+1 from my side.
I also added a comment to HDFS-14268 explaining the reasons to modify ECBlockGroupStats. We can follow up into possible changes there if needed. On Mon, Jun 10, 2019 at 8:10 AM Arpit Agarwal <aagar...@cloudera.com.invalid> wrote: > I scanned the merge payload for changes to non-RBF code. The changes are > minimal, which is good. > > The only commit that I didn’t understand was: > https://issues.apache.org/jira/browse/HDFS-14268 < > https://issues.apache.org/jira/browse/HDFS-14268> > > The jira description doesn’t make it clear why ECBlockGroupStats is > modified. > > +0 apart from that. > > > > On Jun 1, 2019, at 8:40 PM, Brahma Reddy Battula <bra...@apache.org> > wrote: > > > > Dear Hadoop Developers > > > > I would like to propose RBF Branch (HDFS-13891) merge into trunk. We have > > been working on this feature from last several months. > > This feature work received the contributions from different companies. > All > > of the feature development happened smoothly and collaboratively in > JIRAs. > > > > Kindly do take a look at the branch and raise issues/concerns that need > to > > be addressed before the merge. > > > > *Highlights of HDFS-13891 Branch:* > > ============================= > > > > Adding Security to RBF(1) > > Adding Missing Client API's(2) > > Improvements/Bug Fixing > > Critical - HDFS-13637, HDFS-13834 > > > > *Commits:* > > ======== > > > > No of JIRAs Resolved: 72 > > > > All this commits are in RBF Module. No changes in hdfs/common. > > > > *Tested Cluster:* > > ============= > > > > Most of these changes verified at Uber,Microsoft,Huawei and some other > > compaines. > > > > *Uber*: Most changes are running in production @Uber including the > critical > > security changes, HDFS Clusters are 4000+ nodes with 8 HDFS Routers. > > Zookeeper as a state store to hold delegation tokens were also stress > > tested to hold more than 2 Million tokens. --CR Hota > > > > *MicroSoft*: Most of these changes are currently running in production at > > Microsoft.The security has also been tested in a 500 server cluster with > 4 > > subclsuters. --Inigo Goiri > > > > *Huawei* : Deployed all this changes in 20 node cluster with 3 > > routers.Planning deploy 10K production cluster. > > > > *Contributors:* > > =========== > > > > Many thanks to Akira Ajisaka,Mohammad Arshad,Takanobu Asanuma,Shubham > > Dewan,CR Hota,Fei Hui,Inigo Goiri,Dibyendu Karmakar,Fengna Li,Gang > > Li,Surendra Singh Lihore,Ranith Sardar,Ayush Saxena,He Xiaoqiao,Sherwood > > Zheng,Daryn Sharp,VinayaKumar B,Anu Engineer for invloving discussions > and > > contributing to this. > > > > *Future Tasks:* > > ============ > > > > will cleanup the jira's under this umbrella and contiue to work. > > > > Reference: > > 1) https://issues.apache.org/jira/browse/HDFS-13532 > > 2) https://issues.apache.org/jira/browse/HDFS-13655 > > > > > > > > > > --Brahma Reddy Battula > >
