Just to (re)set tone, which may be a problem... this is email... Thank you for trying out the RC Lars. I am going to start a DISCUSS soon on proper etiquette for VOTE threads. In other recent votes I've seen RMs (including myself) doing triage of test environment issues in discussion mixed with votes, which is a practice I'd like to have us stop, as I think it is on the reporter to file JIRAs and do at least a first pass of triage before suggesting there are actual problems with a RC, so I'll get that discussion going.
On Mon, Apr 30, 2018 at 9:09 AM, Andrew Purtell <[email protected]> wrote: > Neither of these seem like real problems. > > LocalHBase works great for tests; they all pass. When I set master port > (and others) in hbase-site the port assignments take effect. Those JIRAs > you mention have long threads, many addendums, and are a can of worms. My > position is I am just a humble RM and you lot need to figure out what you > want, if there is still a lack of consensus, but my VOTE thread is not the > place to do it. Everyone: Please don't raise this issue in a RC VOTE again. > > Please file a JIRA for the LTT thing. Since it seems at best a minor issue > with a test tool I don't see how it is relevant for the RC vote either. > > > On Mon, Apr 30, 2018 at 5:01 AM, Lars Francke <[email protected]> > wrote: > >> I did run this on my local MacBook with embedded ZooKeeper and had two >> issues: >> >> - Master starts up and uses a random port, seems to be related to >> https://issues.apache.org/jira/browse/HBASE-20027 & >> https://issues.apache.org/jira/browse/HBASE-20224 >> - I did run LTT (for the first time ever so I might just miss something) >> and I would get regular errors like these: >> >> 2018-04-30 13:58:03,325 ERROR [HBaseReaderThread_0] >> util.MultiThreadedAction: Error checking data for key >> [c9f0f895fb98ab9159f51fd0297e236d-8], column family [test_cf], column >> [1], >> mutation [null]; value of length 1545 >> 2018-04-30 13:58:03,327 INFO [HBaseReaderThread_1] >> util.MultiThreadedAction: >> xD8\x80\x9C\x81\xF0\x1F\xD5HO\x88Q3\x06{\x91\xFA.\x0F\x0C^\xF5 >> \x84.\xC9\x95G\xB2"\x01p\x8EX\xF7\x17V >> \xD7bX\x19w-\xAC\x83oyH-]\xE7\x98V\xA4\x1D\xF1 >> \x06\x0Am\xA6\xF1\x94\x04\x83\x9D;\x83\x15~\xC7\x9D\xFA\xDAB >> S4\x10\xBBL\xC5n5\xC6ZK"N\x9D\x10\xE0\xE9\xFB#\xE8\x19\x0FD\ >> xC8R\x10q\x98XqN\xC8|\xFC\x04B\x8D\x12\xF5\x7F\xFE\xE7\ >> xAC\xBBcJ\x039\xE6Y\x07e)X\x9F|\xFCJ"]h\xFE\xF3\xFE\x84\ >> x14(\x0C\x1F[\xC8\xC8\x8Fm\xB4\x924\xA3\xF22?\xB2\x83\ >> x88\xBF\xF1\xC5m\x0AV\x96=\x899\xEF~m\x09\x08\xBCZ+`\xEA~ >> \xF9!\xE6p\xEC4/\xC7\xC1n\x8B\xF5.\xB4\xEF\xBF\x9E\xE7\x00\ >> x08\xD6\xEB\x93'\x12h\xFB\xA0\xBB\x5C\x88\xA4\xCB\x07\xF1\ >> xDB\x8D\xE3v\x89\xB8\x16m\x093\xCC\x0DE\xC7\xC3[\x87Q\x94\ >> xA9\x1D\x1C\xBE\xADX\xC4\xFE\x8FG\x88O\xA1\x87\xE3`\xEF-4\ >> xBB\xCD\xFC\x87P\x93\x98\x00\x00\x00\x00\x13\x188f, >> 9bf31c7ff062936a96d3c8bd1f8f2ff3-15/test_cf:increment/152508 >> 9483155/Put/vlen=8/seqid=0/\x00\x00\x00\x00\x14k\x85\xF6, >> 9bf31c7ff062936a96d3c8bd1f8f2ff3-15/test_cf:mutate_info/1525 >> 089483159/Put/vlen=4/seqid=0/#8:0} >> Stale false >> 2018-04-30 13:58:03,327 INFO [HBaseReaderThread_1] >> util.MultiThreadedAction: LOCATION >> region=ltttest4,99999999,1525089476605.30bdabaf0a6c164099c4182b199d55b4., >> hostname=lars-macbook.fritz.box,63296,1525089465510, seqNum=5 >> 2018-04-30 13:58:03,327 ERROR [HBaseReaderThread_1] >> util.MultiThreadedReader: At the time of failure, writer wrote 4 keys >> >> AFAIR this only happens when I run LTT for a second time on the same >> table. >> I'm not sure if it needs to be an empty/non-existent table. >> >> Example command: bin/hbase ltt -batchupdate -bloom ROW >> -data_block_encoding >> FAST_DIFF -num_keys 1000000 -tn ltttest4 -read 100:2 -update 20:2 -write >> 10:2048:2 >> >> I won't vote as I don't know enough about LTT and only had an hour on a >> flight with no internet to do much research. >> >> Cheers, >> Lars >> >> On Mon, Apr 30, 2018 at 12:06 PM Ted Yu <[email protected]> wrote: >> >> > +1 >> > Checked signatures Ran test suite with Java 8 - passedPerformed RAT >> check >> > - passed >> > -------- Original message --------From: Ashish Singhi < >> > [email protected]> Date: 4/29/18 9:45 PM (GMT-08:00) To: >> > [email protected] Subject: Re: [VOTE] The first HBase 1.4.4 release >> > candidate (RC0) is available >> > +1 >> > >> > Downloaded the binary and checked the following, >> > >> > - Checked LICENSE and NOTICE files, looks ok >> > - Apache RAT check, looks good >> > >> > Verified the following in a non-secure setup, >> > >> > - Exercised basic shell commands >> > - Ran LTT with 1M row, checked read and write operations >> > - Poked around webUI >> > - Started Rest server and executed some commands >> > - Checked logs all looks like fine >> > >> > Regards, >> > Ashish >> > >> > On Wed, Apr 25, 2018 at 8:48 AM, Andrew Purtell <[email protected]> >> > wrote: >> > >> > > The first HBase 1.4.4 release candidate (RC0) is available for >> download >> > at >> > > https://dist.apache.org/repos/dist/dev/hbase/hbase-1.4.4RC0/ and >> Maven >> > > artifacts are available in the temporary repository >> > > https://repository.apache.org/content/repositories/orgapache >> hbase-1214/ >> > . >> > > >> > > The git tag corresponding to the candidate is '1.4.4RC0' (fe146eb48c). >> > > >> > > A detailed source and binary compatibility report for this release is >> > > available for your review at >> > > https://dist.apache.org/repos/dist/dev/hbase/hbase-1.4.4RC0/ >> > > compat-check-report.html >> > > . >> > > >> > > A list of the 21 issues resolved in this release can be found at >> > > https://s.apache.org/E6GG . >> > > >> > > Please try out the candidate and vote +1/0/-1. >> > > >> > > This vote will be open for at least 72 hours. Unless objection I will >> try >> > > to close it Monday April 30, 2018 if we have sufficient votes. >> > > >> > > Prior to making this announcement I made the following preflight >> checks: >> > > >> > > RAT check passes (7u80) >> > > Unit test suite passes (7u80) >> > > LTT load 1M rows with 100% verification and 20% updates (8u152) >> > > ITBLL Loop 1 1B rows with serverKilling monkey (8u152) >> > > >> > > -- >> > > Best regards, >> > > Andrew >> > > >> > >> > > > > -- > Best regards, > Andrew > > Words like orphans lost among the crosstalk, meaning torn from truth's > decrepit hands > - A23, Crosstalk > -- Best regards, Andrew Words like orphans lost among the crosstalk, meaning torn from truth's decrepit hands - A23, Crosstalk
