[
https://issues.apache.org/jira/browse/HBASE-8665?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13671078#comment-13671078
]
Elliott Clark commented on HBASE-8665:
--------------------------------------
bq.it's property of the store
Strongly disagree. A major compaction for an important store is not nearly as
important as a compaction to get under the blocking storefile count.
Talked with JD and the compaction priority was put in so that compactions
queued to un-block a flush can be given a higher priority. With that in mind I
think the best solution that I've heard so far, with the least code churn, is
to raise the priority on any non major compactions in the queue when we are
thinking about queueing another.
The other harder more invasive solution that we (JD, Jon, and I) talked about
was to put in real scheduling. Something like the linux O(1) scheduler (with
an active and expired set of queues) would work. The idea is pretty well
tested and shouldn't be too hard to code. Though I don't think that we really
need that yet.
> bad compaction priority behavior in queue can cause store to be blocked
> -----------------------------------------------------------------------
>
> Key: HBASE-8665
> URL: https://issues.apache.org/jira/browse/HBASE-8665
> Project: HBase
> Issue Type: Bug
> Reporter: Sergey Shelukhin
> Assignee: Sergey Shelukhin
>
> Note that this can be solved by bumping up the number of compaction threads
> but still it seems like this priority "inversion" should be dealt with.
> There's a store with 1 big file and 3 flushes (1 2 3 4) sitting around and
> minding its own business when it decides to compact. Compaction (2 3 4) is
> created and put in queue, it's low priority, so it doesn't get out of the
> queue for some time - other stores are compacting. Meanwhile more files are
> flushed and at (1 2 3 4 5 6 7) it decides to compact (5 6 7). This compaction
> now has higher priority than the first one. After that if the load is high it
> enters vicious cycle of compacting and compacting files as they arrive, with
> store being blocked on and off, with the (2 3 4) compaction staying in queue
> for up to ~20 minutes (that I've seen).
> I wonder why we do thing thing where we queue compaction and compact
> separately. Perhaps we should take snapshot of all store priorities, then do
> select in order and execute the first compaction we find. This will need
> starvation safeguard too but should probably be better.
> Btw, exploring compaction policy may be more prone to this, as it can select
> files from the middle, not just beginning, which, given the treatment of
> already selected files that was not changed from the old ratio-based one (all
> files with lower seqNums than the ones selected are also ineligible for
> further selection), will make more files ineligible (e.g. imagine with 10
> blocking files, with 8 present (1-8), (6 7 8) being selected and getting
> stuck). Today I see the case that would also apply to old policy, but
> yesterday I saw file distribution something like this: 4,5g, 2,1g, 295,9m,
> 113,3m, 68,0m, 67,8m, 1,1g, 295,1m, 100,4m, unfortunately w/o enough logs to
> figure out how it resulted.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
