Re: [2.0] slipping date

> On 20 Apr 2016, at 08:03, Daniel Beck wrote: > > >> On 29.03.2016, at 02:54, Kohsuke Kawaguchi wrote: >> >> • April 3: 1.656 release, which will be the last 1.x release >> > > FYI As a test run of the new infra, KK released 1.657 today, which will be > followed by 1.658. > > 2.0 wi

Re: [2.0] slipping date

> On 29.03.2016, at 02:54, Kohsuke Kawaguchi wrote: > > • April 3: 1.656 release, which will be the last 1.x release > FYI As a test run of the new infra, KK released 1.657 today, which will be followed by 1.658. 2.0 will be released later today or tomorrow, so these shouldn't matter t

Re: [2.0] slipping date

On Fri, Apr 8, 2016 at 4:32 AM, Arnaud Héritier wrote: > Why do we use Java 8 for CI ? > Java 7 is the oldest version we are supporting The build already uses `-target 7` and runs Animal Sniffer to prevent (accidental) use of Java 8+ APIs. 8 drops the PermGen, which in my experience makes it much

Re: [2.0] slipping date

Well Antonio might be up to something with the OOMs anyways. I am also experiencing OOMs on my test instance where I just updated the war from 1.650 to 2.0-rc and installed the "recommended plugins" ontop of what I already had, seems like I need to tweak both max heap and perm size from what was wo

Re: [2.0] slipping date

On Fri, Apr 8, 2016 at 1:25 PM, Kanstantsin Shautsou wrote: > Check dmesg, for OOM killer It's not a memory issue I believe, the test is just blocked for some reason. -- Antonio Muñiz Software Engineer CloudBees, Inc. -- You received this message because you are subscribed to the Google Grou

Re: [2.0] slipping date

Check dmesg, for OOM killer > On Apr 8, 2016, at 14:20, Antonio Muñiz wrote: > > One of the forked booters finished, the other one is blocked in a test. Any > idea? > > "Executing > retainMasterLabelWhenNoSlaveDefined(jenkins.model.MasterBuildConfigurationTest)" > prio=10 tid=0x7f47f40078

Re: [2.0] slipping date

One of the forked booters finished, the other one is blocked in a test. Any idea? "Executing retainMasterLabelWhenNoSlaveDefined(jenkins.model.MasterBuildConfigurationTest)" prio=10 tid=0x7f47f4007800 nid=0x3384 in Object.wait() [0x7f47fbe88000] java.lang.Thread.State: WAITING (on obje

Re: [2.0] slipping date

> On Apr 8, 2016, at 13:21, Antonio Muñiz wrote: > > On Fri, Apr 8, 2016 at 11:55 AM, Kanstantsin Shautsou > wrote: >> check what you did with pom and threading in pom/test plugins. > > Nothing was done, just -Xmx1g (which is not required IMO). > > I think celery is not being able to manage t

Re: [2.0] slipping date

https://ci.jenkins-ci.org/job/jenkins_main_trunk/4555 is running. No memory issues so far, but a test seems to be blocked for some other reason. On Fri, Apr 8, 2016 at 12:21 PM, Antonio Muñiz wrote: > On Fri, Apr 8, 2016 at 11:55 AM, Kanstantsin Shautsou > wrote: >> check what you did with pom a

Re: [2.0] slipping date

On Fri, Apr 8, 2016 at 11:55 AM, Kanstantsin Shautsou wrote: > check what you did with pom and threading in pom/test plugins. Nothing was done, just -Xmx1g (which is not required IMO). I think celery is not being able to manage the load and that's the root cause of OOM errors. Arnaud just checke

Re: [2.0] slipping date

Limit threading in build and check what you did with pom and threading in pom/test plugins. > On Apr 8, 2016, at 11:59, Antonio Muñiz wrote: > > On Fri, Apr 8, 2016 at 10:32 AM, Arnaud Héritier wrote: >> Java 7 is the oldest version we are supporting for now it must run with it >> :( > > > `7

Re: [2.0] slipping date

On Fri, Apr 8, 2016 at 10:32 AM, Arnaud Héritier wrote: > Java 7 is the oldest version we are supporting for now it must run with it > :( `7` is set anyways. So building with JDK 8 should be fine. BTW the test suite ran successfully in my local box - max memory consumption during tests: 500MB -

Re: [2.0] slipping date

Why do we use Java 8 for CI ? Java 7 is the oldest version we are supporting for now it must run with it :( On Fri, Apr 8, 2016 at 10:24 AM, Antonio Muñiz wrote: > The build is using JDK 7, could it be configured to use JDK 8? > (https://ci.jenkins-ci.org/job/jenkins_pipeline/branch/2.0 is stabl

Re: [2.0] slipping date

The build is using JDK 7, could it be configured to use JDK 8? (https://ci.jenkins-ci.org/job/jenkins_pipeline/branch/2.0 is stable with JDK 8) On Fri, Apr 8, 2016 at 10:18 AM, Antonio Muñiz wrote: > I'm trying to diagnose the OOME in tests on > https://ci.jenkins-ci.org/job/jenkins_main_trunk af

Re: [2.0] slipping date

I'm trying to diagnose the OOME in tests on https://ci.jenkins-ci.org/job/jenkins_main_trunk after merging 2.0 into master. Running tests locally. On Thu, Apr 7, 2016 at 5:22 PM, Daniel Beck wrote: > > On 07.04.2016, at 17:16, Robert Sandell wrote: > >> Where is the RC? > > Same URL as the betas

Re: [2.0] slipping date

On 07.04.2016, at 17:16, Robert Sandell wrote: > Where is the RC? Same URL as the betas, except for Docker and that's listed on https://hub.docker.com/r/jenkinsci/jenkins/tags/ -- I just need to get the 2.0 page updated and submit the blog post. Haven't had the time to do so yet, unfortunate

Re: [2.0] slipping date

Where is the RC? On Thu, Apr 7, 2016 at 4:26 PM, Daniel Beck wrote: > I pushed the merge to 'master'. So anything targeting 2.1+ can be now > proposed in pull requests to that branch. > > Anything happening on '2.0' branch will be limited to critical fixes for > the 2.0 release specifically. > >

Re: [2.0] slipping date

I pushed the merge to 'master'. So anything targeting 2.1+ can be now proposed in pull requests to that branch. Anything happening on '2.0' branch will be limited to critical fixes for the 2.0 release specifically. On 07.04.2016, at 12:08, Daniel Beck wrote: > > On 07.04.2016, at 02:37, Dani

Re: [2.0] slipping date

On 07.04.2016, at 02:37, Daniel Beck wrote: > Given the late hour, I will merge 2.0 into master tomorrow. The merge commit is here: https://github.com/daniel-beck/jenkins/commit/1fe9cf7b7ada45230f2bc5b8e2f1bdb93175ff9f If you'd like to give it a final review, the time to do so is right now. I

Re: [2.0] slipping date

On 29.03.2016, at 02:54, Kohsuke Kawaguchi wrote: > • April 6: RC. > master will be merged into 2.0. Ideally we're not going to merge crazy stuff > into 'master' between now and next Wednesday to keep the risk of the 2.0 RC > being terrible low. Any unmerged pull requests against '2.0' w

Re: [2.0] slipping date

On 05.04.2016, at 02:06, Kanstantsin Shautsou wrote: > 1.x without fixes for 2 months and 2.x that would be unstable? Not sure what you're referring to. As you can see in the plan posted by Kohsuke, we'll skip just two weekly releases, something we did around New Year's without notable issues

Re: [2.0] slipping date

So, everybody would have 1.x without fixes for 2 months and 2.x that would be unstable? On Tuesday, March 29, 2016 at 3:54:36 AM UTC+3, Kohsuke Kawaguchi wrote: > > (For dev list...) > > This impacts the branching & 1.x release freeze plan >

Re: [2.0] slipping date

(For dev list...) This impacts the branching & 1.x release freeze plan announced by Daniel earlier, too. - April 3: 1.656 release, which will be the last 1.x release - April 6: RC. master will be merged into 2.0. Id

[2.0] slipping date

In today's 2.0 status check call, Daniel Beck as 2.0 release manager had raised that we still have too many issues in flight and it's unrealistic to expect that we can fix them all in RC. So we decided to add one more week till RC.