Thanks for the feedback everyone, it was quite helpful. Regards, Andres
On Thursday, October 15, 2015 at 2:00:22 PM UTC-4, Brian Ray wrote: > > See this earlier thread for the Iterator serialization issue: > > https://groups.google.com/d/msg/jenkinsci-users/LGRv7Jq60YI/ZN-68zuw2loJ > > On Wednesday, October 14, 2015 at 10:33:41 AM UTC-7, Michael Štědrý wrote: >> >> Hi Andres, >> >> I've got one more interesting thing for you. I managed to get your sample >> working even though I had to remove the functions: >> >> def testList = "test1 test2".split() >> def branches = [:] >> for (int i = 0; i < testList.length; i++) { >> def testName = testList[i]; >> branches[testName] = { >> echo testName >> sleep 20 >> } >> } >> parallel branches >> >> What I found out is that I couldn't use the nice for loop because it >> fails with: >> >> java.io.NotSerializableException: java.util.AbstractList$Itr >> >> ... >> >> Caused by: an exception which occurred: >> in field itr >> in field target >> in field continue_ >> in field parent >> in field capture >> in field def >> in field closures >> in object org.jenkinsci.plugins.workflow.cps.CpsThreadGroup@8b36066 >> >> >> And when I put the creation of the branches in a function it started to >> behave strangely. >> >> >> On Sunday, September 27, 2015 at 10:25:47 PM UTC+2, Andres Rodriguez >> wrote: >>> >>> Hi Everyone, >>> >>> First of all, I want to say that I'm really enjoying the workflow >>> plugin. Thanks for all the effort in bringing this feature to production. >>> Specially loading the DSL file from a git repository allows us to auto-test >>> our changes to a job through the same mechanism that we test changes to the >>> code (using Gerrit). >>> >>> I'm currently encountering some weird behavior when generating a list of >>> tests parallel branches dynamically. I've boiled it down to the following >>> small test case that you can paste into the workflow script box: >>> >>> @NonCPS >>> def generateTestBranches (testList) { >>> def branches = [:] >>> for (testName in testList.split()) { >>> branches[testName] = { >>> echo testName >>> sleep 20 >>> } >>> } >>> return branches >>> } >>> >>> def runTests (testList) { >>> def testBranches = generateTestBranches(testList) >>> parallel testBranches >>> } >>> >>> runTests "test1 test2" >>> >>> The output of this job (it can't be aborted): >>> >>> Running: Execute sub-workflows in parallel : Start >>> [test1] Running: Parallel branch: test1 >>> Aborted by anonymous <http://krussell:8080/user/anonymous> >>> Aborted by anonymous <http://krussell:8080/user/anonymous> >>> Aborted by anonymous <http://krussell:8080/user/anonymous> >>> >>> >>> I've also tried replacing the "parallel testBranches" with a simple loop >>> to execute the closures in the map manually (my understanding is that >>> map.each is broken atm so I used a loop instead): >>> >>> for ( test in testBranches) { >>> echo "executing ${test.key}" >>> test.value() >>> } >>> >>> And the output is as follows (note that it prints test2 instead of test1 >>> from inside the closure): >>> >>> Started by user anonymous <http://krussell:8080/user/null> >>> Running: Print Message >>> executing test1 >>> Running: Print Message >>> test2 >>> Running: Print Message >>> executing test2 >>> Running: Print Message >>> test2 >>> Running: End of Workflow >>> Finished: SUCCESS >>> >>> >>> I'm not a Groovy expert (I just started learning for the workflow >>> plugin), so I might be doing something dumb here. I'm trying to read the >>> groovy docs to figure out if the way I'm generating the closure is an >>> issue. But overall the state the system gets into seems to be pretty bad >>> (can't abort the job). So I wanted to send out an email here just in case >>> this is a problem with the Jenkins groovy core. >>> >>> Thanks again for the constant improvements to the workflow plugin. >>> >>> Regards, >>> Andres >>> >>> >>> -- You received this message because you are subscribed to the Google Groups "Jenkins Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/f301bd0b-d892-4a59-9174-fbd45fe598d7%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
