You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Vijay (Created) (JIRA)" <ji...@apache.org> on 2012/03/29 00:43:27 UTC
[jira] [Created] (CASSANDRA-4100) Make scrub and cleanup operations
throttled
Make scrub and cleanup operations throttled
-------------------------------------------
Key: CASSANDRA-4100
URL: https://issues.apache.org/jira/browse/CASSANDRA-4100
Project: Cassandra
Issue Type: Bug
Components: Core
Affects Versions: 1.0.8
Reporter: Vijay
Assignee: Vijay
Priority: Minor
Fix For: 1.0.9
Looks like scrub and cleanup operations are not throttled and it will be nice to throttle else we are likely to run into IO issues while running it on live cluster.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (CASSANDRA-4100) Make scrub and cleanup operations
throttled
Posted by "Vijay (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4100?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Vijay updated CASSANDRA-4100:
-----------------------------
Attachment: 0001-CASSANDRA-4100-v3.patch
Alright, v3 simply adds throttle to scrub and cleanup... Simple refactor to move the throttle to compaction controller.
> Make scrub and cleanup operations throttled
> -------------------------------------------
>
> Key: CASSANDRA-4100
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4100
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: Vijay
> Assignee: Vijay
> Priority: Minor
> Labels: compaction
> Fix For: 1.0.10
>
> Attachments: 0001-CASSANDRA-4100-v2.patch, 0001-CASSANDRA-4100-v3.patch, 0001-CASSANDRA-4100.patch
>
>
> Looks like scrub and cleanup operations are not throttled and it will be nice to throttle else we are likely to run into IO issues while running it on live cluster.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (CASSANDRA-4100) Make scrub and cleanup operations
throttled
Posted by "Jonathan Ellis (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4100?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis updated CASSANDRA-4100:
--------------------------------------
Reviewer: yukim
Affects Version/s: (was: 1.0.8)
Labels: compaction (was: )
> Make scrub and cleanup operations throttled
> -------------------------------------------
>
> Key: CASSANDRA-4100
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4100
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: Vijay
> Assignee: Vijay
> Priority: Minor
> Labels: compaction
> Fix For: 1.0.10
>
> Attachments: 0001-CASSANDRA-4100.patch
>
>
> Looks like scrub and cleanup operations are not throttled and it will be nice to throttle else we are likely to run into IO issues while running it on live cluster.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (CASSANDRA-4100) Make scrub and cleanup operations
throttled
Posted by "Vijay (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4100?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Vijay updated CASSANDRA-4100:
-----------------------------
Attachment: 0001-CASSANDRA-4100.patch
Attached patch adds the throttle for cleanup and scrub.
Note: compaction_throughput_mb_per_sec is used for the throttling, not sure if making a seperate property for cleanup/scrub is better....
> Make scrub and cleanup operations throttled
> -------------------------------------------
>
> Key: CASSANDRA-4100
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4100
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Affects Versions: 1.0.8
> Reporter: Vijay
> Assignee: Vijay
> Priority: Minor
> Fix For: 1.0.10
>
> Attachments: 0001-CASSANDRA-4100.patch
>
>
> Looks like scrub and cleanup operations are not throttled and it will be nice to throttle else we are likely to run into IO issues while running it on live cluster.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (CASSANDRA-4100) Make scrub and cleanup
operations throttled
Posted by "Yuki Morishita (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4100?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13251101#comment-13251101 ]
Yuki Morishita commented on CASSANDRA-4100:
-------------------------------------------
v3 looks good to me, but as Sylvain said, I'm +1 to put this to version 1.1.1 instead of 1.0.10.
> Make scrub and cleanup operations throttled
> -------------------------------------------
>
> Key: CASSANDRA-4100
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4100
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: Vijay
> Assignee: Vijay
> Priority: Minor
> Labels: compaction
> Fix For: 1.0.10
>
> Attachments: 0001-CASSANDRA-4100-v2.patch, 0001-CASSANDRA-4100-v3.patch, 0001-CASSANDRA-4100.patch
>
>
> Looks like scrub and cleanup operations are not throttled and it will be nice to throttle else we are likely to run into IO issues while running it on live cluster.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (CASSANDRA-4100) Make scrub and cleanup
operations throttled
Posted by "Yuki Morishita (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4100?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13250188#comment-13250188 ]
Yuki Morishita commented on CASSANDRA-4100:
-------------------------------------------
OK, so static Throttle is fine here since one compaction_throughput_mb_per_sec is used for all compactions. Then, do we need to divide that by number of active compactions? I'm referring the code inside the implementation of ThroughputFunction:
{code}
totalBytesPerMS / Math.max(1, CompactionManager.instance.getActiveCompactions());
{code}
> Make scrub and cleanup operations throttled
> -------------------------------------------
>
> Key: CASSANDRA-4100
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4100
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: Vijay
> Assignee: Vijay
> Priority: Minor
> Labels: compaction
> Fix For: 1.0.10
>
> Attachments: 0001-CASSANDRA-4100.patch
>
>
> Looks like scrub and cleanup operations are not throttled and it will be nice to throttle else we are likely to run into IO issues while running it on live cluster.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (CASSANDRA-4100) Make scrub and cleanup
operations throttled
Posted by "Yuki Morishita (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4100?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13246329#comment-13246329 ]
Yuki Morishita commented on CASSANDRA-4100:
-------------------------------------------
Looks like patch is against trunk, though fix version is marked as 1.0.10.
Anyway, I think Throttle object in CompcationController should be non-static since compactions may run in parallel.
> Make scrub and cleanup operations throttled
> -------------------------------------------
>
> Key: CASSANDRA-4100
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4100
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: Vijay
> Assignee: Vijay
> Priority: Minor
> Labels: compaction
> Fix For: 1.0.10
>
> Attachments: 0001-CASSANDRA-4100.patch
>
>
> Looks like scrub and cleanup operations are not throttled and it will be nice to throttle else we are likely to run into IO issues while running it on live cluster.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (CASSANDRA-4100) Make scrub and cleanup operations
throttled
Posted by "Vijay (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4100?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Vijay updated CASSANDRA-4100:
-----------------------------
Attachment: 0001-CASSANDRA-4100-v2.patch
Fixed. Thanks!
> Make scrub and cleanup operations throttled
> -------------------------------------------
>
> Key: CASSANDRA-4100
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4100
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: Vijay
> Assignee: Vijay
> Priority: Minor
> Labels: compaction
> Fix For: 1.0.10
>
> Attachments: 0001-CASSANDRA-4100-v2.patch, 0001-CASSANDRA-4100.patch
>
>
> Looks like scrub and cleanup operations are not throttled and it will be nice to throttle else we are likely to run into IO issues while running it on live cluster.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (CASSANDRA-4100) Make scrub and cleanup
operations throttled
Posted by "Vijay (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4100?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13246370#comment-13246370 ]
Vijay commented on CASSANDRA-4100:
----------------------------------
>>> I think Throttle object in CompcationController should be non-static since compactions may run in parallel.
Exactly thats why static is better, Parallel compaction is not a problem per say (ParallelCompactionIterable.getReduced() will take care of it), but compaction running one after the other (lot of small compactions).
Let me know if everything else is ok i will rebase to 1.0.10, if needed. Thanks!
> Make scrub and cleanup operations throttled
> -------------------------------------------
>
> Key: CASSANDRA-4100
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4100
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: Vijay
> Assignee: Vijay
> Priority: Minor
> Labels: compaction
> Fix For: 1.0.10
>
> Attachments: 0001-CASSANDRA-4100.patch
>
>
> Looks like scrub and cleanup operations are not throttled and it will be nice to throttle else we are likely to run into IO issues while running it on live cluster.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (CASSANDRA-4100) Make scrub and cleanup
operations throttled
Posted by "Sylvain Lebresne (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4100?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13250486#comment-13250486 ]
Sylvain Lebresne commented on CASSANDRA-4100:
---------------------------------------------
bq. OK, so static Throttle is fine
I'll have to disagree. I'm pretty sure this patch break throttling. If more than one compaction share the Throttle object, they also share the Throttle.timeAtLastDelay field. Which means that as soon as there is more than 1 compaction running at any given time, the interval on which throttling is computing is bogus, and thus throttling will be bogus.
More generally, I'm -1 on changing code that does not have any known problem on the 1.0 branch (and as far as I know, current throttling works well) as 1.0 is getting really stable and we should start being conservative there (but I'd be fine with a patch that just add throttling to scrub and cleanup for 1.0).
> Make scrub and cleanup operations throttled
> -------------------------------------------
>
> Key: CASSANDRA-4100
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4100
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: Vijay
> Assignee: Vijay
> Priority: Minor
> Labels: compaction
> Fix For: 1.0.10
>
> Attachments: 0001-CASSANDRA-4100-v2.patch, 0001-CASSANDRA-4100.patch
>
>
> Looks like scrub and cleanup operations are not throttled and it will be nice to throttle else we are likely to run into IO issues while running it on live cluster.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Issue Comment Edited] (CASSANDRA-4100) Make scrub and
cleanup operations throttled
Posted by "Vijay (Issue Comment Edited) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4100?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13246370#comment-13246370 ]
Vijay edited comment on CASSANDRA-4100 at 4/4/12 3:41 PM:
----------------------------------------------------------
>>> I think Throttle object in CompcationController should be non-static since compactions may run in parallel.
Exactly thats why static is better, Parallel compaction is not a problem per say (ParallelCompactionIterable.getReduced() will take care of it), but compaction running one after the other (lot of small compactions).
Let me know if everything else is ok i will rebase to 1.0.10 and move away from static (I am ok either ways), if needed. Thanks!
was (Author: vijay2win@yahoo.com):
>>> I think Throttle object in CompcationController should be non-static since compactions may run in parallel.
Exactly thats why static is better, Parallel compaction is not a problem per say (ParallelCompactionIterable.getReduced() will take care of it), but compaction running one after the other (lot of small compactions).
Let me know if everything else is ok i will rebase to 1.0.10, if needed. Thanks!
> Make scrub and cleanup operations throttled
> -------------------------------------------
>
> Key: CASSANDRA-4100
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4100
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: Vijay
> Assignee: Vijay
> Priority: Minor
> Labels: compaction
> Fix For: 1.0.10
>
> Attachments: 0001-CASSANDRA-4100.patch
>
>
> Looks like scrub and cleanup operations are not throttled and it will be nice to throttle else we are likely to run into IO issues while running it on live cluster.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (CASSANDRA-4100) Make scrub and cleanup
operations throttled
Posted by "Sylvain Lebresne (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4100?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13250489#comment-13250489 ]
Sylvain Lebresne commented on CASSANDRA-4100:
---------------------------------------------
And to be clear, I'm not saying that throttling cannot be improved (it is true that currently it is too conservative and could throttle a thread even though the total throughput is below the threshold), but I'm saying that 1) this patch does not fix that correctly and 2) in any case this should be another ticket that shouldn't be targeted at 1.0.
> Make scrub and cleanup operations throttled
> -------------------------------------------
>
> Key: CASSANDRA-4100
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4100
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: Vijay
> Assignee: Vijay
> Priority: Minor
> Labels: compaction
> Fix For: 1.0.10
>
> Attachments: 0001-CASSANDRA-4100-v2.patch, 0001-CASSANDRA-4100.patch
>
>
> Looks like scrub and cleanup operations are not throttled and it will be nice to throttle else we are likely to run into IO issues while running it on live cluster.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira