[
https://issues.apache.org/jira/browse/HBASE-21744?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16749253#comment-16749253
]
Xu Cang commented on HBASE-21744:
---------------------------------
in this while loop :
{code:java}
while (!server.isStopped()) {
{code}
I think we need to have some delay in each iteration. Such as "Thread.sleep()"
Why using nanoSec? It's a bit overkilled for me.
Also, is that possible to write a test case to verify this feature, such as
counting how many times this list gets refreshed within a certain time range?
Thanks.
> timeout for server list refresh calls
> --------------------------------------
>
> Key: HBASE-21744
> URL: https://issues.apache.org/jira/browse/HBASE-21744
> Project: HBase
> Issue Type: Bug
> Affects Versions: 3.0.0
> Reporter: Sergey Shelukhin
> Assignee: Sergey Shelukhin
> Priority: Major
> Attachments: HBASE-21744.patch
>
>
> Not sure why yet, but we are seeing the case when cluster is in overall a bad
> state, where after RS dies and deletes its znode, the notification looks like
> it's lost, so the master doesn't detect the failure. ZK itself appears to be
> healthy and doesn't report anything special.
> After some other change is made to the server list, master rescans the list
> and picks up the stale change. Might make sense to add a config that would
> trigger the refresh if it hasn't happened for a while (e.g. 1 minute).
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
