Filed HBASE-28278, it is part of the reason but since we close the WAL files asynchronously, it is still possible that we consider a WAL file as 'uncleanly closed' even if later it will be closed cleanly.
Maybe we need to change the code where we update the metrics... Anyway, not a blocker for a release. 张铎(Duo Zhang) <palomino...@gmail.com> 于2023年12月21日周四 22:20写道： > > I've done some digging about the test. I do not think it should be > considered as a test issue but it also does not cause real big > troubles in production. > > The problem is that, we will close WAL writer asynchronously for > speeding up log rolling, and writing trailer is part of the closing > operation, so for replication, we may find that there is already a new > file in the queue but when reading the previous file, we can not read > the trailer yet, thus we consider the WAL file as not closed cleanly. > > Let me check how do we deal with this on master and branch-3. > > Thanks. > > Andrew Purtell <apurt...@apache.org> 于2023年12月21日周四 02:03写道： > > > > Please file or promote an existing JIRA for this test issue (I seem to > > recall some have been opened and closed over the previous months) so we can > > track them. If you set a fix version of 2.5.8 it will be an effective > > reminder. > > > > On Wed, Dec 20, 2023 at 9:59 AM Andrew Purtell <apurt...@apache.org> wrote: > > > > > This test is a known flake. It has been flaky from the time it was > > > introduced. I cannot claim there is definitively no real issue here but I > > > am certain the test should be rewritten. > > > > > > It is certainly no reason to stop this release. > > > > > > On Wed, Dec 20, 2023 at 5:45 AM 张铎(Duo Zhang) <palomino...@gmail.com> > > > wrote: > > > > > >> Run UTs locally, TestBasicWALEntryStreamAsyncFSWAL failed for me several > > >> times. > > >> > > >> Checked flaky dashboard > > >> > > >> > > >> https://nightlies.apache.org/hbase/HBase-Find-Flaky-Tests/branch-2.5/837/output/dashboard.html > > >> > > >> TestBasicWALEntryStreamAsyncFSWAL and TestBasicWALEntryStreamFSHLog > > >> fails nearly 100% of times. > > >> > > >> I think we need to find out whether this is just a test issue, if not > > >> we need to find out the root cause and fix it. > > >> > > >> Thanks. > > >> > > >> Andrew Purtell <apurt...@apache.org> 于2023年12月19日周二 05:38写道： > > >> > > > >> > Please vote on this Apache HBase release candidate, 2.5.7RC0 > > >> > > > >> > The VOTE will remain open for at least 72 hours. > > >> > > > >> > [ ] +1 Release this package as Apache HBase 2.5.7 > > >> > [ ] -1 Do not release this package because ... > > >> > > > >> > The tag to be voted on is 2.5.7RC0: > > >> > > > >> > https://github.com/apache/hbase/tree/2.5.7RC0 > > >> > > > >> > This tag currently points to git reference > > >> > > > >> > 6788f98356dd70b4a7ff766ea7a8298e022e7b95 > > >> > > > >> > The release files, including signatures, digests, as well as CHANGES.md > > >> > and RELEASENOTES.md included in this RC can be found at: > > >> > > > >> > https://dist.apache.org/repos/dist/dev/hbase/2.5.7RC0/ > > >> > > > >> > The API compatibility report comparing this RC to the previous release > > >> > can be found at: > > >> > > > >> > > > >> > > > >> https://dist.apache.org/repos/dist/dev/hbase/2.5.7RC0/api_compare_2.5.6_to_2.5.7RC0.html > > >> > > > >> > As explained in the recent email to dev@ titled '[NOTICE] Recent > > >> changes > > >> > to ZooKeeper dependencies due to CVE issues has a compatibility > > >> > impact', > > >> > due to a recent upgrade of our ZooKeeper dependencies via HBASE-28153 > > >> > there has been an allowable and expected impact to the shaded jars we > > >> > distribute for the convenience of our downstream users. Naturally > > >> because > > >> > the version of ZooKeeper shaded into these jars has changed, so has the > > >> > compatibility of the included classes with earlier releases. > > >> > > > >> > There is also a source only compatibility impact to the Public > > >> > annotated > > >> > Admin API because of HBASE-25549, which has this release note: > > >> > > > >> > New APIs are added to Admin, AsyncAdmin, and hbase shell to allow > > >> > modifying a table without reopening all regions... > > >> > > > >> > Maven artifacts are available in a staging repository at: > > >> > > > >> > > > >> https://repository.apache.org/content/repositories/orgapachehbase-1529/ > > >> > > > >> > Maven artifacts for hadoop3 are available in a staging repository at: > > >> > > > >> > > > >> https://repository.apache.org/content/repositories/orgapachehbase-1530/ > > >> > > > >> > Artifacts were signed with the 0xD5365CCD key which can be found in: > > >> > > > >> > https://downloads.apache.org/hbase/KEYS > > >> > > > >> > To learn more about Apache HBase, please see > > >> > > > >> > http://hbase.apache.org/ > > >> > > > >> > Thanks, > > >> > Your HBase Release Manager > > >> > > > > > > > > > -- > > > Best regards, > > > Andrew > > > > > > Unrest, ignorance distilled, nihilistic imbeciles - > > > It's what we’ve earned > > > Welcome, apocalypse, what’s taken you so long? > > > Bring us the fitting end that we’ve been counting on > > > - A23, Welcome, Apocalypse > > > > > > > > > -- > > Best regards, > > Andrew > > > > Unrest, ignorance distilled, nihilistic imbeciles - > > It's what we’ve earned > > Welcome, apocalypse, what’s taken you so long? > > Bring us the fitting end that we’ve been counting on > > - A23, Welcome, Apocalypse
