cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Joel Knighton (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (CASSANDRA-11168) Hint Metrics are updated even if hinted_hand-offs=false
Date Thu, 10 Mar 2016 20:19:40 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-11168?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15189764#comment-15189764
] 

Joel Knighton edited comment on CASSANDRA-11168 at 3/10/16 8:19 PM:
--------------------------------------------------------------------

The 3.0 patch is pretty close - it just needs a few tweaks.

# It looks like the boolean condition around hintedHandoffEnabled() is inverted - if we're
moving all the following code inside the if, it should only be run when hinted handoff is
enabled, not the opposite.
# It looks like we still increment the metric when we decide not to hint because hints are
disabled for a specific datacenter. We shouldn't do this.

The version from your comment above doesn't have these problems; if you want me to make a
patch out of that, I can.

EDIT: Both patches (2.2 and revised 3.0) look good. 3.0 merges cleanly through 3.5 and trunk.
CI here:
||branch||testall||dtest||
|[11168-2.2|https://github.com/jkni/cassandra/tree/11168-2.2]|[testall|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-11168-2.2-testall]|[dtest|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-11168-2.2-dtest]|
|[11168-3.0|https://github.com/jkni/cassandra/tree/11168-3.0]|[testall|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-11168-3.0-testall]|[dtest|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-11168-3.0-dtest]|



was (Author: jkni):
The 3.0 patch is pretty close - it just needs a few tweaks.

# It looks like the boolean condition around hintedHandoffEnabled() is inverted - if we're
moving all the following code inside the if, it should only be run when hinted handoff is
enabled, not the opposite.
# It looks like we still increment the metric when we decide not to hint because hints are
disabled for a specific datacenter. We shouldn't do this.

The version from your comment above doesn't have these problems; if you want me to make a
patch out of that, I can.

EDIT: The 2.2 patch looks good. I've pushed it to CI here:
||branch||testall||dtest||
|[11168-2.2|https://github.com/jkni/cassandra/tree/11168-2.2]|[testall|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-11168-2.2-testall]|[dtest|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-11168-2.2-dtest]|


> Hint Metrics are updated even if hinted_hand-offs=false
> -------------------------------------------------------
>
>                 Key: CASSANDRA-11168
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-11168
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Anubhav Kale
>            Assignee: Anubhav Kale
>            Priority: Minor
>         Attachments: 0001-Hinted-Handoff-Fix.patch, 0001-Hinted-Handoff-fix-2_2.patch,
0001-Hinted-handoff-metrics.patch, 0001-Hinted-handoffs-fix.patch
>
>
> In our PROD logs, we noticed a lot of hint metrics even though we have disabled hinted
handoffs.
> The reason is StorageProxy.ShouldHint has an inverted if condition. 
> We should also wrap the if (hintWindowExpired) block in if (DatabaseDescriptor.hintedHandoffEnabled()).
> The fix is easy, and I can provide a patch.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message