You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "daishi (JIRA)" <ji...@apache.org> on 2009/05/09 00:44:45 UTC
[jira] Created: (CASSANDRA-157) make cassandra not allow itself to
run out of memory during sustained inserts
make cassandra not allow itself to run out of memory during sustained inserts
-----------------------------------------------------------------------------
Key: CASSANDRA-157
URL: https://issues.apache.org/jira/browse/CASSANDRA-157
Project: Cassandra
Issue Type: Bug
Reporter: daishi
Tv on IRC pointed out to me that the issue that I've been encountering
is probably point 2. in this roadmap:
http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
I was unable to find any existing issue for this topic, so I'm creating a new one.
Since this issue would block our use of Cassandra I'm happy to look into it,
but if this is a known issue perhaps there's already a plan for addressing it
that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-157) make cassandra not allow itself
to run out of memory during sustained inserts
Posted by "daishi (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12716474#action_12716474 ]
daishi commented on CASSANDRA-157:
----------------------------------
Sorry I haven't had a chance to follow up on this.
If I remember correctly the main unbounded memory growth
that caused OOMs did seem to go away, but that the exposed
properties didn't control the memory usage in a predictable way.
In particular I think the objCount property would schedule
a Memtable.flush(), but ColumnFamilyStore.memtablesPendingFlush
would be actually flushed according to a different schedule.
I haven't read enough of the code to know whether it would
be possible/make sense to allow user-control of latter schedule,
but that was the direction I was thinking in when I had to drop
this for a bit (maybe you can let me know if that makes no
sense or if there's a better way to think about things).
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
> Assignee: daishi
> Attachments: Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-157) make cassandra not allow itself
to run out of memory during sustained inserts
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12708853#action_12708853 ]
Hudson commented on CASSANDRA-157:
----------------------------------
Integrated in Cassandra #73 (See [http://hudson.zones.apache.org/hudson/job/Cassandra/73/])
unregister mbean on flush. patch by daishi; reviewed by jbellis for
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
> Assignee: daishi
> Attachments: Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-157) make cassandra not allow itself
to run out of memory during sustained inserts
Posted by "Brandon Williams (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12735817#action_12735817 ]
Brandon Williams commented on CASSANDRA-157:
--------------------------------------------
+1 The heap usage is much improved for me with this patch.
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: daishi
> Assignee: Jonathan Ellis
> Attachments: 157.patch, Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (CASSANDRA-157) make cassandra not allow itself to
run out of memory during sustained inserts
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis updated CASSANDRA-157:
-------------------------------------
Attachment: 157.patch
Found the (a?) problem.
memtablesPendingFlush used NonBlockingHashSets to store the memtables-to-flush. But NBHS uses a NBHMap under the hood, which when remove() is called, assigns a tombstone value to the key instead of actually removing it. (See http://sourceforge.net/tracker/?func=detail&aid=2828100&group_id=194172&atid=948362.)
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: daishi
> Assignee: Jonathan Ellis
> Attachments: 157.patch, Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-157) make cassandra not allow itself
to run out of memory during sustained inserts
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12716437#action_12716437 ]
Jonathan Ellis commented on CASSANDRA-157:
------------------------------------------
Is this still an issue?
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
> Assignee: daishi
> Attachments: Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Assigned: (CASSANDRA-157) make cassandra not allow itself to
run out of memory during sustained inserts
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis reassigned CASSANDRA-157:
----------------------------------------
Assignee: Jonathan Ellis (was: daishi)
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: daishi
> Assignee: Jonathan Ellis
> Attachments: Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Resolved: (CASSANDRA-157) make cassandra not allow itself to
run out of memory during sustained inserts
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis resolved CASSANDRA-157.
--------------------------------------
Resolution: Fixed
committed
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: daishi
> Assignee: Jonathan Ellis
> Attachments: 157.patch, Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-157) make cassandra not allow itself
to run out of memory during sustained inserts
Posted by "Eric Evans (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12708121#action_12708121 ]
Eric Evans commented on CASSANDRA-157:
--------------------------------------
I think we need a wiki page which details what we've learned about tuning thresholds to stay within memory constraints (and yes, I suppose I am volunteerin :).
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-157) make cassandra not allow itself
to run out of memory during sustained inserts
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12707586#action_12707586 ]
Jonathan Ellis commented on CASSANDRA-157:
------------------------------------------
Ah, I see you've had this discussion with urandom on irc.
Like he said, play with jconsole, it may be your limits are still not low enough or possibly there is a bug in its internal accounting.
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-157) make cassandra not allow itself
to run out of memory during sustained inserts
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12707585#action_12707585 ]
Jonathan Ellis commented on CASSANDRA-157:
------------------------------------------
if you're running out of memory, start by decreasing these settings in conf/storage-conf. svn up first because objectcount used to default to 0.1 (i.e. 100k objects).
<!--
The maximum amount of data to store in memory before flushing to
disk. Note: There is one memtable per column family, and this threshold
is based solely on the amount of data stored, not actual heap memory
usage (there is some overhead in indexing the columns).
-->
<MemtableSizeInMB>32</MemtableSizeInMB>
<!--
The maximum number of columns in millions to store in memory
before flushing to disk. This is also a per-memtable setting.
Use with MemtableSizeInMB to tune memory usage.
-->
<MemtableObjectCountInMillions>0.02</MemtableObjectCountInMillions>
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-157) make cassandra not allow itself
to run out of memory during sustained inserts
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12708533#action_12708533 ]
Jonathan Ellis commented on CASSANDRA-157:
------------------------------------------
great fix! committed.
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
> Attachments: Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-157) make cassandra not allow itself
to run out of memory during sustained inserts
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12716482#action_12716482 ]
Jonathan Ellis commented on CASSANDRA-157:
------------------------------------------
That's strange. Let me know if you can reproduce that behavior still with 0.3 or trunk. It shouldn't be queueing up unless the flusher is busy with a different memtable.
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
> Assignee: daishi
> Attachments: Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-157) make cassandra not allow itself
to run out of memory during sustained inserts
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12716475#action_12716475 ]
Jonathan Ellis commented on CASSANDRA-157:
------------------------------------------
> ColumnFamilyStore.memtablesPendingFlush would be actually flushed according to a different schedule
flushes are processed one at a time, first come first served, to avoid swamping the system with IO.
does that jive with what you saw?
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
> Assignee: daishi
> Attachments: Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Issue Comment Edited: (CASSANDRA-157) make cassandra not
allow itself to run out of memory during sustained inserts
Posted by "Eric Evans (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12708121#action_12708121 ]
Eric Evans edited comment on CASSANDRA-157 at 5/11/09 10:32 AM:
----------------------------------------------------------------
I think we need a wiki page which details what we've learned about tuning thresholds (and yes, I suppose I am volunteerin :).
was (Author: urandom):
I think we need a wiki page which details what we've learned about tuning thresholds to stay within memory constraints (and yes, I suppose I am volunteerin :).
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-157) make cassandra not allow itself
to run out of memory during sustained inserts
Posted by "Eric Evans (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12709943#action_12709943 ]
Eric Evans commented on CASSANDRA-157:
--------------------------------------
http://wiki.apache.org/cassandra/MemtableThresholds
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
> Assignee: daishi
> Attachments: Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (CASSANDRA-157) make cassandra not allow itself to
run out of memory during sustained inserts
Posted by "daishi (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
daishi updated CASSANDRA-157:
-----------------------------
Attachment: Cassandra-157_Unregister_Memtable_MBean.diff
This patch removes the MBean reference to Memtable objects,
reducing the memory pressure under sustained inserts somewhat.
(This helps, but I think there are other remaining issues).
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
> Attachments: Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Assigned: (CASSANDRA-157) make cassandra not allow itself to
run out of memory during sustained inserts
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis reassigned CASSANDRA-157:
----------------------------------------
Assignee: daishi
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
> Assignee: daishi
> Attachments: Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-157) make cassandra not allow itself
to run out of memory during sustained inserts
Posted by "daishi (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12716479#action_12716479 ]
daishi commented on CASSANDRA-157:
----------------------------------
> flushes are processed one at a time, first come first served, to avoid swamping the system with IO.
> does that jive with what you saw?
I guess that depends on what "first come first served" means.
Basically what I observed was that I could control how
frequently Memtable.flush() happened, but that only
after N memtables had been "queued up" to flush
would they actually be flushed to disk (I presume -
my view is based only on observing memory usage).
I think N was 10-20, but I don't remember exactly right now.
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Reporter: daishi
> Assignee: daishi
> Attachments: Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-157) make cassandra not allow itself
to run out of memory during sustained inserts
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12736066#action_12736066 ]
Hudson commented on CASSANDRA-157:
----------------------------------
Integrated in Cassandra #151 (See [http://hudson.zones.apache.org/hudson/job/Cassandra/151/])
use CSLS instead of NBHS for memtablesPendingFlush. See explanation here: http://sourceforge.net/tracker/?func=detail&aid=2828100&group_id=194172&atid=948362
patch by jbellis; reviewed by Brandon Williams for
> make cassandra not allow itself to run out of memory during sustained inserts
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-157
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Reporter: daishi
> Assignee: Jonathan Ellis
> Attachments: 157.patch, Cassandra-157_Unregister_Memtable_MBean.diff
>
>
> Tv on IRC pointed out to me that the issue that I've been encountering
> is probably point 2. in this roadmap:
> http://www.mail-archive.com/cassandra-dev@incubator.apache.org/msg00160.html
> I was unable to find any existing issue for this topic, so I'm creating a new one.
> Since this issue would block our use of Cassandra I'm happy to look into it,
> but if this is a known issue perhaps there's already a plan for addressing it
> that could be clarified?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.