[
https://issues.apache.org/jira/browse/HBASE-16664?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15567630#comment-15567630
]
stack commented on HBASE-16664:
-------------------------------
Operation timeout has always been a murky story full of surprises seemingly
going out of its way to keep the hapless operator baffled. The descriptions in
Table on how the operation timeout works (or will work) promises to clean up
this mess.
Operation timeout should apply whether a fat batch or a small Get.
Above it is suggested that if you need a short timeout for small Gets but a
long one for big batches of MultiGets or Puts, then you could use two Tables
differently configured. That'd work It might frustrate or surprise the operator
some but it is 'understandable'.
For MultiPuts, could we encourage folks to use the BufferedMutator if doing
large Puts? We already do I suppose. Could we add an operation timeout on the
BufferedMutator, one that would override the Table's operation timeout? Would
that help?
That'd leave multiget. AP does this in parallel? Maybe then it could live
within the operation timeout for the table as long as the multiget was many
small ones rather than a fat one?
Would allowing the setting of a timeout on each Get/Mutate object be crazy?
This timeout would override the Table setting. Would 'work' for the single
object case; e.g. a Get or Put. Harder to reason about for the Lists and
batches. Could just get messy fast and be hard to reason about.
> Timeout logic in AsyncProcess is broken
> ---------------------------------------
>
> Key: HBASE-16664
> URL: https://issues.apache.org/jira/browse/HBASE-16664
> Project: HBase
> Issue Type: Bug
> Reporter: Phil Yang
> Assignee: Phil Yang
> Attachments: 1.patch, HBASE-16664-branch-1-v1.patch,
> HBASE-16664-branch-1-v1.patch, HBASE-16664-branch-1.1-v1.patch,
> HBASE-16664-branch-1.2-v1.patch, HBASE-16664-branch-1.3-v1.patch,
> HBASE-16664-v1.patch, HBASE-16664-v2.patch, HBASE-16664-v3.patch,
> HBASE-16664-v4.patch, HBASE-16664-v5.patch, testhcm.patch
>
>
> Have not checked the root cause, but I think timeout of all operations in
> AsyncProcess is broken
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
