> On Nov. 17, 2016, 11:28 p.m., Joshua Cohen wrote: > > I'm seeing intermittent failures running the e2e tests with this patch > > applied. That's not necessarily alarming, as we've had reports of this on > > master as well, however the behavior in the event of these failures is > > different with this patch applied. See this gist which contains the full > > output: https://gist.github.com/jcohen/ee69208fbca714d3f3437e4039961edf. > > Note the failure about 80% through, followed by some strange compile issues? > > John Sirois wrote: > Can you click on a line number and paste that url? - not seeing it. > > Joshua Cohen wrote: > I think it's just reviewboard counting the period as part of the url? try > this: https://gist.github.com/jcohen/ee69208fbca714d3f3437e4039961edf > > John Sirois wrote: > Ah - my miscommunication. URL worked fine, I thought clicking on a line > number would ammend the url to an anchor that would skip me down to the error > you saw. Moot since the gist is truncated and you have to view it raw to see > the whole thing (as I now realize). > > > So: > ``` > :commons:generateThriftResources UP-TO-DATE > :commons:processResources > :commons:classes > :commons:jar > :compileJava > > /home/vagrant/aurora/src/main/java/org/apache/aurora/scheduler/storage/log/WriteAheadStorage.java:82: > error: cannot find symbol > class WriteAheadStorage extends WriteAheadStorageForwarder implements > ^ > symbol: class WriteAheadStorageForwarder > error: Problem loading existing flags on compile time classpath: > java.io.EOFException: Unexpected end of ZLIB input stream > at > java.util.zip.ZipFile$ZipFileInflaterInputStream.fill(ZipFile.java:417) > at > java.util.zip.InflaterInputStream.read(InflaterInputStream.java:158) > at java.io.FilterInputStream.read(FilterInputStream.java:133) > at sun.nio.cs.StreamDecoder.readBytes(StreamDecoder.java:284) > at sun.nio.cs.StreamDecoder.implRead(StreamDecoder.java:326) > at sun.nio.cs.StreamDecoder.read(StreamDecoder.java:178) > at java.io.InputStreamReader.read(InputStreamReader.java:184) > at com.google.common.io.MultiReader.read(MultiReader.java:55) > at com.google.common.io.LineReader.readLine(LineReader.java:77) > at > com.google.common.io.CharStreams.readLines(CharStreams.java:146) > at > org.apache.aurora.common.args.apt.Configuration.load(Configuration.java:409) > at > org.apache.aurora.common.args.apt.Configuration.load(Configuration.java:325) > at > org.apache.aurora.common.args.apt.CmdLineProcessor$1.get(CmdLineProcessor.java:103) > at > org.apache.aurora.common.args.apt.CmdLineProcessor$1.get(CmdLineProcessor.java:100) > at > com.google.common.base.Suppliers$MemoizingSupplier.get(Suppliers.java:120) > at > org.apache.aurora.common.args.apt.CmdLineProcessor.process(CmdLineProcessor.java:174) > at > com.sun.tools.javac.processing.JavacProcessingEnvironment.callProcessor(JavacProcessingEnvironment.java:794) > at > com.sun.tools.javac.processing.JavacProcessingEnvironment.discoverAndRunProcs(JavacProcessingEnvironment.java:705) > at > com.sun.tools.javac.processing.JavacProcessingEnvironment.access$1800(JavacProcessingEnvironment.java:91) > at > com.sun.tools.javac.processing.JavacProcessingEnvironment$Round.run(JavacProcessingEnvironment.java:1035) > at > com.sun.tools.javac.processing.JavacProcessingEnvironment.doProcessing(JavacProcessingEnvironment.java:1176) > at > com.sun.tools.javac.main.JavaCompiler.processAnnotations(JavaCompiler.java:1170) > at > com.sun.tools.javac.main.JavaCompiler.compile(JavaCompiler.java:856) > at com.sun.tools.javac.main.Main.compile(Main.java:523) > ``` > > Which has to be unrelated to these changes? > > John Sirois wrote: > Basically the error above is an annotation processor trying to read > classpath resources and encountering a jar on the classpath that is truncated. > > Joshua Cohen wrote: > The odd thing is I'm able to reproduce about 50% of the time with a clean > vagrant image (i.e. vagrant destroy before running the e2e tests). But it > only seems to be failing in this particular way with this patch applied. > > Probably we need a broader ticket to investigate the recent e2e flakiness > though, as regardless of this patch there are issues on master. > > John Sirois wrote: > OK - Since I seem to be blocked on this e2e flakiness, I'll see if I can > find some time in the next few days to file and take a look at the problem.
fwiw, in passing I traced the above e2e error to gradle daemon, when I get some time I'll investigate if there's a fix other than disabling the daemon. Until then, I don't think it should block this change. - Joshua ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/53836/#review156262 ----------------------------------------------------------- On Nov. 17, 2016, 4:46 a.m., John Sirois wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/53836/ > ----------------------------------------------------------- > > (Updated Nov. 17, 2016, 4:46 a.m.) > > > Review request for Aurora, Joshua Cohen, Stephan Erb, and Zameer Manji. > > > Repository: aurora > > > Description > ------- > > This required an upgrade to the latest pants dev release to correct > an issue with the setup.py binary packager we use to generate sdists. > > This is for sanity sake, and, once the TODO in > `build-support/thrift/prepare_binary.sh` is addressed, it will also > allow pants-patch-free addition of new platforms (thinking ARM). > > api/src/main/thrift/org/apache/aurora/gen/BUILD | 3 > +++ > api/src/main/thrift/org/apache/thermos/BUILD | 1 + > build-support/thrift/.gitignore | 1 + > build-support/thrift/AURORA-1727.compiler.cpp.Makefile.in.patch | 42 > +++++++++++++------------------ > build-support/thrift/AURORA-1727.lib.py.setup.py.patch | 45 > +++++++++++++++++++++++++++++++++ > {api/src/main/thrift/org/apache/thermos => build-support/thrift}/BUILD | 18 > ++++--------- > build-support/thrift/Makefile | 2 > +- > build-support/thrift/prepare_binary.sh | 68 > ++++++++++++++++++++++++++++++++++++++++++++++++++ > build-support/thrift/thriftw | 12 > +++++++-- > pants.ini | 7 > +++++- > 10 files changed, 158 insertions(+), 41 deletions(-) > > > Diffs > ----- > > api/src/main/thrift/org/apache/aurora/gen/BUILD > f4004a3ca36078c6d24991db4beb68903a05652c > api/src/main/thrift/org/apache/thermos/BUILD > cb655a2fd35d22f7d7e80c4311742bad763d8614 > build-support/thrift/.gitignore 9a3adb69210ba3dbf3c1f408895561da37e8f4c3 > build-support/thrift/AURORA-1727.compiler.cpp.Makefile.in.patch > b69e3fef137cd73c6f2b73201463a0705ef8082a > build-support/thrift/AURORA-1727.lib.py.setup.py.patch PRE-CREATION > build-support/thrift/BUILD PRE-CREATION > build-support/thrift/Makefile 48b174ad622288d2738a5fa37bbb72385fcc3a27 > build-support/thrift/prepare_binary.sh PRE-CREATION > build-support/thrift/thriftw 50d6dfdeb16ca8bf14aaff7aa826e3d69c5e13f0 > pants.ini cecdb277f327f77b2652f76a30fc8d4ffd9ff1db > > Diff: https://reviews.apache.org/r/53836/diff/ > > > Testing > ------- > > Locally green: > ``` > rm -rf ~/.cache/pants .cache/ > git clean -fdx build-support/thrift > ./pants clean-all > ./build-support/jenkins/build.sh > > vagrant ssh --command "rm -rf ~/.cache/pants && cd aurora && rm -rf .cache/ > && ./pants clean-all" > ./src/test/sh/org/apache/aurora/e2e/test_end_to_end.sh > > ./build-support/python/make-pycharm-virtualenv > ``` > > > Thanks, > > John Sirois > >
