[
https://issues.apache.org/jira/browse/SOLR-7121?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14324993#comment-14324993
]
Shawn Heisey commented on SOLR-7121:
------------------------------------
I see hard-coded comparison values in the new code. If the threshold values do
not come from configuration with documented defaults, I think it can be a
serious problem.
I will also say that all time measurements should be using nanoTime, not
currentTimeMillis. Trust me when I say that the discussion has been done to
death on this topic, and nanoTime is what you'll find everybody switching to,
because currentTimeMillis is not monotonic.
http://stackoverflow.com/a/2979239
In general I like the ideas expressed here, though I haven't looked into how
things are calculated, other than noticing currentTimeMillis, so I don't know
if the approach is good. I also haven't determined whether it is on or off by
default - it should be off, so upgrading users are not surprised by completely
new behavior.
> Solr nodes should go down based on configurable thresholds and not rely on
> resource exhaustion
> ----------------------------------------------------------------------------------------------
>
> Key: SOLR-7121
> URL: https://issues.apache.org/jira/browse/SOLR-7121
> Project: Solr
> Issue Type: New Feature
> Reporter: Sachin Goyal
> Attachments: SOLR-7121.patch
>
>
> Currently, there is no way to control when a Solr node goes down.
> If the server is having high GC pauses or too many threads or is just getting
> too many queries due to some bad load-balancer, the cores in the machine keep
> on serving unless they exhaust the machine's resources and everything comes
> to a stall.
> Such a slow-dying core can affect other cores as well by taking huge time to
> serve their distributed queries.
> There should be a way to specify some threshold values beyond which the
> targeted core can its ill-health and proactively go down to recover.
> When the load improves, the core should come up automatically.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
