

Source code confidentiality on EC2

[Jan Goyvaerts]

We're investigating the possibilities of cloud-based building with Jenkins
as our local hardware gets overwhelmed. Among which is Amazon's EC2. I am
asking it here because maybe somebody is already doing this. The aim would
be to move the system management to the provider while keeping the
possibility of scaling anytime to larger configurations.

We're - unfortunately - not an open source company. Hence the potential
confidentiality issues:

   1. Making the source repository reachable through the internet.
   2. The source code transits to an uncontrolled external site.

Are there safe solutions to these concerns on EC2 ? Or any other provider
for that matter...

TIA !

Jan

Trigger action just before a build is aborted.

[Michael]

*Use Case*

A java process is causing the build to hang and as a result the build 
times out and the build is aborted.

What I would like to happen is that a pre-abort action be triggered, for 
example run jstack on all the java processes. I understand that 
there could be a problem of this action itself hanging causing the 
build to hang and never terminate.  

Done a quick search on the news groups and could not find anything
that fits the bill.

Anyone any ideas? 

Editing Sectioned View broken with Status Code: 500

[Dirk Kuypers]

Hi,

I just wanted to add another list-view section in a sectioned view
existing for some time now. I am using Jenkins 1.474, all plugins
latest. As the sectioned view plugin was updated last year I suspect
that there broke something in Jenkins core?! Can anyone confirm this,
then I would file a JIRA ticket.

BR
Dirk

PS: This is the stack trace I get when trying to save:

Status Code: 500

Exception: net.sf.json.JSONException: JSONObject["useincluderegex"] is
not a JSONObject.
Stacktrace:
javax.servlet.ServletException: net.sf.json.JSONException:
JSONObject["useincluderegex"] is not a JSONObject.
at org.kohsuke.stapler.Stapler.tryInvoke(Stapler.java:616)
at org.kohsuke.stapler.Stapler.invoke(Stapler.java:659)
at org.kohsuke.stapler.MetaClass$6.doDispatch(MetaClass.java:241)
at 
org.kohsuke.stapler.NameBasedDispatcher.dispatch(NameBasedDispatcher.java:53)
at org.kohsuke.stapler.Stapler.tryInvoke(Stapler.java:574)
at org.kohsuke.stapler.Stapler.invoke(Stapler.java:659)
at org.kohsuke.stapler.MetaClass$12.dispatch(MetaClass.java:384)
at org.kohsuke.stapler.Stapler.tryInvoke(Stapler.java:574)
at org.kohsuke.stapler.Stapler.invoke(Stapler.java:659)
at org.kohsuke.stapler.MetaClass$6.doDispatch(MetaClass.java:241)
at 
org.kohsuke.stapler.NameBasedDispatcher.dispatch(NameBasedDispatcher.java:53)
at org.kohsuke.stapler.Stapler.tryInvoke(Stapler.java:574)
at org.kohsuke.stapler.Stapler.invoke(Stapler.java:659)
at org.kohsuke.stapler.Stapler.invoke(Stapler.java:488)
at org.kohsuke.stapler.Stapler.service(Stapler.java:162)
at javax.servlet.http.HttpServlet.service(HttpServlet.java:45)
at winstone.ServletConfiguration.execute(ServletConfiguration.java:248)
at winstone.RequestDispatcher.forward(RequestDispatcher.java:333)
at winstone.RequestDispatcher.doFilter(RequestDispatcher.java:376)
at 
hudson.util.PluginServletFilter$1.doFilter(PluginServletFilter.java:95)
at 
net.bull.javamelody.MonitoringFilter.doFilter(MonitoringFilter.java:202)
at 
net.bull.javamelody.MonitoringFilter.doFilter(MonitoringFilter.java:175)
at 
net.bull.javamelody.PluginMonitoringFilter.doFilter(PluginMonitoringFilter.java:86)
at 
org.jvnet.hudson.plugins.monitoring.HudsonMonitoringFilter.doFilter(HudsonMonitoringFilter.java:84)
at 
hudson.util.PluginServletFilter$1.doFilter(PluginServletFilter.java:98)
at 
hudson.plugins.greenballs.GreenBallFilter.doFilter(GreenBallFilter.java:58)
at 
hudson.util.PluginServletFilter$1.doFilter(PluginServletFilter.java:98)
at hudson.util.PluginServletFilter.doFilter(PluginServletFilter.java:87)
at winstone.FilterConfiguration.execute(FilterConfiguration.java:194)
at winstone.RequestDispatcher.doFilter(RequestDispatcher.java:366)
at hudson.security.csrf.CrumbFilter.doFilter(CrumbFilter.java:47)
at winstone.FilterConfiguration.execute(FilterConfiguration.java:194)
at winstone.RequestDispatcher.doFilter(RequestDispatcher.java:366)
at 
hudson.security.ChainedServletFilter$1.doFilter(ChainedServletFilter.java:84)
at 
hudson.security.UnwrapSecurityExceptionFilter.doFilter(UnwrapSecurityExceptionFilter.java:51)
at 
hudson.security.ChainedServletFilter$1.doFilter(ChainedServletFilter.java:87)
at 
org.acegisecurity.ui.ExceptionTranslationFilter.doFilter(ExceptionTranslationFilter.java:166)
at 
hudson.security.ChainedServletFilter$1.doFilter(ChainedServletFilter.java:87)
at 
org.acegisecurity.providers.anonymous.AnonymousProcessingFilter.doFilter(AnonymousProcessingFilter.java:125)
at 
hudson.security.ChainedServletFilter$1.doFilter(ChainedServletFilter.java:87)
at 
org.acegisecurity.ui.rememberme.RememberMeProcessingFilter.doFilter(RememberMeProcessingFilter.java:142)
at 
hudson.security.ChainedServletFilter$1.doFilter(ChainedServletFilter.java:87)
at 
org.acegisecurity.ui.AbstractProcessingFilter.doFilter(AbstractProcessingFilter.java:271)
at 
hudson.security.ChainedServletFilter$1.doFilter(ChainedServletFilter.java:87)
at 
org.acegisecurity.ui.basicauth.BasicProcessingFilter.doFilter(BasicProcessingFilter.java:173)
at 
hudson.security.ChainedServletFilter$1.doFilter(ChainedServletFilter.java:87)
at jenkins.security.ApiTokenFilter.doFilter(ApiTokenFilter.java:63)
at 
hudson.security.ChainedServletFilter$1.doFilter(ChainedServletFilter.java:87)
at 
org.acegisecurity.context.HttpSessionContextIntegrationFilter.doFilter(HttpSessionContextIntegrationFilter.java:249)
at 
hudson.security.HttpSessionContextIntegrationFilter2.doFilter(HttpSessionContextIntegrationFilter2.java:66)
at 
hudson.security.ChainedServletFilter$1.doFilter(ChainedServletFilter.java:87)
at 
hudson.security.ChainedServletFilter.doFilter(ChainedSer

Re: Sherry's HotList

[Andrew Melo]

This is a list for people wanting support for a specific product. Barring that, 
the targeted positions don't really line up with what (perhaps I'm wrong) 
people do.


Andrew Melo
Sent from my secret fortress.

On Jul 17, 2012, at 2:12, Varghese Renny  wrote:

> Not getting what you are trying to say..
> 
> Is it posting requirments or something related to jenkins?

Re: TestNG plugin - no results displayed

[Sami Tikka]

If you look at the message from Jenkins, it seems Jenkins is not happy with the 
last-modified attribute of the xml file. If Jenkins sees the xml file is older 
than the build, it thinks the file might be a left-over from an earlier build. 
This is a common mistake if people assume build always start with an empty 
workspace. It doesn't, unless you specifically configure it to be so.

Another possibility is the xml file has actually been checked into the version 
control by someone.

Yet another possibility is clocks aren't synchronized in your Jenkins server, 
your file server and your version control server. Unsynchronized clocks are an 
easy way to generate strange errors that are loads of fun to debug!

-- Sami

Nig kirjoitti 16.7.2012 kello 12.27:

> Hi,
> 
> I've installed  testng-6.5.2 and version 0.32 of the pulgin but no results 
> are displayed.
>  
> The following is logged in the Console Output:
> TestNG Reports Processing: START
> Looking for TestNG results report in workspace using pattern: 
> **/testng-results.xml
> testng-results.xml was last modified before this build started. Ignoring it.
> Saving reports...
> Found matching files but did not find any TestNG results.
>  
> The testng-results.xml file is attached.
>  
> Am I doing something wrong ?
> 
> Thanks !
> 
> 

Re: Custom login screen message

[Sami Tikka]

I'm not sure if it is possible to customize the actual login screen. However, 
aren't your users typically landing on the Jenkins top page first? Could you 
click the "Add description" link on the top page and write the custom message. 
It would appear at the top of the Jenkins top page. 

This is exactly what I did when I installed the ActiveDirectory plugin and I 
needed to instruct my users to start using their Windows login and password to 
log into Jenkins. It seemed to work reasonably well. No-one ever asked me how 
to log into Jenkins.

Additionally, these days in Jenkins global configuration there's a place for 
"System message", which will appear at the top of every Jenkins page. You could 
also use that.

-- Sami

Sorin Ionuț Sbârnea kirjoitti 16.7.2012 kello 17.06:

> Is there a solution for this? /thanks
> 
> On Friday, July 15, 2011 1:53:58 PM UTC+1, Martin B. wrote:
> Hi all!
> I have disabled anonymous access to our Jenkins instance.
> Is it somehow possible to display a custom message on the login screen 
> when not logged in?
> 
> 
> 
> 
> 
> 

Re: Is it called "Project" or is it called "Job"?

[Sami Tikka]

Currently it seems Jenkins UI uses "project" and "job" as synonyms. A single 
run of a "job" or "project" is usually named a "build". It would be nice to get 
some consistency on that.

By the way, if someone feels deeply enough about this, go to 
https://github.com/jenkinsci/jenkins, fork it, fix it and submit a pull request.

-- Sami

Re: Deploying application cause error in tomcat

[Sami Tikka]

Something is probably different in the environment when Tomcat is started by 
Jenkins.

I would start by comparing the environment variables in your shell where it 
works and environment variables in the shell used by ant.

You might want to execute the command "env" to see all the environment 
variables and their values.

You could add a build step into your job that executes "env" alone. Even better 
would be to use ant to run "env" command, just in case ant affects the 
environment variables in some way.

-- Sami

Varghese Renny kirjoitti 17.7.2012 kello 8.07:

> 
> While i trying to execute the tomcat startup from jenkins i am getting an 
> error like 
> log4j: Error occured while  converting date.
> i started tomcat using ant task.
> 
> When i manually do startup, it is not showing any problem application is 
> getting up.
> Version of tomcat i am using is tomcat  6.0.30
> 
> 
> Please help me , i am deploying it in my slave system where workspace exists.
> 
> 
> 
> 
> Thanks in Advance
> varghese 

Re: Sherry's HotList

[Varghese Renny]

Not getting what you are trying to say..
>
> Is it posting requirments or something related to jenkins?