[
https://issues.apache.org/jira/browse/HBASE-9861?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13809309#comment-13809309
]
stack commented on HBASE-9861:
------------------------------
Makes sense.
Could we get stuck here if new location is bad? Or will the new location be
the old location next time through?
+ if (oldLocation == null ||
!source.getServerName().equals(oldLocation.getServerName())) {
+ // There is no such location in the cache (it's been removed already)
or
+ // the cache has already been refreshed with a different location. =>
nothing to do
This is great
+ if (cause instanceof RegionTooBusyException || cause instanceof
RegionOpeningException) {
+ // We know that the region is still on this region server
+ return;
+ }
Fix this comment:
+ * Look for an excpetion we know in the remove exception:
findException looks to duplicate a bunch of exception parsing that we do in
multiple places. We should do a walk through one of these days to clean it all
up.
.... oh, I take it back. You are doing cleanup. How'd we have that dupe of
exception parsing in RegionMoved and RegionOpening?
+1 caveat the little concern up top.
Nice [~nkeywal]
> Location does not have to be refreshed on regionTooBusy
> -------------------------------------------------------
>
> Key: HBASE-9861
> URL: https://issues.apache.org/jira/browse/HBASE-9861
> Project: HBase
> Issue Type: Bug
> Components: Client
> Affects Versions: 0.98.0, 0.96.0
> Reporter: Nicolas Liochon
> Assignee: Nicolas Liochon
> Priority: Minor
> Fix For: 0.96.1
>
> Attachments: 9861.v1.patch
>
>
> Minor improvement. There is already some code to manage the exception, so
> I've change it to share between the classes.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
