Hi all We are also considering how to improve MTTR these days and have a plan. I just notice this thread, hoping I am not late :)
My original thought is simple: Why we must put entries to MemStore of new RS? We can read/write WAL entries only once in failover, from log entries to HFiles, which is similar with Enis's thought. My idea is to flush them to HFiles in their region's directory at the end of splitting, each HFile for each region. And then assign the region to a RS. For the new RS it doesn't replay any logs, it can open the region directly. The difference from the new RS and the crashed RS is the new RS has one or some small HFiles, and its MemStore is empty. If we have too many small HFiles we can submit a minor compaction task but when we compact the region is already online Although this thought need writing many small HFiles rather than one big HFile for each log file, now for ReferenceFile we also write a small file containing region and split key to region's directory. Duo Zhang told me this is because HDFS doesn't support hard link. So if we write them to many HFiles the number of files in HDFS will not be increased. And if we write entries to many HFiles, we can use current logic for all steps which is simple to implement. For example, we can use a MemStore to save entries for one region, and use flush logic to write it to the directory. We don't need new HFile format and we can still check if we have split logs so we have compatibility with old version, even if the RSs in cluster have different split logic, we can still recover all regions. Thanks, Phil 2016-10-22 6:24 GMT+08:00 Enis Söztutar <enis....@gmail.com>: > A bit late, but let me give my perspective. This can also be moved to jira > or dev@ I think. > > DLR was a nice and had pretty good gains for MTTR. However, dealing with > the sequence ids, onlining regions etc and the replay paths proved to be > too difficult in practice. I think the way forward would be to not bring > DLR back, but actually fix long standing log split problems. > > The main gains in DLR is that we do not create lots and lots of tiny files, > but instead rely on the regular region flushes, to flush bigger files. This > also helps with handling requests coming from different log files etc. The > only gain that I can think of that you get with DLR, but not with log split > is the online enabling of writes while the recovery is going on. However, I > think it is not worth having DLR just for this feature. > > Now, what are the problems with Log Split you ask. The problems are > - we create a lot of tiny files > - these tiny files are replayed sequentially when the region is assigned > - The region has to replay and flush all data sequentially coming from > all these tiny files. > > In terms of IO, we pay the cost of reading original WAL files, and writing > this same amount of data into many small files where the NN overhead is > huge. Then for every region, we do serially sort the data by re-reading the > tiny WAL files (recovered edits) and sorting them in memory and flushing > the data. Which means we do 2 times the reads and writes that we should do > otherwise. > > The way to solve our log split bottlenecks is re-reading the big table > paper and implement the WAL recovery as described there. > - Implement an HFile format that can contain data from multiple regions. > Something like a concatinated HFile format where each region has its own > section, with its own sequence id, etc. > - Implement links to these files where a link can refer to this data. This > is very similar to our ReferenceFile concept. > - In each log splitter task, instead of generating tiny WAL files that are > recovered edits, we instead buffer up in memory, and do a sort (this is the > same sort of inserting into the memstore) per region. A WAL is ~100 MB on > average, so should not be a big problem to buffer up this. At the end of > the WAL split task, write an hfile containing data from all the regions as > described above. Also do a multi NN request to create links in regions to > refer to these files (Not sure whether NN has a batch RPC call or not). > > The reason this will be on-par or better than DLR is that, we are only > doing 1 read and 1 write, and the sort is parallelized. The region opening > does not have to block on replaying anything or waiting for flush, because > the data is already sorted and in HFile format. These hfiles will be used > the normal way by adding them to the KVHeaps, etc. When compactions run, we > will be removing the links to these files using the regular mechanisms. > > Enis > > On Tue, Oct 18, 2016 at 6:58 PM, Ted Yu <yuzhih...@gmail.com> wrote: > > > Allan: > > One factor to consider is that the assignment manager in hbase 2.0 would > be > > quite different from those in 0.98 and 1.x branches. > > > > Meaning, you may need to come up with two solutions for a single problem. > > > > FYI > > > > On Tue, Oct 18, 2016 at 6:11 PM, Allan Yang <allan...@163.com> wrote: > > > > > Hi, Ted > > > These issues I mentioned above(HBASE-13567, HBASE-12743, HBASE-13535, > > > HBASE-14729) are ALL reproduced in our HBase1.x test environment. > Fixing > > > them is exactly what I'm going to do. I haven't found the root cause > yet, > > > but I will update if I find solutions. > > > what I afraid is that, there are other issues I don't know yet. So if > > you > > > or other guys know other issues related to DLR, please let me know > > > > > > > > > Regards > > > Allan Yang > > > > > > > > > > > > > > > > > > > > > > > > At 2016-10-19 00:19:06, "Ted Yu" <yuzhih...@gmail.com> wrote: > > > >Allan: > > > >I wonder how you deal with open issues such as HBASE-13535. > > > >From your description, it seems your team fixed more DLR issues. > > > > > > > >Cheers > > > > > > > >On Mon, Oct 17, 2016 at 11:37 PM, allanwin <allan...@163.com> wrote: > > > > > > > >> > > > >> > > > >> > > > >> Here is the thing. We have backported DLR(HBASE-7006) to our 0.94 > > > >> clusters in production environment(of course a lot of bugs are > fixed > > > and > > > >> it is working well). It is was proven to be a huge gain. When a > large > > > >> cluster crash down, the MTTR improved from several hours to less > than > > a > > > >> hour. Now, we want to move on to HBase1.x, and still we want DLR. > This > > > >> time, we don't want to backport the 'backported' DLR to HBase1.x, > but > > it > > > >> seems like that the community have determined to remove DLR... > > > >> > > > >> > > > >> The DLR feature is proven useful in our production environment, so I > > > think > > > >> I will try to fix its issues in branch-1.x > > > >> > > > >> > > > >> > > > >> > > > >> > > > >> > > > >> At 2016-10-18 13:47:17, "Anoop John" <anoop.hb...@gmail.com> wrote: > > > >> >Agree with ur observation.. But DLR feature we wanted to get > > removed.. > > > >> >Because it is known to have issues.. Or else we need major work to > > > >> >correct all these issues. > > > >> > > > > >> >-Anoop- > > > >> > > > > >> >On Tue, Oct 18, 2016 at 7:41 AM, Ted Yu <yuzhih...@gmail.com> > wrote: > > > >> >> If you have a cluster, I suggest you turn on DLR and observe the > > > effect > > > >> >> where fewer than half the region servers are up after the crash. > > > >> >> You would have first hand experience that way. > > > >> >> > > > >> >> On Mon, Oct 17, 2016 at 6:33 PM, allanwin <allan...@163.com> > > wrote: > > > >> >> > > > >> >>> > > > >> >>> > > > >> >>> > > > >> >>> Yes, region replica is a good way to improve MTTR. Specially if > > one > > > or > > > >> two > > > >> >>> servers are down, region replica can improve data availability. > > But > > > >> for big > > > >> >>> disaster like 1/3 or 1/2 region servers shutdown, I think DLR > > still > > > >> useful > > > >> >>> to bring regions online more quickly and with less IO usage. > > > >> >>> > > > >> >>> > > > >> >>> Regards > > > >> >>> Allan Yang > > > >> >>> > > > >> >>> > > > >> >>> > > > >> >>> > > > >> >>> > > > >> >>> > > > >> >>> At 2016-10-17 21:01:16, "Ted Yu" <yuzhih...@gmail.com> wrote: > > > >> >>> >Here was the thread discussing DLR: > > > >> >>> > > > > >> >>> >http://search-hadoop.com/m/YGbbOxBK2n4ES12&subj=Re+ > > > >> >>> DISCUSS+retiring+current+DLR+code > > > >> >>> > > > > >> >>> >> On Oct 17, 2016, at 4:15 AM, allanwin <allan...@163.com> > > wrote: > > > >> >>> >> > > > >> >>> >> Hi, All > > > >> >>> >> DLR can improve MTTR dramatically, but since it have many > bugs > > > like > > > >> >>> HBASE-13567, HBASE-12743, HBASE-13535, HBASE-14729(any more > > I'don't > > > >> know?), > > > >> >>> it was proved unreliable, and has been deprecated almost in all > > > >> branches > > > >> >>> now. > > > >> >>> >> > > > >> >>> >> > > > >> >>> >> My question is, is there any other way other than DLR to > > improve > > > >> MTTR? > > > >> >>> 'Cause If a big cluster crashes, It takes a long time to bring > > > regions > > > >> >>> online, not to mention it will create huge pressure on the IOs. > > > >> >>> >> > > > >> >>> >> > > > >> >>> >> To tell the truth, I still want DLR back, if the community > > don't > > > >> have > > > >> >>> any plan to bring back DLR, I may want to figure out the > problems > > in > > > >> DLR > > > >> >>> and make it working and reliable, Any suggests for that? > > > >> >>> >> > > > >> >>> >> > > > >> >>> >> sincerely > > > >> >>> >> Allan Yang > > > >> >>> > > > >> > > > > > >
