cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jonathan Ellis (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-5255) dsnitch severity is not correctly set for compaction info
Date Mon, 18 Feb 2013 15:19:12 GMT


Jonathan Ellis commented on CASSANDRA-5255:

bq. Another possible option: Create a way to get IOStat data into the JVM and measure the
IOWait time but the problem is how do we support MS Windows? not sure if there is a generic
way to do so.

I like that idea a lot better than trying to special-case streaming and compaction.

No idea how to get that info on Windows though.  Guess we could fall back to MBPs compaction/streaming
there.  (Yes, this is not perfect if you mix SSDs + HDDs, but it's a LOT better than total
MB of compaction.)
> dsnitch severity is not correctly set for compaction info
> ---------------------------------------------------------
>                 Key: CASSANDRA-5255
>                 URL:
>             Project: Cassandra
>          Issue Type: Bug
>    Affects Versions: 1.2.0 beta 1
>            Reporter: Brandon Williams
>            Assignee: Vijay
>            Priority: Minor
>             Fix For: 1.2.2
>         Attachments: 0001-CASSANDRA-5255.patch
> We're doing two things wrong in CI.  First, load can change between calls, which can
cause a negative severity even though it meant to subtract whatever it added before.  Second,
we should report based on how much IO we're using, since a 1T throttled to 5MB/s is less impactful
than a 100MB running at full speed.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see:

View raw message