[
https://issues.apache.org/jira/browse/KUDU-1460?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15310065#comment-15310065
]
Todd Lipcon commented on KUDU-1460:
-----------------------------------
This was fixed in master by
6fff1b2a2e0927e738bea5216352d14f73ac182e. Just opened a cherry-pick to try to
get it into 0.9.0 as well.
> StartRemoteBootstrap trace can be many MB large
> -----------------------------------------------
>
> Key: KUDU-1460
> URL: https://issues.apache.org/jira/browse/KUDU-1460
> Project: Kudu
> Issue Type: Bug
> Components: tserver
> Affects Versions: 0.8.0
> Reporter: Todd Lipcon
> Assignee: Todd Lipcon
>
> Currently, StartRemoteBootstrap runs the entire bootstrap process inside the
> RPC handler thread. This includes replaying all the operations from the log,
> which generate trace statements. The RPC trace is then many MB large, and it
> gets sampled for rpcz, holdnig memory and making /rpcz slow to load.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
