[jira] Commented: (ZOOKEEPER-831) BookKeeper: Throttling improved for reads
[ https://issues.apache.org/jira/browse/ZOOKEEPER-831?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12910965#action_12910965 ] Hudson commented on ZOOKEEPER-831: -- Integrated in ZooKeeper-trunk #940 (See [https://hudson.apache.org/hudson/job/ZooKeeper-trunk/940/]) ZOOKEEPER-831. BookKeeper: Throttling improved for reads BookKeeper: Throttling improved for reads - Key: ZOOKEEPER-831 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-831 Project: Zookeeper Issue Type: Bug Components: contrib-bookkeeper Affects Versions: 3.3.1 Reporter: Flavio Junqueira Assignee: Flavio Junqueira Fix For: 3.4.0 Attachments: ZOOKEEPER-831.patch, ZOOKEEPER-831.patch, ZOOKEEPER-831.patch, ZOOKEEPER-831.patch Reads and writes in BookKeeper are asymmetric: a write request writes one entry, whereas a read request may read multiple requests. The current implementation of throttling only counts the number of read requests instead of counting the number of entries being read. Consequently, a few read requests reading a large number of entries each will spawn a large number of read-entry requests. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Commented: (ZOOKEEPER-831) BookKeeper: Throttling improved for reads
[ https://issues.apache.org/jira/browse/ZOOKEEPER-831?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12909236#action_12909236 ] Ivan Kelly commented on ZOOKEEPER-831: -- +1 BookKeeper: Throttling improved for reads - Key: ZOOKEEPER-831 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-831 Project: Zookeeper Issue Type: Bug Components: contrib-bookkeeper Affects Versions: 3.3.1 Reporter: Flavio Junqueira Assignee: Flavio Junqueira Fix For: 3.4.0 Attachments: ZOOKEEPER-831.patch, ZOOKEEPER-831.patch, ZOOKEEPER-831.patch, ZOOKEEPER-831.patch Reads and writes in BookKeeper are asymmetric: a write request writes one entry, whereas a read request may read multiple requests. The current implementation of throttling only counts the number of read requests instead of counting the number of entries being read. Consequently, a few read requests reading a large number of entries each will spawn a large number of read-entry requests. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Commented: (ZOOKEEPER-831) BookKeeper: Throttling improved for reads
[
https://issues.apache.org/jira/browse/ZOOKEEPER-831?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12908724#action_12908724
]
Ivan Kelly commented on ZOOKEEPER-831:
--
submitOrdered can throw RejectedExecutionException (Im guessing rare to the
order of if this happens the machine will die soon) or NullPointerException
(unlikely). However there's no harm putting a try { } catch (Exception e) {
opCounterSem.release(); } around it. the handler will never run except if the
job is never submitted.
Also, you need a release inside the if (metadata.isClosed) . Otherwise I think
it should be fine. As PendingAddOp should be able to handle it no matter what
error occurs.
BookKeeper: Throttling improved for reads
-
Key: ZOOKEEPER-831
URL: https://issues.apache.org/jira/browse/ZOOKEEPER-831
Project: Zookeeper
Issue Type: Bug
Components: contrib-bookkeeper
Affects Versions: 3.3.1
Reporter: Flavio Junqueira
Assignee: Flavio Junqueira
Fix For: 3.4.0
Attachments: ZOOKEEPER-831.patch, ZOOKEEPER-831.patch,
ZOOKEEPER-831.patch
Reads and writes in BookKeeper are asymmetric: a write request writes one
entry, whereas a read request may read multiple requests. The current
implementation of throttling only counts the number of read requests instead
of counting the number of entries being read. Consequently, a few read
requests reading a large number of entries each will spawn a large number of
read-entry requests.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (ZOOKEEPER-831) BookKeeper: Throttling improved for reads
[ https://issues.apache.org/jira/browse/ZOOKEEPER-831?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12907144#action_12907144 ] Ivan Kelly commented on ZOOKEEPER-831: -- asyncAddEntry - why did you move the semaphore acquire in this? If an exception occurs during the adding of the operation, is the completion (and therefore the release) guaranteed? Otherwise the patch looks good. BookKeeper: Throttling improved for reads - Key: ZOOKEEPER-831 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-831 Project: Zookeeper Issue Type: Bug Components: contrib-bookkeeper Affects Versions: 3.3.1 Reporter: Flavio Junqueira Assignee: Flavio Junqueira Fix For: 3.4.0 Attachments: ZOOKEEPER-831.patch, ZOOKEEPER-831.patch Reads and writes in BookKeeper are asymmetric: a write request writes one entry, whereas a read request may read multiple requests. The current implementation of throttling only counts the number of read requests instead of counting the number of entries being read. Consequently, a few read requests reading a large number of entries each will spawn a large number of read-entry requests. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Commented: (ZOOKEEPER-831) BookKeeper: Throttling improved for reads
[
https://issues.apache.org/jira/browse/ZOOKEEPER-831?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12907149#action_12907149
]
Ivan Kelly commented on ZOOKEEPER-831:
--
You've added a public method to the API purely for testing. Is this necessary?
You can access the private members using reflection, which would save you
having to add anything to the API which could be difficult to remove in the
future.
@suppressWarning(unchecked)
int getAvailablePermits(LedgerHandle lh) {
Field field = LedgerHandle.class.getDeclaredField(opCounterSem);
field.setAccessible(true);
return ((Semaphore)field.get(lh)).getAvailablePermits();
}
BookKeeper: Throttling improved for reads
-
Key: ZOOKEEPER-831
URL: https://issues.apache.org/jira/browse/ZOOKEEPER-831
Project: Zookeeper
Issue Type: Bug
Components: contrib-bookkeeper
Affects Versions: 3.3.1
Reporter: Flavio Junqueira
Assignee: Flavio Junqueira
Fix For: 3.4.0
Attachments: ZOOKEEPER-831.patch, ZOOKEEPER-831.patch
Reads and writes in BookKeeper are asymmetric: a write request writes one
entry, whereas a read request may read multiple requests. The current
implementation of throttling only counts the number of read requests instead
of counting the number of entries being read. Consequently, a few read
requests reading a large number of entries each will spawn a large number of
read-entry requests.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (ZOOKEEPER-831) BookKeeper: Throttling improved for reads
[ https://issues.apache.org/jira/browse/ZOOKEEPER-831?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12895440#action_12895440 ] Hadoop QA commented on ZOOKEEPER-831: - +1 overall. Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12451268/ZOOKEEPER-831.patch against trunk revision 980576. +1 @author. The patch does not contain any @author tags. +1 tests included. The patch appears to include 3 new or modified tests. +1 javadoc. The javadoc tool did not generate any warning messages. +1 javac. The applied patch does not increase the total number of javac compiler warnings. +1 findbugs. The patch does not introduce any new Findbugs warnings. +1 release audit. The applied patch does not increase the total number of release audit warnings. +1 core tests. The patch passed core unit tests. +1 contrib tests. The patch passed contrib unit tests. Test results: http://hudson.zones.apache.org/hudson/job/Zookeeper-Patch-h7.grid.sp2.yahoo.net/107/testReport/ Findbugs warnings: http://hudson.zones.apache.org/hudson/job/Zookeeper-Patch-h7.grid.sp2.yahoo.net/107/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html Console output: http://hudson.zones.apache.org/hudson/job/Zookeeper-Patch-h7.grid.sp2.yahoo.net/107/console This message is automatically generated. BookKeeper: Throttling improved for reads - Key: ZOOKEEPER-831 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-831 Project: Zookeeper Issue Type: Bug Components: contrib-bookkeeper Affects Versions: 3.3.1 Reporter: Flavio Junqueira Assignee: Flavio Junqueira Fix For: 3.4.0 Attachments: ZOOKEEPER-831.patch Reads and writes in BookKeeper are asymmetric: a write request writes one entry, whereas a read request may read multiple requests. The current implementation of throttling only counts the number of read requests instead of counting the number of entries being read. Consequently, a few read requests reading a large number of entries each will spawn a large number of read-entry requests. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
