[
https://issues.apache.org/jira/browse/SOLR-7270?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14370096#comment-14370096
]
Mark Miller commented on SOLR-7270:
-----------------------------------
bq. May be I am misunderstanding this then, wouldn't the stack trace then have
the URL because the exception message has it?
I'm talking about when you dump jstacks or something and just see the current
stack trace for every thread. Or when you use a tool like YourKit. In many
cases, users find a client application or Solr or an integration point stuck
and so we start looking at stack traces to see what is happening. I would love
to be able to following one of these update threads right to the node its
waiting on and look at those stack traces. Sometimes I have logs from around
that point as well and sometimes I don't. But even then, logs take a lot of
time to work through.
bq. That seems reasonable to look into.
I guess part of what also kept me from that is that I didn't really want the
base name to fluctuate before and after hitting the executor - so I tried to
keep the name the same and just have this extra data. Given what we are doing,
I guess that is not the biggest concern if the full name fluctuates though.
> Add URL of remote host in distrib update thread names to improve
> debuggability via stack traces.
> ------------------------------------------------------------------------------------------------
>
> Key: SOLR-7270
> URL: https://issues.apache.org/jira/browse/SOLR-7270
> Project: Solr
> Issue Type: Improvement
> Reporter: Mark Miller
> Assignee: Mark Miller
> Attachments: SOLR-7270.patch, SOLR-7270.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
