Re: [VOTE FAILED] Release Apache Cassandra 3.0.16

2018-02-14 Thread Michael Kjellman
No worries.. it looks like we didn't update the circleci config in tree for the cassandra-3.0 and cassandra-3.11 branches. We should do that -- i'll take that as an action item on me... for now though you can grab the same one from trunk: CircleCI Configuration: https://github.com/apache/cassan

Re: [VOTE FAILED] Release Apache Cassandra 3.0.16

2018-02-14 Thread Michael Shuler
I apologize for being naive. The test runs do not come from the in-tree circle.yml file and I need to do something different? On 02/14/2018 12:21 PM, Michael Kjellman wrote: > please use the latest circleci config that is in trunk. looking at the config > you used in your run you’re using the old

Re: [VOTE FAILED] Release Apache Cassandra 3.0.16

2018-02-14 Thread Michael Kjellman
please use the latest circleci config that is in trunk. looking at the config you used in your run you’re using the old circleci 1.0 based config. > On Feb 14, 2018, at 10:16 AM, Michael Shuler wrote: > > So far, I have had very unpredictable test runs from CirceCI and ASF > Jenkins. Commit 89

Re: [VOTE FAILED] Release Apache Cassandra 3.0.16

2018-02-14 Thread Michael Shuler
So far, I have had very unpredictable test runs from CirceCI and ASF Jenkins. Commit 890f319 resulted in 4 completely different test failures for me today in CircleCI. https://circleci.com/gh/mshuler/cassandra/150 I trust results from the static ASF slaves even less. `ant test-all -Dtest.name=Co

Re: [VOTE FAILED] Release Apache Cassandra 3.0.16

2018-02-14 Thread Michael Kjellman
the tests are writing something to java.io.tmpdir which for whatever reason isn’t writable. the same exact tests work locally and in the cassandra-test docker image. given everyone can partake on circleci runs and asf jenkins i think we should send out those links and base the vote off those ru

Re: [VOTE FAILED] Release Apache Cassandra 3.0.16

2018-02-14 Thread Michael Shuler
This is an internal Jenkins instance that's not reachable on the internet. What's the permissions issue? The test runs on this internal instance are exactly the way cassci used to run - launch a scratch machine, check out, build, run tests. I'll see if I can repro locally. Michael On 02/14/2018

Re: [VOTE FAILED] Release Apache Cassandra 3.0.16

2018-02-14 Thread Michael Shuler
On 02/14/2018 11:40 AM, Michael Shuler wrote: > Thanks for the feedback on 3.0.16 tentative release. > > Commit 890f319 (current cassandra-3.0 branch HEAD) fails only one test > (in both standard and -compression suites) in CI for me. This test has > failed 20 times in the last 20 runs. Test outpu

Re: [VOTE FAILED] Release Apache Cassandra 3.0.16

2018-02-14 Thread Michael Kjellman
i looked at this a few weeks back. this is asf jenkins right? if so, it’s a permissions issue on the build executors > On Feb 14, 2018, at 9:40 AM, Michael Shuler wrote: > > Thanks for the feedback on 3.0.16 tentative release. > > Commit 890f319 (current cassandra-3.0 branch HEAD) fails only

Re: [VOTE FAILED] Release Apache Cassandra 3.0.16

2018-02-14 Thread Jason Brown
Hi Michael, Which CI environment? iirc, I had several clean runs in circleci with the latest commits (will run again today). Thanks, -Jasoin On Wed, Feb 14, 2018 at 9:40 AM, Michael Shuler wrote: > Thanks for the feedback on 3.0.16 tentative release. > > Commit 890f319 (current cassandra-3.0

[VOTE FAILED] Release Apache Cassandra 3.0.16

2018-02-14 Thread Michael Shuler
Thanks for the feedback on 3.0.16 tentative release. Commit 890f319 (current cassandra-3.0 branch HEAD) fails only one test (in both standard and -compression suites) in CI for me. This test has failed 20 times in the last 20 runs. Test output attached. Do we wish to fix this before the next cut,