You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Jonathan Ellis (Created) (JIRA)" <ji...@apache.org> on 2012/01/02 15:54:31 UTC
[jira] [Created] (CASSANDRA-3687) Local range scans are not run on
the read stage
Local range scans are not run on the read stage
-----------------------------------------------
Key: CASSANDRA-3687
URL: https://issues.apache.org/jira/browse/CASSANDRA-3687
Project: Cassandra
Issue Type: Bug
Reporter: Jonathan Ellis
Priority: Minor
Fix For: 1.2
Running directly on the client request/StorageProxy thread means we're now allowing one range scan per thrift thread instead of one per read stage thread [which may be more, or less, depending on thrift server mode], and it bypasses the "drop hopeless requests" overcapacity protection built in there.
--
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-3687) Local range scans are not run
on the read stage
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-3687?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13281930#comment-13281930 ]
Jonathan Ellis commented on CASSANDRA-3687:
-------------------------------------------
We've already checked that size == 1, so position 0 is the only possible place for it to be. :)
> Local range scans are not run on the read stage
> -----------------------------------------------
>
> Key: CASSANDRA-3687
> URL: https://issues.apache.org/jira/browse/CASSANDRA-3687
> Project: Cassandra
> Issue Type: Bug
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Priority: Minor
> Fix For: 1.2
>
>
> Running directly on the client request/StorageProxy thread means we're now allowing one range scan per thrift thread instead of one per read stage thread [which may be more, or less, depending on thrift server mode], and it bypasses the "drop hopeless requests" overcapacity protection built in there.
--
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-3687) Local range scans are not run
on the read stage
Posted by "Vijay (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-3687?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13412493#comment-13412493 ]
Vijay commented on CASSANDRA-3687:
----------------------------------
+1
> Local range scans are not run on the read stage
> -----------------------------------------------
>
> Key: CASSANDRA-3687
> URL: https://issues.apache.org/jira/browse/CASSANDRA-3687
> Project: Cassandra
> Issue Type: Bug
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Priority: Minor
> Fix For: 1.2
>
>
> Running directly on the client request/StorageProxy thread means we're now allowing one range scan per thrift thread instead of one per read stage thread [which may be more, or less, depending on thrift server mode], and it bypasses the "drop hopeless requests" overcapacity protection built in there.
--
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-3687) Local range scans are not run
on the read stage
Posted by "Vijay (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-3687?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13278463#comment-13278463 ]
Vijay commented on CASSANDRA-3687:
----------------------------------
+10, one thing which i noticed was will it be better to remove (coz the localnode might not be in position 0).
{code}
if (handler.endpoints.size() == 1 && handler.endpoints.get(0).equals(FBUtilities.getBroadcastAddress())
{code}
and add it to
{code}
for (InetAddress endpoint : handler.endpoints)
{
if (endpoint.equals(FBUtilities.getBroadcastAddress())
{
logger.debug("reading data locally");
StageManager.getStage(Stage.READ).execute(new LocalRangeSliceRunnable(nodeCmd, handler));
}
{code}
similar to fetchRows?
> Local range scans are not run on the read stage
> -----------------------------------------------
>
> Key: CASSANDRA-3687
> URL: https://issues.apache.org/jira/browse/CASSANDRA-3687
> Project: Cassandra
> Issue Type: Bug
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Priority: Minor
> Fix For: 1.2
>
>
> Running directly on the client request/StorageProxy thread means we're now allowing one range scan per thrift thread instead of one per read stage thread [which may be more, or less, depending on thrift server mode], and it bypasses the "drop hopeless requests" overcapacity protection built in there.
--
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-3687) Local range scans are not run
on the read stage
Posted by "Jonathan Ellis (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-3687?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13178419#comment-13178419 ]
Jonathan Ellis commented on CASSANDRA-3687:
-------------------------------------------
Look at SP.LocalReadRunnable for how to do this safely. Simplest fix would be to just continue routing all range scans over MessagingService.
> Local range scans are not run on the read stage
> -----------------------------------------------
>
> Key: CASSANDRA-3687
> URL: https://issues.apache.org/jira/browse/CASSANDRA-3687
> Project: Cassandra
> Issue Type: Bug
> Reporter: Jonathan Ellis
> Priority: Minor
> Fix For: 1.2
>
>
> Running directly on the client request/StorageProxy thread means we're now allowing one range scan per thrift thread instead of one per read stage thread [which may be more, or less, depending on thrift server mode], and it bypasses the "drop hopeless requests" overcapacity protection built in there.
--
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