Let's back to verify the release and vote. +1 (binding)
> source - verify file (signature, MD5, SHA) -- source, tar.gz : OK -- source, zip : OK - extract file -- source, tar.gz : OK -- source, zip : OK - diff-ing extracted files between tar.gz and zip : OK - build source with JDK 7 -- source, tar.gz : integration-test failed, others are OK - build source dist -- source, tar.gz : OK - build binary dist -- source, tar.gz : OK > binary - verify file (signature, MD5, SHA) -- binary, tar.gz : OK -- binary, zip : OK - extract file -- binary, tar.gz : OK -- binary, zip : OK - diff-ing extracted files between tar.gz and zip : OK - launch daemons : OK - run RollingTopWords (local) : OK - run RollingTopWords (remote) : OK - activate / deactivate / rebalance / kill : OK - logviewer (worker dir, daemon dir) : OK - change log level : OK - thread dump, heap dump, restart worker : OK - log search : OK I don't see odd numbers while testing, but I don't have stage/production level of cluster/use case, hence someone might be able to see the behavior what Alexandre encountered. Thanks, Jungtaek Lim (HeartSaVioR) 2018년 1월 24일 (수) 오전 10:19, Jungtaek Lim <kabh...@gmail.com>님이 작성: > Alexandre, > > Please file an issue with screenshot and reproducible step (if only > possible). It would be very appreciated if you could spend time to dive > into the codebase and find the cause, and fix and submit a patch (only when > you could get it). > Open source community can't live without contributors. I think reporting > issue itself is great contribution, but I feel we don't have enough code > contributors who could help driving the community. > > Thanks, > Jungtaek Lim (HeartSaVioR) > > 2018년 1월 24일 (수) 오전 9:57, P. Taylor Goetz <ptgo...@gmail.com>님이 작성: > >> Yes, that’s the same error I got, and I think we both just shaved the >> same yak. ;) >> >> I imagine infra is enforcing TLS > 1.0 now. >> >> -Taylor >> >> > On Jan 23, 2018, at 7:46 PM, Jungtaek Lim <kabh...@gmail.com> wrote: >> > >> > Stig, the script doesn't also work for me, but that's not because of >> script >> > or jira module error. >> > I've encountered TLSV1_ALERT_PROTOCOL_VERSION error and my python2.7 is >> > unfortunately coupled with OpenSSL 0.9.8zh which doesn't support >> TLSv1.2. >> > My python3.6 is coupled with OpenSSL 1.0.2l but the script is not >> > compatible with python 3. Maybe I need to modify the script to be >> > compatible with python3.6. >> > >> > cc. to Taylor, assuming that we are getting same error. >> > >> > - Jungtaek Lim (HeartSaVioR) >> > >> > 2018년 1월 24일 (수) 오전 8:21, Stig Rohde Døssing <stigdoess...@gmail.com>님이 >> 작성: >> > >> >> Taylor, >> >> >> >> The release notes script appears to work fine for me. There are a >> couple of >> >> issues with fix version 1.2.0 that are not resolved, which we should >> fix. >> >> Note that 2710 is the release 1.2.0 epic, we might want to not mark >> that >> >> with a fix version so it isn't included in the release notes. >> >> >> >> dev-tools/release_notes.py 1.2.0 >> >> The release is not completed since unresolved issues or improperly >> resolved >> >> issues were found still tagged with this release as the fix version: >> >> Unresolved issue: STORM-2904 None >> >> https://issues.apache.org/jira/browse/STORM-2904 >> >> Unresolved issue: STORM-2710 None >> >> https://issues.apache.org/jira/browse/STORM-2710 >> >> Unresolved issue: STORM-2153 None >> >> https://issues.apache.org/jira/browse/STORM-2153 >> >> >> >> If I ignore the unresolved issues check, I get the expected release >> notes >> >> >> >> dev-tools/release_notes.py 1.2.0 > release-1.2.0.html produces >> >> https://pste.eu/p/ZvbF.html >> >> >> >> 2018-01-24 0:09 GMT+01:00 Alexandre Vermeerbergen < >> >> avermeerber...@gmail.com> >> >> : >> >> >> >>> Hello, >> >>> >> >>> I'm afraid I my vote in 1.2.0 RC1 is a -1: >> >>> >> >>> Indeed metrics displayed in Storm UI from 1.2.0 RC1 are obviously >> wrong. >> >>> >> >>> See for example attached picture showing "Assigned Mem (MB)" for one >> >>> of our topologies: >> >>> - On the left hand side we have Storm 1.1.0 showing 2112 MB on each >> >>> host, which sounds "normal" to us (in line with what we had with >> >>> previous Storm 1.0.3 version) >> >>> - On the right hand side we have Storm 1.2.0 RC1 showing 65 MB on >> >>> each host, which sound completely wrong ! >> >>> >> >>> And I have similar concerns on the statistics on bolts, for example on >> >>> a bolt of our topology in charge of writing logs into HBase, we have: >> >>> >> >>> With Storm 1.1.0, capacity (last 10 min): 0.090 ; Execute Latency >> (ms): >> >>> 0.029 >> >>> With Storm 1.2.0, capacity (last 10 min): 438.956 ; Execute Latency >> >>> (ms): 197.840 >> >>> >> >>> Am I the only one to find weird numbers in Storm UI 1.2.0 ? >> >>> >> >>> Best regards, >> >>> Alexandre Vermeerbergen >> >>> >> >> >> >
