Re: fonts in junit trend graphs are unreadable

2012-04-20 Thread Didier Durand

Hello,

You should read 
https://wiki.jenkins-ci.org/display/JENKINS/Jenkins+got+java.awt.headless+problem

it may help you solve your issue

regards

didier

Le lundi 16 avril 2012 22:57:20 UTC+2, Alex Dean a écrit :

 I've just set up Jenkins to run PHPUnit tests. Everything is working 
 fine except that the junit trend graphs are very difficult to read. 
 The font is all wrong. 

 https://img.skitch.com/20120416-deuen41cigddrgt7nfyhmyi3f4.png 

 I'm unsure how to further research this issue. (Searches for 'trend 
 report font'  similar haven't produced anything useful.) Can anyone 
 offer any pointers? 

 thanks, 
 alex 


 Centos 5.6 

 $ java -version 
 java version 1.6.0_20 
 OpenJDK Runtime Environment (IcedTea6 1.9.8) (rhel-1.22.1.9.8.el5_6- 
 x86_64) 
 OpenJDK 64-Bit Server VM (build 19.0-b09, mixed mode) 

 $ rpm -qa | grep jenkins 
 jenkins-1.460-1.1



Re: ERR_CONTENT_DECODING_FAILED when try to access job configuration pages

2012-04-16 Thread Didier Durand
Hi Jason,

I had this same problem with 1.458 on CentOS (due to the gzip page 
compression introduced then) 

see 
https://groups.google.com/forum/?fromgroups#!searchin/jenkinsci-users/cobertura/jenkinsci-users/OJxWOXpSsMs/ReNTRVSFkFAJ

I upgraded this morning to 1.460 and the problem disappeared: you may also 
try the update.

regards

didier

Le mardi 17 avril 2012 06:45:12 UTC+2, Jason Axelson a écrit :

 Hi,

 I'm getting Error 330 (net::ERR_CONTENT_DECODING_FAILED): Unknown
 error. in Chrome when I try to access job configuration pages.
 Firefox gives a more generic error but also cannot access it.

 This is on Jenkins 1.459 running on Ubuntu 11.10. Nothing is showing
 up in the Jenkins Log. This happens on all my projects even when I
 create a new job (can't even set any of the settings but it still
 creates the empty job). Does anyone have ideas what could cause this?

 I am about to try restarting after some builds finish. And it works
 after restarting but I still want to send this out in case anyone gets
 a similar error.

 It may be related to just installing 1.459 last Friday, not sure
 exactly which version I was on before that but it was probably around
 a month old.

 Thanks,
 Jason



Re: Simple newbie question on how Post-build actions work...

2012-03-29 Thread Didier Durand
Hi Richard,

Generally speaking, you should never check-in results of tools produced 
elsewhere for use on Jenkins. You should rather implement those tools on 
your Jenkins server, run them there to obtain the results that you need in 
further steps of your Jenkins jobs.

regards

didier

Le jeudi 29 mars 2012 09:21:28 UTC+2, Richard Berger a écrit :

 OK, I think I figured it out... the key was that in the Advanced 
 settings for the Invoke Ant task, for the Build File I had put the 
 path to my eclipsestuff/workspace/project/build.xml - so all the 
 references inside that build.xml where being interpreted as relative 
 to eclipsestuff. 

 Realizing that my build.xml will also exist in the Jenkins workspace, 
 I can specify THAT build.xml in the build file line.  I am running 
 into some other problems so I haven't fully verified this works.  But 
 it *feels* more correct. 

 Thanks for listening... (and if anyone can confirm that this the right 
 approach that would be much appreciated - on the other hand, if I am 
 totally lost, feel free to let me wander aimlessly for a 
 while :) :) ). 

 RB 

 On Mar 28, 11:58 pm, Richard Berger richardlan...@gmail.com wrote: 
  I am playing with Jenkins (which seems really cool) and have a 
  question on coordinating Post-build actions, for example Publish 
  FindBugs analysis results. 
  
  I started this project in Eclipse and the code resides in the 
  directory eclipsestuff/workspace/project (not the real name).  The 
  code is also checked into SVN.  When I update a file, this correctly 
  triggers a build.  In my build.xml file there is the FindBugs task, 
  which creates an output file of output.xml.  From within Eclipse, it 
  all works fine. 
  
  Now, when I set up the project in Jenkins, I have to enter something 
  for the FindBugs results file.  And, more specifically, that file 
  name has to be in my Jenkins workspace, which is pulled from SVN (or 
  at least I couldn't find a way to specify a path to eclipsestuff/ 
  workspace/project/output.xml).  So I think that my output.xml results 
  need to get checked into SVN so that Jenkins can then pull them back 
  out to do the FindBugs analysis.  But that doesn't seem right either. 
  
  I solved this by having the ant task (inside my build.xml) create 
  the output.xml in trunk/Project/output.xml.  But that seems somehow 
  wrong. 
  
  Before I apply this solution to checkstyle and Junit, I thought I 
  should check in and get some advice.  I have a feeling I am missing 
  something very obvious.  (My question seems a little similar tohttp://
 groups.google.com/group/jenkinsci-users/browse_thread/thread/b..., 
  but even simpler). 
  
  Thanks!!! 
  RB



Trailing WInstone error message on job configure page

2012-03-28 Thread Didier Durand
Hello,

Since 1.456 (also in 1.457), I get the following error message from 
winstone in the Job configuration page for each job after current 
configuration is displayed

Status Code: nullException: 
Stacktrace: 

(none)


--
*Generated by Winstone Servlet Engine v0.9.10 at Wed Mar 28 08:00:18 CEST 
2012*

Do I have a config issue specific to my machine or is it a new problem for 
those releases for which I should open a JIRA ticket ?

regards

didier


Re: Change svn check-out strategy for automatic/manuel build

2012-03-25 Thread Didier Durand
Hi Chitech,

Agreed with Jan: what you propose is a lot of work compared to my solution 
achieved in a few clicks...

Up to you

regards

didier

Le samedi 24 mars 2012 11:09:34 UTC+1, chitech a écrit :

 We found a way around it.

 1.0 Make a cron job to modify the config.xml to Clean workspace
 1.1 Call reload configuration via disk from cron job
 2.0 The hudson run the nightly builds
 3.0 Make a cron job to modify the config.xml back to emualte clean
 3.1 Call reload configuration via disk from cron job

  

 On Thursday, March 22, 2012 1:34:16 PM UTC+1, chitech wrote:

 When an user schedule a build we would like to use svn check-out 
 strategy: 
 ‘Emulate clean checkout by first deleting unversioned/ignored files, 
 then 'svn update'’ 

 But for nightly build we would like to use this strategy: 
 Clean workspace and then checkout 

 We want this because the build time will be shorter in the daytime. 
 Can we 
 somehow achieve this setup?


 On Thursday, March 22, 2012 1:34:16 PM UTC+1, chitech wrote:

 When an user schedule a build we would like to use svn check-out 
 strategy: 
 ‘Emulate clean checkout by first deleting unversioned/ignored files, 
 then 'svn update'’ 

 But for nightly build we would like to use this strategy: 
 Clean workspace and then checkout 

 We want this because the build time will be shorter in the daytime. 
 Can we 
 somehow achieve this setup?


 On Thursday, March 22, 2012 1:34:16 PM UTC+1, chitech wrote:

 When an user schedule a build we would like to use svn check-out 
 strategy: 
 ‘Emulate clean checkout by first deleting unversioned/ignored files, 
 then 'svn update'’ 

 But for nightly build we would like to use this strategy: 
 Clean workspace and then checkout 

 We want this because the build time will be shorter in the daytime. 
 Can we 
 somehow achieve this setup?


 On Thursday, March 22, 2012 1:34:16 PM UTC+1, chitech wrote:

 When an user schedule a build we would like to use svn check-out 
 strategy: 
 ‘Emulate clean checkout by first deleting unversioned/ignored files, 
 then 'svn update'’ 

 But for nightly build we would like to use this strategy: 
 Clean workspace and then checkout 

 We want this because the build time will be shorter in the daytime. 
 Can we 
 somehow achieve this setup?


 On Thursday, March 22, 2012 1:34:16 PM UTC+1, chitech wrote:

 When an user schedule a build we would like to use svn check-out 
 strategy: 
 ‘Emulate clean checkout by first deleting unversioned/ignored files, 
 then 'svn update'’ 

 But for nightly build we would like to use this strategy: 
 Clean workspace and then checkout 

 We want this because the build time will be shorter in the daytime. 
 Can we 
 somehow achieve this setup?


 On Thursday, March 22, 2012 1:34:16 PM UTC+1, chitech wrote:

 When an user schedule a build we would like to use svn check-out 
 strategy: 
 ‘Emulate clean checkout by first deleting unversioned/ignored files, 
 then 'svn update'’ 

 But for nightly build we would like to use this strategy: 
 Clean workspace and then checkout 

 We want this because the build time will be shorter in the daytime. 
 Can we 
 somehow achieve this setup?



Re: Change svn check-out strategy for automatic/manuel build

2012-03-22 Thread Didier Durand
Hi,

I guess that you have to create 2 jobs:

   - daily: with trigger on each change and simple svn update
   - nightly: with time scheduling and and fresh svn checkout

That means that you'll have a separate workspace for each of those jobs

regards


didier

Le jeudi 22 mars 2012 13:34:16 UTC+1, chitech a écrit :


When an user schedule a build we would like to use svn check-out 
 strategy: 
 ‘Emulate clean checkout by first deleting unversioned/ignored files, 
 then 'svn update'’ 

 But for nightly build we would like to use this strategy: 
 Clean workspace and then checkout 

 We want this because the build time will be shorter in the daytime. 
 Can we 
 somehow achieve this setup?



Re: How to make variables available to all build steps/action in a job

2012-03-19 Thread Didier Durand
Hi,

You can go into Jenkins Configure  Global Properties  there set env 
variables that are available to all jobs.

regards

didier

Le lundi 19 mars 2012 09:59:09 UTC+1, Jan Seidel a écrit :

 Hi Shanz, 

 *uhm* mkey... I am not familiar with python but I will give it a try. 
 Seems to be quite convenient to adapt the properties to my needs. 

 Thanks for your feedback :) 

 Take care 
 Jan 

 On 16 Mrz., 16:34, shanz duncan.perr...@gmail.com wrote: 
  I found that a python script gave me more flexibility in what I could 
  attach to an email... 
  
  import smtplib 
  import os 
  from email.MIMEMultipart import MIMEMultipart 
  from email.MIMEBase import MIMEBase 
  from email.MIMEText import MIMEText 
  from email.Utils import COMMASPACE, formatdate 
  from email import Encoders 
  
  myRelease = os.environ.get(BUILD_STRING) 
  
 files=[myData.xml,myDataNightly_Previous.xml,myDataRelease_Previous.xml,
  

  myReport.html] 
  to=[an.ema...@address.com,an.ema...@address.com,
 an.ema...@address.com] 
  text1 = To everyone\n\nmyData.xml has changed following the latest 
  build :-  
  text2 = myRelease 
  text3 = \r\n\nSee attached files\r\n\r\nReleases are in\netdrive 
  \Product source code\Tagged\, \r\n\nNightly files are in  
  \netdrive\Product source code\Nightly\\r\n\nRegards,\r\n\nJenkins\r 
  \n 
  text = text1+text2+text3 
  
  # Send the email via our own SMTP server. 
  def send_mail(to, subject, text, fromWho=, files=[], cc=[], bcc=[], 
  server=10.192.00.000): 
  assert type(to)==list 
  assert type(files)==list 
  assert type(cc)==list 
  assert type(bcc)==list 
  
  message = MIMEMultipart() 
  message['From'] = fromWho 
  message['To'] = COMMASPACE.join(to) 
  message['Date'] = formatdate(localtime=True) 
  message['Subject'] = subject 
  message['Cc'] = COMMASPACE.join(cc) 
  
  message.attach(MIMEText(text)) 
  
  for f in files: 
  part = MIMEBase('application', 'octet-stream') 
  part.set_payload(open(f, 'rb').read()) 
  Encoders.encode_base64(part) 
  part.add_header('Content-Disposition', 'attachment; 
  filename=%s' % os.path.basename(f)) 
  message.attach(part) 
  
  addresses = [] 
  for x in to: 
  addresses.append(x) 
  for x in cc: 
  addresses.append(x) 
  for x in bcc: 
  addresses.append(x) 
  
  smtp = smtplib.SMTP(server) 
  smtp.sendmail(fromWho, addresses, message.as_string()) 
  smtp.close() 
  
  send_mail(to, myData.xml, text, jenkins, files)



Re: Jenkins vs. Hudson

2012-03-19 Thread Didier Durand
Hi,

A recent ppt by Cloudbees clearly demonstrated (via manay figures) that the 
community has switched from Hudson to Jenkins.

http://www.cloudbees.com/sites/default/files/whitepapers/Jenkins_Safe_Investment_Final.pdf
 
- many figures there.

http://blog.cloudbees.com/2011/12/jenkins-community-survey-results-82.html

regards

didier


Le lundi 19 mars 2012 09:19:34 UTC+1, zakyn a écrit :

 Hi, 

 I would like to ask you how jenkins and hudson community cooperates or 
 where is the main development. 

 I have not met any information about the synergy of both or not. Of 
 course I am interested in plugins development and so on. 

 Thank you. 

 Vladimir



Re: Jenkins - getting slower and slower

2012-03-18 Thread Didier Durand
Hi,

To disable plugin via the HDD, you can have to delete them in 
$JENKINS_HOME/plugins (bot .hpi file and directory with same name

regards

didier

Le dimanche 18 mars 2012 22:16:44 UTC+1, zakyn a écrit :

 Hello, 

 I have problem that Jenkins is getting slower and slower and it is not 
 possible to manage plugins. So I would lilke to ask you how I can 
 disable plugin directly on HDD. 

 Thank you. 

 Best regards, 

 Vladimir



Re: Archive only X month old builds

2012-03-16 Thread Didier Durand
Hi,

In each job configuration, you have Discard old builds (among the 1st 
checkmarks) where you select either the number of jobs or how many days you 
want to archive.

regards

didier


Le vendredi 16 mars 2012 13:44:49 UTC+1, zakyn a écrit :

 Hello, 

 I am looking for the plugin which allows me to archive only 1 or 2 
 month old builds. I know I can specify  the number of archived builds 
 (not only artifacts but the builds themselves) but it is not enough 
 for me. 

 The reason for that is that I am facing now to the problem with java 
 heap space and i think it is because of the history of builds. 

 Thank you for your help. 

 Vladimir



Re: Instable builds related to low memory

2012-03-09 Thread Didier Durand
Hi,

Did you try to try Java heap size via the Java options (advanced section) 
of the step config parameters using Xnx and Xms ?

That should solve your issue. [I dodn't think the size of physical real 
memory really matters for completion: even if you allocate more heap than 
physical memory, your server will paginate and extend duration but it 
should go to end with the right settings of Xmx / Xms.

I would also recommend you to install the Jenkins monitoring plugin based 
on Java Melody: very useful to follow memory comsumption

regards

didier


Le vendredi 9 mars 2012 12:47:25 UTC+1, dacay a écrit :

 Hi everyone,

 There is a problem I have been struggling with Jenkins. I tried 
 different CI servers but none of them offers the same features Jenkins 
 has, so I want to use it.

 I have a small Rackspace CloudServer instance with 256MB RAM and running 
 Jenkins with Apache Tomcat 7 in that server. Using Maven 2 and Git 
 plugins of Jenkins.

 The problem is, either Maven JVM exits with the code 137 (when forking 
 is enabled) or I get mysterious crashes. I know that the 256MB of RAM 
 may not be enough for Jenkins; but the thing is: even when I am running 
 Jenkins, I can build the project via command line in a short period of 
 time. My project involves some heavy servers, and generally the test are 
 failing with Jenkins.

 If I can build the project via command line just fine when the Jenkins 
 server is running (meaning more than one JVM), isn't that a problem with 
 the Jenkins? I tried different combinations of fork configurations and 
 Maven JVM settings, even when build is OK for once, then it fails in 
 subsequent builds.

 I even set maximum number of builds to keep to 3, but it didn't help.

 Any comments are appreciated.

 Best regards,

 Deniz



Re: Jenkins Hung

2012-03-08 Thread Didier Durand
Hi,

you can access the system log via Jenkins Homepage  Manage Jenkins  
System log

regards

didier

Le mercredi 7 mars 2012 20:24:16 UTC+1, John Hinnegan a écrit :

 Jenkins hung on me again. It just stops responding to requests. I wish it 
 would just die so it would be restarted.

 Anyway, following along: 
 https://wiki.jenkins-ci.org/display/JENKINS/Jenkins+is+hanging

 Here's the JStack: http://pastebin.com/ZSNQgCm7
 Generated with sudo jstack -F 64871

 Heres's the JMap: http://pastebin.com/Mytcsnpv
 Generated with sudo jmap -heap 64871

 It's OSX. My version of JMap didn't seem to understand what the 
 -dump:format=b,file=heap.hprof tip in the wiki was going for. Can someone 
 give me the command line to see how that would work?

 Also, I tried to get the core dump, but it seems on OS X they are 
 suppressed by default, at least according to: 
 http://stackoverflow.com/questions/2080918/where-are-core-dumps-written-to-in-mac-os-x

 I didn't know to run the ulimit command first, so it seems I don't have a 
 dump to share. 

 Suggestions on what to do next time to help diagnose the issue appreciated.




Re: Maintain Jobs in SCM

2012-03-06 Thread Didier Durand
Hi,

You can insert env variables in your paths on Jenkins SCM section in
job configuration

So, what you can do to avoid recreating jobs for branches is to put a
variable for the path of the branch and set the value of the variable
just before lanching build step ot the job. [I use this technique to
be able to easily change at at once the URL root of the source code I
am building in my jobs]

Something that I didn't try is to use a matrix job to set this
variable successively to various values (all the possible branch
names) and then have Jenkins run them automatically but the odds are
good that it should work.

regards

didier

On Mar 7, 2:10 am, Conormull conor.mul...@gmail.com wrote:
 Hi all

 As we achieve more advanced build pipeline features with out Jenkins
 jobs, we would now like to maintain those jobs in our SCM. Each time
 we cut a new branch to continue developer we now find ourselves
 recreating our Jenkins jobs to point to the new branch. I was just
 wondering if anyone is doing something similar and who you may have
 gone about it from an end to end perspective. When and how are you
 uploading the newest version of the job etc. Any help would be much
 appreciated.

 Kind Regards
 Conor


Having a given job running absolutely alone on my jenkins: how ?

2012-03-02 Thread Didier Durand
Hello,

During night, I want to launch a jenkins job that makes all backups
for all the ressources (database, svn, etc) that all other jenkins
jobs use.

So, I have to make sure that this job is running absolutely alone.

I already have tenths of jobs configured and I don't want to change
something in their config because of this new backup job.

What is the best way to achieve that ?

My current plan:
   a) schedule the job at 02:00 am
   b) make the job reduce the number of executors to 1 for itself only
   c) query the queue and executors to wait that all other job finish
   d) run the backup activities
   e) restore the executors to their original number
   f) finish the backup job

Questions:
  1) what is the correct way (API) to reduce the executors
programmatically ?
  2) is there anything better than my plan ? (I studied LockLatch and
Throttle plugin but they don't seem appropriate for that purpose)

Thanks a lot in advance !

regards

didier


Re: Having a given job running absolutely alone on my jenkins: how ?

2012-03-02 Thread Didier Durand
Hi Jess  Matt,

Thanks a lot for your quick answers !

I'll try the Exclusive Execution Plugin first as it is my exact use
case. Then, the others if needed.

regards

didier


On Mar 2, 5:41 pm, matthew.web...@diamond.ac.uk wrote:
 I've not used it, 
 buthttps://wiki.jenkins-ci.org/display/JENKINS/Exclusive+Execution+Pluginmatches
  your use case exactly.









  -Original Message-
  From: jenkinsci-users@googlegroups.com 
  [mailto:jenkinsci-users@googlegroups.com] On Behalf Of Didier
  Durand
  Sent: 02 March 2012 16:37
  To: Jenkins Users
  Subject: Having a given job running absolutely alone on my jenkins: how ?

  Hello,

  During night, I want to launch a jenkins job that makes all backups
  for all the ressources (database, svn, etc) that all other jenkins
  jobs use.

  So, I have to make sure that this job is running absolutely alone.

  I already have tenths of jobs configured and I don't want to change
  something in their config because of this new backup job.

  What is the best way to achieve that ?

  My current plan:
     a) schedule the job at 02:00 am
     b) make the job reduce the number of executors to 1 for itself only
     c) query the queue and executors to wait that all other job finish
     d) run the backup activities
     e) restore the executors to their original number
     f) finish the backup job

  Questions:
    1) what is the correct way (API) to reduce the executors
  programmatically ?
    2) is there anything better than my plan ? (I studied LockLatch and
  Throttle plugin but they don't seem appropriate for that purpose)

  Thanks a lot in advance !

  regards

  didier

 --
 This e-mail and any attachments may contain confidential, copyright and or 
 privileged material, and are for the use of the intended addressee only. If 
 you are not the intended addressee or an authorised recipient of the 
 addressee please notify us of receipt by returning the e-mail and do not use, 
 copy, retain, distribute or disclose the information in or attached to the 
 e-mail.
 Any opinions expressed within this e-mail are those of the individual and not 
 necessarily of Diamond Light Source Ltd.
 Diamond Light Source Ltd. cannot guarantee that this e-mail or any 
 attachments are free from viruses and we cannot accept liability for any 
 damage which you may sustain as a result of software viruses which may be 
 transmitted in or with the message.
 Diamond Light Source Limited (company no. 4375679). Registered in England and 
 Wales with its registered office at Diamond House, Harwell Science and 
 Innovation Campus, Didcot, Oxfordshire, OX11 0DE, United Kingdom


Re: Port issue

2012-02-27 Thread Didier Durand
Hi,

When you say that port 8080 gets down does it mean that the jenkins
server is down: i.e do you find it  via ps -ef on linux console ?

Other question; is your jenkins running natively (suning embedded
Winstone) or within a container (tomcat or other)?

regards

didier

On Feb 27, 9:19 am, Suri suresh1256...@gmail.com wrote:
 Hi

 Our Jenkins server is Red Hat Linux.
 We are frequently getting Webpage issue in Jenkins. i have checked
 Jenkins service in server, When webpage is down.
 Jenkins service status showing as running . but the webpage is down.
 Not able to access Jenkins Webpage.

 I have found port 8080 is not running in server. It is frequently
 getting down automatically.
 We  don’t know, why the port is getting down?
 I have checked from Server and Network end. It seems to be well.

 For temporarily solution We have created a crontab entry with a
 Jenkins service startup script, it will check the 8080 status, when
 ever the 8080 port is down it will automatically restart the Jenkins
 service
 Can anyone please advise me, Why the port is getting down
 automatically.


Re: Jenkins (JVM) crashes after few minutes after start

2012-02-24 Thread Didier Durand
Hi,

Do you have the Java stack trace of the crash?

regards

didier

On Feb 24, 4:04 pm, Domen Kožar do...@dev.si wrote:
 I have been fighting with this one for a month, would be really
 pleased if someone can give tips or request more data:

 Crosspost:http://serverfault.com/questions/355516/jenkins-jvm-crashes-after-few...


Re: anonymous disables jobs

2012-02-24 Thread Didier Durand
Hi,

Was the number of executors changed in the Jenkins config to 0 at some
point ?

regards

didier

On Feb 24, 5:32 pm, noa kuki...@gmail.com wrote:
 hi,

 I have a promblem with my jenkins server, some of my clients
 complained about their jobs being disabled and after looking into it I
 saw the jobs are being disabled by an anonymous user.
 the wierd thing is that my jenkins server has permissions and the user
 anonymous has no permissions at all.
 the authenthincation is with an ldap server.

 I couldn't find anything about that problem on the internet so any
 help would be appriciated.

 thank you :)


Re: JMS trigger

2012-02-23 Thread Didier Durand
Hi,

If you go to the plugins page and check the Build Triggers section,
you'll see that many plugins have the same intent as yours: FileFound
plugin, FST plugin, IRC plugin, URL change plugin.

So, you should definitely do it.

Btw, the plugins mentionned probably provide good source code
skeletons for you

regards

didier

On Feb 24, 1:51 am, tommyB thomas.p.bar...@gmail.com wrote:
 So I have been using Jenkins for a couple of years to periodically run
 some etl/batch jobs.  Most of the jobs are simple enough that I just
 need a groovy script or two to take care business.  So anyway, I am
 starting to get requests to start dealing with real time data from
 JMS.  The data processing is fairly simple, and Jenkins has been a
 life saver for notification, logging and its restful api.  If I were
 to do what I need to consume and process data from a message queue, I
 basically would be creating a crappy version of Jenkins.  So I
 thought, why not extend Jenkins?  So here's my proposed use case...
 please tell me if this is insane.  I was thinking I could make a JMS
 trigger that can fire off a job if messages are in a queue; the job
 would run until all messages have been consumed.  I haven't seen
 anyone do this so I wonder if this is beyond the scope of
 Jenkins.


Re: non-recursively checkout out an svn tree

2012-02-21 Thread Didier Durand
Hi,

Is it possible for you to create a dir source/Makedir and move
Makefile into it ?
At least can you create Makedir and make a symlink to makefile into
it.

Then, you configure 2 checkout location in you jenkins job config: 1
for source/cpp and 1 for source/Makedir and you've reached your
objective.

regards

didier

On Feb 21, 6:30 pm, Chris Withers ch...@simplistix.co.uk wrote:
 Hi All,

 I have a project with a tree structure roughly like this:

 source
 source/Makefile
 source/python
 source/cpp
 source/sql
 source/lotsofotherstuff

 I have a job that builds the C++ release and then runs its unit tests.
 This requires:

 source/Makefile
 source/cpp

 Since Makefile is a file, not a folder, and one that for arcane reasons
 I can't move from its location, I can't *only* checkout Makefile and the
 cpp folder. This is resulting in loads of spurious executions of this
 build job as a result of changes to other parts of the tree that don't
 affect the C++ release at all.

 How can I *just* checkout the Makefile and cpp folders and have those
 used as the things that jenkins looks at for changes?

 cheers,

 Chris

 --
 Simplistix - Content Management, Batch Processing  Python Consulting
              -http://www.simplistix.co.uk


Re: jenkins svn update

2012-02-16 Thread Didier Durand
Hi Fiona,

The place to increase memory is Job Configuration of your job  Build
 Invoke Ant  Advanced  Java Options

What I can suggest to see if the issue comes from your CI machine or
the SVN server: connect to the CI machine via ssh, run the checkout in
the terminal session via the native svn client using 'svn co etc.'
command and see if time differs much from what you get when Jenkins
itself does the checkout via the SvnKit java library.

What I can say is that checkout is long is you have lots of code: I
regurlarly work on a project with 10+ millions lines of code. It takes
4 minutes to check this project out (jenkins and svn hosted on same
very powerful machine from I/O standpoint: all disks are SSDs)

regards

didier


On Feb 16, 7:02 am, fiona bingxue0...@hotmail.com wrote:
 hi,
 I open my job config, there no place to increase the -Xmx -Xms , I
 increase it in my enviornment variable JAVA_OPTS,the Xmx is 1024m, Xms
 is 128m, it seems like no effect.
 BTW,my source code is too much.

 On 2月16日, 下午1时20分, Didier Durand durand.did...@gmail.com wrote:







  Hi,

  Maybe not enough memory for the JVM running the job to handle the svn
  protocol efficiently

  Try to increase the -Xmx -Xms on the Java Options of your job config.

  I would be interested to read about your results with this proposal or
  any other.

  regards

  didier

  On Feb 16, 6:15 am, fiona bingxue0...@hotmail.com wrote:

   if my source code is managed by svn, when a new build start, it will
   update first,but I find it very slow, if I execute command 'svn
   update' alone, it is update fast. what factor maybe affect the speed?- 
   隐藏被引用文字 -

  - 显示引用的文字 -


Re: jenkins svn update

2012-02-15 Thread Didier Durand
Hi,

Maybe not enough memory for the JVM running the job to handle the svn
protocol efficiently

Try to increase the -Xmx -Xms on the Java Options of your job config.

I would be interested to read about your results with this proposal or
any other.

regards

didier

On Feb 16, 6:15 am, fiona bingxue0...@hotmail.com wrote:
 if my source code is managed by svn, when a new build start, it will
 update first,but I find it very slow, if I execute command 'svn
 update' alone, it is update fast. what factor maybe affect the speed?


Re: Job dependencies

2012-02-15 Thread Didier Durand
Hi,

You could create a job C that is triggered by the same events as B

C will only do 1 thing: have A a downstream job and A will have B as
downstream job

So, when C triggers (based on the triggers that you have chosen) it
launches A. When A finishes, it launches B

That's 1 way: maybe will people come with smarter solutions

regards

didier

On Feb 15, 4:57 pm, Alper Tekinalp alper.tekin...@gmail.com wrote:
 Any idea please?

 2012/2/15 Alper alper.tekin...@gmail.com

  Hi.
  I have two jobs A, B and B is dependent to A. So how can I say every
  time before building job B build job A?
  Best regards.

 --
 Alper Tekinalp
 Hermes İletişim
 atekinalp.blogspot.com


Re: Source code metrics plugin ?

2012-02-06 Thread Didier Durand
Hi,

I personally use SLOCCOUNT: if you work in an environment with
multiple languages, it's particularly well suited to get measures in
such an heterogeneous environment.

regards

didier


On Feb 6, 1:49 pm, Jan Goyvaerts java.arti...@gmail.com wrote:
 AHA ! NOW they're hiding ! :-)

 Error formatting macro: contentbylabel:
 java.util.ConcurrentModificationException

 When looking to the reporting plugins.







 On Mon, Feb 6, 2012 at 13:44, Jan Goyvaerts java.arti...@gmail.com wrote:
  It was of course a figure of speech. :-) I meant that there's so much
  choice ! ;-)

  Thanks !

  On Mon, Feb 6, 2012 at 13:41, Fred G fred.g...@googlemail.com wrote:

  Hi,

  here is the link to all the reporting plugins:
 https://wiki.jenkins-ci.org/display/JENKINS/Plugins#Plugins-Buildreports

  One of the most popular plug-ins is this one:
 https://wiki.jenkins-ci.org/display/JENKINS/Static+Code+Analysis+Plug...

  BTW: The plug-ins are not hiding, they can easily be found with the
  search feature of the Jenkins wiki. ;)

  Hth,

  Fred