Hi Wei-Chi, There is bunch of performance jiras in Hadoop. This one is not marked as blocker and seems to be hard to reproduce. And I see that you are still discussing the problem and the solution with Daryn in the jira. I don't think it should block the release. If this or anything else will prove to be a serious problem we can follow up with a maintenance release containing hotfixes only.
Thanks, --Konst On Fri, Aug 4, 2017 at 9:24 AM, Wei-Chiu Chuang <weic...@cloudera.com> wrote: > Hi, > I'm sorry coming to this vote late. > Daryn mentioned in HDFS-12136 > <https://issues.apache.org/jira/browse/HDFS-12136> that HDFS-11160 > <https://issues.apache.org/jira/browse/HDFS-11160> has a performance > regression at DataNode due to the way it locks dataset lock. > > HDFS-11160 is in Hadoop 2.7.4. Would it be critical enough to warrant a > stop for release? > > I myself can't reproduce the performance regression (assuming it only > occurs under extreme workload). Would Daryn or other Yahoo folks comment? > > On Thu, Aug 3, 2017 at 10:31 PM, Akira Ajisaka <aajis...@apache.org> > wrote: > >> +1 (binding) >> >> - Verified the checksum and the signature of the source tarball >> - Built from source with CentOS 7.2 and OpenJDK 1.8.0_141 >> - Built Hive 2.1.0/2.3.0 and Tez 0.8.5/0.9.0 with Hadoop 2.7.4 artifacts >> - Built single node cluster and ran some Hive on Tez queries successfully >> >> Regards, >> Akira >> >> >> On 2017/08/04 0:25, Kuhu Shukla wrote: >> >>> +1 (non-binding) >>> >>> 1. Verified signatures and digests. >>> 2. Built source. >>> 3. Installed on a pseudo-distributed cluster. >>> 4. Ran sample MR jobs and Tez example jobs like orderedwordcount >>> successfully. >>> >>> Thank you Konstantin and others for this release. >>> >>> Regards, >>> Kuhu >>> >>> >>> >>> On Thursday, August 3, 2017, 7:19:07 AM CDT, Sunil G <sun...@apache.org> >>> wrote: >>> >>> >>> Thanks Konstantin >>> >>> +1 (binding) >>> >>> 1. Build tar ball from source package >>> 2. Ran basic MR jobs and verified UI. >>> 3. Enabled node labels and ran sleep job. Works fine. >>> 4. Verified CLI commands related to node labels and its working fine. >>> 5. RM WorkPreserving restart cases are also verified, and looks fine >>> >>> Thanks >>> Sunil >>> >>> >>> >>> On Sun, Jul 30, 2017 at 4:59 AM Konstantin Shvachko < >>> shv.had...@gmail.com> >>> wrote: >>> >>> Hi everybody, >>>> >>>> Here is the next release of Apache Hadoop 2.7 line. The previous stable >>>> release 2.7.3 was available since 25 August, 2016. >>>> Release 2.7.4 includes 264 issues fixed after release 2.7.3, which are >>>> critical bug fixes and major optimizations. See more details in Release >>>> Note: >>>> http://home.apache.org/~shv/hadoop-2.7.4-RC0/releasenotes.html >>>> >>>> The RC0 is available at: http://home.apache.org/~shv/hadoop-2.7.4-RC0/ >>>> >>>> Please give it a try and vote on this thread. The vote will run for 5 >>>> days >>>> ending 08/04/2017. >>>> >>>> Please note that my up to date public key are available from: >>>> https://dist.apache.org/repos/dist/release/hadoop/common/KEYS >>>> Please don't forget to refresh the page if you've been there recently. >>>> There are other place on Apache sites, which may contain my outdated >>>> key. >>>> >>>> Thanks, >>>> --Konstantin >>>> >>>> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: hdfs-dev-unsubscr...@hadoop.apache.org >> For additional commands, e-mail: hdfs-dev-h...@hadoop.apache.org >> >> > > > -- > A very happy Clouderan >
