Hi,
On Wed, Jun 5, 2013 at 4:31 PM, Apache Jenkins Server
wrote:
> Check console output at https://builds.apache.org/job/Jackrabbit-trunk/2131/
> to view the results.
Looks like a Jenkins issue:
error=12, Not enough space
BR,
Jukka Zitting
Hi,
On Tue, May 7, 2013 at 3:54 PM, Jukka Zitting wrote:
> On Tue, May 7, 2013 at 1:59 PM, Olivier Lamy wrote:
>> IMHO/Maybe there are some zombies process running in the solaris
>> instances which eat the memory.
>
> Can someone with access to the slave check that?
Anyon
x27;s the bit I wasn't sure about; whether the error was
coming from the fork or the setup before that.
> IMHO/Maybe there are some zombies process running in the solaris
> instances which eat the memory.
Can someone with access to the slave check that?
BR,
Jukka Zitting
`/home/buildslave21/slave21/oak-trunk/build/analyze-hprof.pl':
Permission denied
rm: cannot remove
`/home/buildslave21/slave21/oak-trunk/build/oak-lucene/pom.xml':
Permission denied
[...]
BR,
Jukka Zitting
-- Forwarded message ------
From: Jukka Zitting
Date: M
o
work with.
For now I updated the build to use MAVEN_OPTS="-Xmx1g
-XX:MaxPermSize=300m" to cover for the latter case. Let's see what
happens.
BR,
Jukka Zitting
Hi,
On Mon, Mar 18, 2013 at 9:36 AM, wrote:
> The Buildbot has detected a new failure on builder oak-trunk while building
> ASF Buildbot.
> Full details are available at:
> http://ci.apache.org/builders/oak-trunk/builds/1846
Looks like some permission problem on osiris_ubuntu.
Hi,
[cc += builds@]
On Mon, Jan 21, 2013 at 9:16 AM, wrote:
> http://ci.apache.org/builders/tika-trunk/builds/1020
Looks like a buildbot issue: "cp: writing
`/home/buildmaster/master1/public_html/projects/tika/rat-output.xml':
No space left on device"
BR,
Jukka Zitting
al. People and projects who want
to use Apache snapshots (without building them locally) need to
explicitly add the repository.apache.org repository to their Maven
settings.
I haven't used the IRC notification setup, but perhaps someone else
can help with that.
BR,
Jukka Zitting
pository" option (with
"https://repository.apache.org/content/repositories/snapshots"; as the
URL and "apache.snapshots.https" as the ID).
BR,
Jukka Zitting
; Could someone with jenkins admin karma please help these
> folks out and disable their build?
Done. Normal jobadmin karma is enough for this.
BR,
Jukka Zitting
test3 without
> lib directory
BR,
Jukka Zitting
-- Forwarded message --
From: Apache Jenkins Server
Date: Fri, Jun 3, 2011 at 6:12 PM
Subject: Build failed in Jenkins: Tika-trunk #559
To: d...@tika.apache.org
See <https://builds.apache.org/hudson/job/Tika-trunk/559/changes
min for the projects I'm involved in, but as seen in some recent
cases I'm no longer up to date enough to track down and fix issues in
the Jenkins installation itself.
BR,
Jukka Zitting
Hi,
Something wrong with the Jenkins slave on solaris2, see below.
BR,
Jukka Zitting
Original Message
Subject: Build failed in Jenkins: Tika-trunk #513
Date: Mon, 11 Apr 2011 13:01:19 +0100
From: Apache Hudson Server
Reply-To: d...@tika.apache.org
To: d...@tika.apache.org
(ThreadPoolExecutor.java:603)
at java.lang.Thread.run(Thread.java:636)
I can't ssh there, so perhaps someone else can fix this?
BR,
Jukka Zitting
as there the commit account setup has already been done. AFAIUI the
site builds are fairly straightforward, so there should be no need for
all the extra features Jenkins has over Buildbot.
BR,
Jukka Zitting
d it puts on Jenkins. Is this OK?
Doesn't sound like a very good approach.
The Apache CMS setup running on Buildbot simply commits the generated
site to svn from where it's picked up by the public web servers. I
guess a similar setup could be done also with Maven site builds
running in Hudson.
BR,
Jukka Zitting
s is restored.
Good, thanks!
> Keep up man :)
Yeah, I've spent way too much time writing in the past few months.
Luckily I'm just about done already. :-)
BR,
Jukka Zitting
the hudson master and my direct login on minerva no longer
works.
BR,
Jukka Zitting
it some more, thanks
I was thinking of just adding something like this:
ProxyPass /jenkins http://internal-server-name.apache.org/jenkins/
BR,
Jukka Zitting
e URL is IMHO quite OK, as we support
multiple CI systems and the product name is the best way to tell them
apart. And it's not like Jenkins will be changing its name again
anytime soon.
BR,
Jukka Zitting
/load-statistics?type=hour
BR,
Jukka Zitting
Hi,
On Thu, Dec 2, 2010 at 10:50 PM, Jukka Zitting wrote:
> Also the Jackrabbit-trunk build failed with an out of space error. It
> looks like the culprit is the /var/tmp directory, that contains almost
> 1.7M files! I'm clearing up that directory...
The vast majority of these f
> Weird, there is plenty of space on all file systems on that slave.
Also the Jackrabbit-trunk build failed with an out of space error. It
looks like the culprit is the /var/tmp directory, that contains almost
1.7M files! I'm clearing up that directory...
BR,
Jukka Zitting
ration
option just allows you to add more things to be archived.
See http://issues.hudson-ci.org/browse/HUDSON-3922 for the relevant
Hudson issue.
BR,
Jukka Zitting
al/Hudson#What_should_I_do_with_the_build_and_test_tools_that_my_project_needs.3F
BR,
Jukka Zitting
lve
this issue without the need for a separate release repository.
BR,
Jukka Zitting
> like to be able to run the process all the way through to release
> without 'releasing' anything [...]
Isn't that the very purpose of the staging repository?
> [...] or even showing up in the staging repo.
Why? You can always simply drop those deployments that you don't want released.
BR,
Jukka Zitting
Hi,
On Sun, Oct 24, 2010 at 11:50 PM, Niklas Gustavsson
wrote:
> I finally bit the bullet and rearranged the views in Hudson into a
> nested structure.
Nice work, thanks!
BR,
Jukka Zitting
ool.
I changed the freebsd1 configuration to "tied jobs only" until we have
JDK 1.5 there, as I believe quite a few projects are using Java 5 as
their baseline build environment.
BR,
Jukka Zitting
or directory
PS. I tried to do something about this, but it looks like I still lack
sudo access to the hudson account on aegis. Can someone grant me the
missing karma? I have OPIE properly set up.
[1] https://hudson.apache.org/hudson/job/Jackrabbit-trunk/1300/console
BR,
Jukka Zitting
Hi,
Restarted.
If we can't fix the ssh problems, perhaps trying another slave launch
method might help?
BR,
Jukka Zitting
ime limit, would it make more sense to only disable
those jobs where >n of the last builds have failed?
BR,
Jukka Zitting
,
they don't really consume build resources. As soon as someone gets
around to fixing the pending errors, I expect the CI build to start up
again automatically to verify the fix.
I suggest that we only disable *periodic* building of codebases that
have been failing for a long time.
BR,
Jukka Zitting
Hi,
On Mon, Sep 13, 2010 at 8:21 PM, Stefan Seelmann
wrote:
> ubuntu1 and ubuntu2 seem to hang again.
Both slaves are now restarted.
BR,
Jukka Zitting
I just modified the PDFBox-trunk build configuration to archive the
pdfbox.log file for all builds.
BR,
Jukka Zitting
Hi,
On Mon, Aug 16, 2010 at 4:19 PM, Antoine Toulme wrote:
> Very cool! It looks like it broke our email notifications though.
>
> Can someone add hud...@hudson.apache.org to c...@buildr.apache.org please ?
Your list moderators should be able to do that.
BR,
Jukka Zitting
e sudo rights
I'll be able to help out where needed.
BTW, there's a Hudson security release 1.371 out. I'm upgrading
hudson.zones.apache.org now, and it would be good to upgrade the new
master too before switching over.
BR,
Jukka Zitting
.
Agreed, it's not much effort. My point is just that someone should at
least be aware of the build and how to get it adjusted if or when
needed. Now we had a case where nobody on our team knew about the
build when Buildbot started sending failure notifications to d...@tika.
That's what I want to avoid going forward.
BR,
Jukka Zitting
build, I think it's better to remove it.
[1] http://hudson.zones.apache.org/hudson/job/Tika-trunk/
BR,
Jukka Zitting
d command from "mvn clean install site" to just "mvn clean
install", or disable the failure notifications to d...@tika?
BR,
Jukka Zitting
advisories@ mailing list.
[1] http://infradna.com/content/security-advisory-2010-07-05
[2] http://wiki.hudson-ci.org/display/HUDSON/Security+Advisories
BR,
Jukka Zitting
g related to
ZOOKEEPER-778.
BR,
Jukka Zitting
better capture thread and stack dumps in case we need
to troubleshoot the Hudson instance.
BTW, how do I get started with OPIE on aegis?
BR,
Jukka Zitting
Hi,
The build [1] was apparently stuck in the integration test phase.
[1] http://hudson.zones.apache.org/hudson/job/Karaf/20/
BR,
Jukka Zitting
rently
running builds have completed.
BR,
Jukka Zitting
Hi,
The build [1] was stuck for over eight hours.
[1] http://hudson.zones.apache.org/hudson/job/james-server-v2.3-m2/803/
BR,
Jukka Zitting
ic warnings less
effective. Most of the time the instance recovers by itself, so we
don't react to each warning too quickly.
There's a new server in place that'll hopefully ease the overload
problems once we get Hudson up and running there.
BR,
Jukka Zitting
Hi,
On Mon, Jun 21, 2010 at 12:49 PM, Gav... wrote:
> Try now.
I'm in now. Thanks!
BR,
Jukka Zitting
d.
[1] http://hudson.zones.apache.org/hudson/job/CXF-Trunk-JDK15/1270/
BR,
Jukka Zitting
Hi,
On Fri, Jun 18, 2010 at 7:09 AM, Gav... wrote:
> Jukka, does it work for you?
Nope, I get the same error as Niclas saw
I'm using the "ju...@jzitting.day.com" key that I also use for
accessing people.apache.org.
BR,
Jukka Zitting
Hi,
The build [1] was successful, but got stuck for two days waiting for
build results to be transferred to Hudson master. Most likely another
overload problem with the master.
[1] http://hudson.zones.apache.org/hudson/job/Lucene-3.x/46/
BR,
Jukka Zitting
Hi,
On Thu, Jun 10, 2010 at 5:22 AM, Antoine Toulme wrote:
> it looks Hudson is still going through hiccups on vesta.
>
> Can it be restarted there please ?
Not sure if someone already did something about this, but now
everything seems to work fine on vesta.
BR,
Jukka Zitting
://hudson.zones.apache.org/hudson/job/Lucene-3.x/34/
BR,
Jukka Zitting
Hi,
On Mon, May 31, 2010 at 1:25 PM, Justin Mason wrote:
> ok, that would be great if it's doable. We'd also need to ensure auth traffic
> uses HTTPS if we're transmitting "real" ASF passwords.
Would it make sense to set up OpenID for something like this?
BR,
Jukka Zitting
we need to restart
Hudson.
BR,
Jukka Zitting
AFAIUI Hudson avoids this problem as long as the context path of the
Hudson webapp matches the local part of the proxy base URL. So we
should we fine as long as we use .../hudson in both places. I can
check that on a local Hudson installation to be sure.
BR,
Jukka Zitting
YI, as discussed on d...@pdfbox.a.o, I've enabled RAT checks with
up-to-date exclude rules in the PDFBox-trunk Hudson build.
We're using the RAT Maven plugin which only outputs text reports. I'll
see if I can update the plugin to output also XML.
BR,
Jukka Zitting
Hi,
The Hudson slave on lucene.zones.apache.org was failing due to a ping
timeout. I restarted the Hudson slave process and everything seems to
be back to normal.
BR,
Jukka Zitting
ment details in Hudson URLs.
Other than that the migration shouldn't cause too much trouble.
PS. Do you think the new server could handle Sonar as well (see
INFRA-2551)? I'd be happy to set it up.
[1] http://wiki.hudson-ci.org/display/HUDSON/Running+Hudson+behind+Apache
BR,
Jukka Zitting
/
BR,
Jukka Zitting
Hi,
On Mon, May 24, 2010 at 7:50 PM, Jukka Zitting wrote:
> Hudson was having trouble, so I restarted it.
Restarted it again, as we were receiving PermGen errors and other
oddness from Jackrabbit builds.
BR,
Jukka Zitting
Hi,
Until now we've configured minerva with two executors and vesta with
just one to avoid resource contention on the build servers. However,
it now seems that both nodes are configured to run four executors. Was
this a conscious change or an unintended side-effect of something
else?
BR,
Hi,
Hudson was having trouble, so I restarted it. Things should be back to
normal soon.
BR,
Jukka Zitting
Hi,
The build was taking over 19 hours, see [1]. I modified the
CXF-2.2.x-JDK15 configuration to automatically abort builds that take
longer than two hours.
[1] http://hudson.zones.apache.org/hudson/job/CXF-2.2.x-JDK15/425/
BR,
Jukka Zitting
Hi,
On Mon, May 17, 2010 at 2:02 PM, Niklas Gustavsson wrote:
> Next up, how do one get karma to the infra/Hudson project/component in
> JIRA so that I can close issues? :-)
By asking. :-) You now have the karma.
BR,
Jukka Zitting
n admin work, I've just been a bit
occupied during the last few weeks (my girlfriend had a surgery).
BR,
Jukka Zitting
Hi,
This build was stuck for over two days.
http://hudson.zones.apache.org/hudson/job/Chukwa-trunk/349/
BR,
Jukka Zitting
se/hudson/1268866800
BR,
Jukka Zitting
Hi,
On Wed, Mar 24, 2010 at 11:20 AM, Jukka Zitting wrote:
> On Wed, Mar 24, 2010 at 5:09 AM, Giridharan Kesavan
> wrote:
>> Can we work on moving the build off hudson master.
>
> I think we should do that, ideally reserving the master to tied jobs
> only.
I have changed
uilds and reconnected to the
slave.
BR,
Jukka Zitting
ooks like Hudson is getting confused by the conditional activation of
the JDK15 profile in the Maven build.
I added an explicit -PJDK15 option to the build, which seems to have
worked around this problem.
BR,
Jukka Zitting
Hi,
On Fri, Mar 26, 2010 at 10:06 AM, Jukka Zitting wrote:
> I'm seeing various problems (ssh polls failing, second executor on
> minerva dead, builds stuck, etc.) on Hudson. I'll do a restart and see
> if I can find more details in the logs.
Hudson was running out of memory.
Hi,
I'm seeing various problems (ssh polls failing, second executor on
minerva dead, builds stuck, etc.) on Hudson. I'll do a restart and see
if I can find more details in the logs.
BR,
Jukka Zitting
Hi,
On Thu, Mar 18, 2010 at 11:54 AM, Jukka Zitting wrote:
> Please configure the build job to discard old builds that are no
> longer needed, as excess historical records adversely affect Hudson
> performance. A good default setting is to discard old builds after 10
> or so days.
ir}
expands to "/home/hudson/hudson-slave/workspace/Jakarta BSF
3/bsf3.x").
BR,
Jukka Zitting
That's probably because JAVA_HOME is not set and there is no java on
the default PATH.
I just added JAVA_HOME=/home/hudson/tools/java/latest/bin/java to the
default environment of the slave process.
BR,
Jukka Zitting
on some custom tools installed in personal home
directories.
BR,
Jukka Zitting
aster
struts2
xwork2
Zookeeper-Patch-Admin
Which ones of these really must be run on the master?
BR,
Jukka Zitting
Hi,
On Wed, Mar 17, 2010 at 8:43 PM, Jukka Zitting wrote:
> I've just configured the zone as a Hudson slave and will be running a
> few builds there to see whether everything works OK. Once all the
> issues have been worked out, I'm planning to switch the zone from
> "
on the slave, so this should now be fixed.
BR,
Jukka Zitting
/job/struts2/
BR,
Jukka Zitting
lly
non-Maven) to hammer out any remaining issues before we make the zone
available for untied jobs.
BR,
Jukka Zitting
new slave. This should
release most of the pressure we're seeing on the Hudson master.
BR,
Jukka Zitting
Hi,
The Shiro build [1] was polling svn once every minute. I changed the
polling frequency to once an hour to match the infra recommendation.
[1] http://hudson.zones.apache.org/hudson/job/Shiro/
BR,
Jukka Zitting
Hi,
On Sun, Mar 14, 2010 at 6:37 AM, Patrick Hunt wrote:
> Can someone look into this? Pig on h7 seems to be stuck for >> 1 day. I
> tried killing it but no luck (using the hudson UI).
I restarted the Hudson slaves on h6 and h7. Both seem to be back to normal now.
BR,
Jukka Zitting
x27;d change the configuration so that
*untied* builds would no longer by default be running on hudson.zones
(currently they run on hudson.zones, minerva or vesta, depending on
where an available executor is found).
BR,
Jukka Zitting
t ~hudson/tools/maven/latest2/bin/mvn.
> Or perhaps a Maven guru can suggest how to rewrite the BSF pom
> eliminate the nested Maven call ?
As commented in the build.xml file, using the Maven Assembly plugin
for this seems like the obvious solution.
BR,
Jukka Zitting
get year).
BR,
Jukka Zitting
Hi,
This build [1] was stuck since yesterday.
PS. The subversion-1.6.x-solaris job is also running on the Lucene
zone. Is this intentional?
[1] http://hudson.zones.apache.org/hudson/job/subversion-1.6.x-solaris/88/
BR,
Jukka Zitting
x27;ve noticed that too. Is there anything we can do about that?
BR,
Jukka Zitting
Hi,
On Fri, Feb 19, 2010 at 5:37 PM, Justin Mason wrote:
> On Fri, Feb 19, 2010 at 14:42, Jukka Zitting wrote:
>> That's 30mins. Not sure what we can do to speed that up.
>
> Perhaps we should consider archiving old builds at this stage -- we
> have lots, alright.
Loaded all jobs
That's 30mins. Not sure what we can do to speed that up.
BR,
Jukka Zitting
Hi,
On Fri, Feb 19, 2010 at 12:59 PM, Kristian Waagan
wrote:
> Caused by: java.lang.OutOfMemoryError: GC overhead limit exceeded
I'm seeing this with other jobs as well. I'll restart Hudson as soon
as the currently running builds are done.
BR,
Jukka Zitting
but he is experienced
> with Hudson.
>
> Uwe's Apache account id is uschindler
I'll take care of this, see
https://issues.apache.org/jira/browse/INFRA-2501 for progress.
BR,
Jukka Zitting
Hi,
On Tue, Feb 16, 2010 at 11:14 AM, Justin Mason wrote:
> On Tue, Feb 16, 2010 at 09:58, Jukka Zitting wrote:
>> Perhaps we should install the audit trail plugin [1] to avoid
>> confusion like this.
>>
>> [1] http://wiki.hudson-ci.org/display/HUDSON/Audit+Tr
Hi,
On Mon, Feb 15, 2010 at 5:14 AM, Giridharan Kesavan wrote:
> This is the second time.
Perhaps we should install the audit trail plugin [1] to avoid
confusion like this.
[1] http://wiki.hudson-ci.org/display/HUDSON/Audit+Trail+Plugin
BR,
Jukka Zitting
Hi,
On Wed, Feb 10, 2010 at 2:24 PM, Jukka Zitting wrote:
> On Wed, Feb 10, 2010 at 2:21 PM, Justin Mason wrote:
>> will this break permalinks to existing projects and builds? e.g. from
>> mails, jira, wikis etc.
>
> Yes. :-(
Actually, it only breaks links to exi
Hi,
On Wed, Feb 10, 2010 at 2:21 PM, Justin Mason wrote:
> will this break permalinks to existing projects and builds? e.g. from
> mails, jira, wikis etc.
Yes. :-(
BR,
Jukka Zitting
r, see for example
http://hudson.zones.apache.org/hudson/view/J/view/Jackrabbit/ for the
updated Jackrabbit view.
WDYT, should we do this also for the other letters of the alphabet?
[1] http://wiki.hudson-ci.org/display/HUDSON/Nested+View+Plugin
BR,
Jukka Zitting
Hi,
On Mon, Feb 8, 2010 at 11:58 PM, Jukka Zitting wrote:
> Yep, I'll do the upgrade as soon as there's a good window between
> long-running build jobs.
The upgrade is now done.
BR,
Jukka Zitting
1 - 100 of 151 matches
Mail list logo
