You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Michael McCandless (JIRA)" <ji...@apache.org> on 2011/06/23 21:56:47 UTC
[jira] [Created] (LUCENE-3235) TestDoubleBarrelLRUCache hangs under
Java 1.5, 3.x and trunk, likely JVM bug
TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
----------------------------------------------------------------------------
Key: LUCENE-3235
URL: https://issues.apache.org/jira/browse/LUCENE-3235
Project: Lucene - Java
Issue Type: Bug
Reporter: Michael McCandless
Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Mark Miller (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13054433#comment-13054433 ]
Mark Miller commented on LUCENE-3235:
-------------------------------------
bq. +1 to drop 1.5...
+1.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Michael McCandless (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13054071#comment-13054071 ]
Michael McCandless commented on LUCENE-3235:
--------------------------------------------
Still hangs if I run -client; but it looks like -Xint prevents the hang (235 iterations so far on beast).
3.2 also hangs.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Michael McCandless (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149728#comment-13149728 ]
Michael McCandless commented on LUCENE-3235:
--------------------------------------------
+1 for the safe-but-slow Java 5 only workaround....
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Michael McCandless (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13054349#comment-13054349 ]
Michael McCandless commented on LUCENE-3235:
--------------------------------------------
VERY interesting! Is anyone able to repro this hang besides me...?
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149941#comment-13149941 ]
Uwe Schindler commented on LUCENE-3235:
---------------------------------------
...and Solaris? The JVM BUG seems to only affect Solaris (according to the sun reports).
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149585#comment-13149585 ]
Uwe Schindler commented on LUCENE-3235:
---------------------------------------
I am currently preparing a patch.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Michael McCandless (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150917#comment-13150917 ]
Michael McCandless commented on LUCENE-3235:
--------------------------------------------
{quote}
If the whole point is to prevent double-seeks for queries doing IDF then getting termdocs,
we can just put a tiny LRU cache in the already-existing threadlocal...
{quote}
In fact this is what we used to do (LinkedHashMap), and then in LUCENE-2075 we moved to a cache shareable across threads.
But I agree -- risk of hangs is not worth it. Let's just move back to thread-private cache, but let's make it tiny (8? 16? 13?) in size?
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150905#comment-13150905 ]
Uwe Schindler commented on LUCENE-3235:
---------------------------------------
Robert, I agree, a simple LRUCache based on LinkedHashMap with
{code:java}
@Override
protected boolean removeEldestEntry(Map.Entry<K,V> eldest) {
return size() > MAX_ENTRIES;
}
{code}
would be fine. Because of this bugs and heavy use in Solr of other concurrent classes, we should maybe add a warning to the startup based on my findings before.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150561#comment-13150561 ]
Uwe Schindler commented on LUCENE-3235:
---------------------------------------
This testcase fails, but we are using concurrent also in ParallelMultiSearcher (die, die, die) and other places (even the indexer was partly upgraded to use ConcurrentLock). In my opinion we should not change our code to work around that issue. Just because one test case hangs its not guaranteed that other uses will work correctly. It brings a false security and slows down VMs that work correctly. And it only affects very modern processors.
If we would have a logging framework in Lucene (maybe Solr could do this): It could parse the args of Java (from system property) and look for -XX:+UseMembar, if its Java 1.5 it should print a warning to Solr/Lucene log file.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Updated] (LUCENE-3235) TestDoubleBarrelLRUCache hangs under
Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Robert Muir updated LUCENE-3235:
--------------------------------
Attachment: LUCENE-3235.patch
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13054067#comment-13054067 ]
Uwe Schindler commented on LUCENE-3235:
---------------------------------------
LOL, no comment.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149954#comment-13149954 ]
Robert Muir commented on LUCENE-3235:
-------------------------------------
and intel cpu
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13054070#comment-13054070 ]
Robert Muir commented on LUCENE-3235:
-------------------------------------
i ran the test with the same version as mike (1.5.0_22) in two ways on windows:
* -Dtests.iter=100
* in a loop from a script, 100 times with its own ant run.
i can't reproduce it on windows.
in my eyes, there isn't even an argument about whether or not we should support java5: its not possible, if bugs are not getting fixed.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149921#comment-13149921 ]
Robert Muir commented on LUCENE-3235:
-------------------------------------
Hoss you are right: we should also check Constants.SUN ?
otherwise lets not do the hack...
But i'm for the change because there is nothing slower than a hang/deadlock...
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Updated] (LUCENE-3235) TestDoubleBarrelLRUCache hangs under
Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Uwe Schindler updated LUCENE-3235:
----------------------------------
Attachment: LUCENE-3235.patch
Patch.
We should forward port the deprecation/removal of useless Constants.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149581#comment-13149581 ]
Uwe Schindler commented on LUCENE-3235:
---------------------------------------
An easy fix would be to use Collections.synchronizedMap(new HashMap()) in the ctor to initializer cache1 and cache2 (if Java 5 is detected)? If people are using Java 5 they get not-the best-performance.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150003#comment-13150003 ]
Uwe Schindler commented on LUCENE-3235:
---------------------------------------
That blog is as cool as the generics policeman ones... :-)
Thanks Simon, I think we should list this bug and its workaround in the wiki page and close this report.
Mike can you try -XX:+UseMembar ?
Uwe
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Simon Willnauer (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149631#comment-13149631 ]
Simon Willnauer commented on LUCENE-3235:
-----------------------------------------
bq. An easy fix would be to use Collections.synchronizedMap(new HashMap()) in the ctor to initializer cache1 and cache2 (if Java 5 is detected)? If people are using Java 5 they get not-the best-performance.
I like that too...
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13054336#comment-13054336 ]
Robert Muir commented on LUCENE-3235:
-------------------------------------
Mike, i installed 1.5.0_22 (amd64) on my linux machine, and i can't reproduce there either (i ran like 500 iterations).
Maybe my hardware isn't concurrent enough? or maybe you should un-overclock? :)
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150813#comment-13150813 ]
Robert Muir commented on LUCENE-3235:
-------------------------------------
{quote}
How are we slowing down other VMs...? We scope the workaround?
{quote}
I agree, according to the bug report it seems all 1.5's are affected and some 1.6's?
Doesn't seem like solaris is related either, especially since Mike hit it on linux
So the current patch is actually way under-scoped.
Sure, some 1.6's are affected, and if we want it to be even better,
we should likely improve Constants a little bit to make the minor
version more easily accessible, from the bug report it seems we
should at least consider doing something for < 1.6.0u21 ? And we should
remove the 'Solaris' check, but keep it little-endian because the bug
report mentions its way more likely to happen on those cpus.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149562#comment-13149562 ]
Robert Muir commented on LUCENE-3235:
-------------------------------------
wait, this statement makes no sense.
if 1.5 is no longer supported, then 1.5 should no longer be supported, and we should be
free to use 1.6 code everywhere.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150896#comment-13150896 ]
Robert Muir commented on LUCENE-3235:
-------------------------------------
Maybe we should remove this global cache completely?
If the whole point is to prevent double-seeks for queries doing IDF then getting termdocs,
we can just put a tiny LRU cache in the already-existing threadlocal...
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Michael McCandless (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13054363#comment-13054363 ]
Michael McCandless commented on LUCENE-3235:
--------------------------------------------
Indeed java -client -version shows it's still using server VM -- you're right!
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149919#comment-13149919 ]
Uwe Schindler commented on LUCENE-3235:
---------------------------------------
We are using ConcurrentHashMap also at other places, should we replace all of them or where is the bug that this happens only here?
It also appears to happen on Mike's machine, so maybe its hardware-related (Solaris?) as the Sun bugreport seems to tell us.
I am also +0 to apply the patch. I just showed one possibility how to fix this.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Issue Comment Edited] (LUCENE-3235)
TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM
bug
Posted by "Uwe Schindler (Issue Comment Edited) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150878#comment-13150878 ]
Uwe Schindler edited comment on LUCENE-3235 at 11/15/11 11:27 PM:
------------------------------------------------------------------
This ois the way to get the runtime args: [http://download.oracle.com/javase/1.5.0/docs/api/java/lang/management/RuntimeMXBean.html#getInputArguments()]:
{code:java}
RuntimeMXBean m = ManagementFactory.getRuntimeMXBean();
List<String> vmargs = m.getInputArguments();
{code}
This prints all args for me:
{noformat}
[junit] vmargs=[-Xmx512M, -Dtests.verbose=false, -Dtests.infostream=false, -Dtests.lockdir=C:\Users\Uwe, Schindler\Projects\lucene\trunk-lusolr1\lucene\build, -Dtests.codec=random, -Dtests.postingsformat=random, -Dtests.locale=random, -Dtests.timezone=random, -Dtests.directory=random, -Dtests.linedocsfile=europarl.lines.txt.gz, -Dtests.iter=1, -Dtests.iter.min=1, -Dtests.seed=random, -Dtests.luceneMatchVersion=4.0, -Dtests.cleanthreads=perMethod, -Djava.util.logging.config.file=/dev/null, -Dtests.nightly=false, -Dtests.asserts.gracious=false, -Dtests.multiplier=1, -DtempDir=C:\Users\Uwe, Schindler\Projects\lucene\trunk-lusolr1\lucene\build\test\1, -Dlucene.version=4.0-SNAPSHOT, -Dtestmethod=, -Djetty.testMode=1, -Djetty.insecurerandom=1, -Dsolr.directoryFactory=org.apache.solr.core.MockDirectoryFactory, -ea:org.apache.lucene..., -ea:org.apache.solr...]
{noformat}
was (Author: thetaphi):
This ois the way to get the runtime args: [http://download.oracle.com/javase/1.5.0/docs/api/java/lang/management/RuntimeMXBean.html#getInputArguments()]:
{code:java}
RuntimeMXBean m = ManagementFactory.getRuntimeMXBean();
List<String> vmargs = getInputArguments();
{code}
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150904#comment-13150904 ]
Robert Muir commented on LUCENE-3235:
-------------------------------------
actually i think even a tiny queue would work.
this would prevent the double-seeks. I dont think there is a point to worrying about anything else.
the LRU-ness seems stupid: for 'common' terms that appear over and over like stopwords, those are gonna be slow anyway.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Updated] (LUCENE-3235) TestDoubleBarrelLRUCache hangs under
Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Uwe Schindler updated LUCENE-3235:
----------------------------------
Attachment: LUCENE-3235.patch
Updated patch after LUCENE-3574 was committed. I also added a System.out.println to the test (VERBOSE only).
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Michael McCandless (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150484#comment-13150484 ]
Michael McCandless commented on LUCENE-3235:
--------------------------------------------
That's a nice blog post! What a scary platform-specific JVM bug...
I still hit hit this hang reasonably often when running 3.x tests. It's always the DBLRU cache, so far anyway.
Because this is our most intense use of a CHM... I still think the workaround (scoped down to 1.5, Sun JVM, little endian arch) makes sense? I agree it won't fully work around the JVM bug, since in theory other uses of java.util.concurrent.* could hit it, but it can prevent the most common occurrence? The patch seems minimal and worth it... a hang is a truly awful.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13054066#comment-13054066 ]
Robert Muir commented on LUCENE-3235:
-------------------------------------
+1 to drop java 5
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Resolved] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Simon Willnauer (Resolved) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Simon Willnauer resolved LUCENE-3235.
-------------------------------------
Resolution: Won't Fix
we moved to 1.6 on trunk seems we can't do much about it on 3.x - folks should run their stuff on 1.6 jvms or newer
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Simon Willnauer (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149979#comment-13149979 ]
Simon Willnauer commented on LUCENE-3235:
-----------------------------------------
here is a very detailed writeup for this:
http://blogs.oracle.com/dave/entry/a_race_in_locksupport_park
some interesting facts:
* The problem would only manifest when we were using the -UseMembar optimization that lets us remove fences from certain hot thread state transitions paths that need to coordinate safepoints between mutator threads and the JVM. This feature is enabled by default, but we can turn it off with the -XX:+UseMembar switch, which causes the JVM to emit normal fence instructions in the state transitions paths.
* The bug is a "day-one" bug and present in all versions of HotSpot.
* Parker::park() and unpark() reside in os_linux.cpp, os_solaris.cpp and os_windows.cpp for Linux, Solaris and Windows, respectively.
* The built-in synchronized implementation uses a different park mechanism (PlatformPark::) whereas the java.util.concurrent infrastructure uses Parker::. Only Parker:: is vulnerable.
* The bug will not manifest on uniprocessors or environments where threads are otherwise constrained to just a single processor.
I think the only reasonable fix for this is to recommend people to use -XX:+UseMembar if they are running on a vulnerable JVM
simon
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149967#comment-13149967 ]
Robert Muir commented on LUCENE-3235:
-------------------------------------
I modified the patch, the thing is that Mike was seeing this on Linux I think too...
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149565#comment-13149565 ]
Uwe Schindler commented on LUCENE-3235:
---------------------------------------
I agree with Robert. This issue is still existent in 3.x as we officially support Java 5.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Reopened] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (Reopened) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Robert Muir reopened LUCENE-3235:
---------------------------------
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Dawid Weiss (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13054366#comment-13054366 ]
Dawid Weiss commented on LUCENE-3235:
-------------------------------------
I'm same as Robert: +1 to drop 1.5...
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150878#comment-13150878 ]
Uwe Schindler commented on LUCENE-3235:
---------------------------------------
This ois the way to get the runtime args: [http://download.oracle.com/javase/1.5.0/docs/api/java/lang/management/RuntimeMXBean.html#getInputArguments()]
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Simon Willnauer (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150363#comment-13150363 ]
Simon Willnauer commented on LUCENE-3235:
-----------------------------------------
I updated the wiki...
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150874#comment-13150874 ]
Uwe Schindler commented on LUCENE-3235:
---------------------------------------
Yes, also 1.6.0_17 is affected. As always, 1.6.0_18 is the last and only good JVM :-)
My problem with the patch is that it only affects very few users, most users will have a working environment also with broken JVMs. The fix in the patch is very heavy, as, if we apply it correctly, will also slowdown <1.6.0_18.
As I said before, we should at least instruct Solr to print a WARN in the log if a JVM < 1.6.0_18 is started and the JVM parameter -XX:+UseMembar is missing. In Lucene we have no way to tell this the user as we have no logging framework, alternatively we could throw an Error is one of the central classes in Lucene is loaded by classloader and the JVM parameter is not given (static initializer e.g. in Constants.java). The same way we could tell the user: Dont use Java 7 GA.
As far as I know, the JVM command can be checked with a System-property and a simple regex should help.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Issue Comment Edited] (LUCENE-3235)
TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM
bug
Posted by "Uwe Schindler (Issue Comment Edited) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150878#comment-13150878 ]
Uwe Schindler edited comment on LUCENE-3235 at 11/15/11 11:21 PM:
------------------------------------------------------------------
This ois the way to get the runtime args: [http://download.oracle.com/javase/1.5.0/docs/api/java/lang/management/RuntimeMXBean.html#getInputArguments()]:
{code:java}
RuntimeMXBean m = ManagementFactory.getRuntimeMXBean();
List<String> vmargs = getInputArguments();
{code}
was (Author: thetaphi):
This ois the way to get the runtime args: [http://download.oracle.com/javase/1.5.0/docs/api/java/lang/management/RuntimeMXBean.html#getInputArguments()]
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Simon Willnauer (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149577#comment-13149577 ]
Simon Willnauer commented on LUCENE-3235:
-----------------------------------------
well then we should fix it - I will mark it as 3.5
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Updated] (LUCENE-3235) TestDoubleBarrelLRUCache hangs under
Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Simon Willnauer (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Simon Willnauer updated LUCENE-3235:
------------------------------------
Affects Version/s: 3.0
3.1
3.2
3.3
3.4
Fix Version/s: 3.5
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Mark Miller (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150545#comment-13150545 ]
Mark Miller commented on LUCENE-3235:
-------------------------------------
bq. I still think the workaround (scoped down to 1.5, Sun JVM, little endian arch) makes sense?
+1 Doesn't hurt other JVMs, improves things on the Sun JVM (something that hangs a lot and then does not hang is a big improvement in my book), and putting a workaround command to use in the wiki just seems a whole lot less user friendly to me. It doesn't mean we have to try and address every use of java.util.concurrent to work around this specific issue, does it?
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Hoss Man (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149906#comment-13149906 ]
Hoss Man commented on LUCENE-3235:
----------------------------------
+0
if anyone else suggested that we should add a "slow hack" to work around a Sun JRE bug i would argue that we were being unfair to people using other JRE (ie: does IBM's JRE have this bug? -- do IBM java 1.5 users deserve slower performance because Sun's JRE has a bug?) but since rmuir is the biggest proponent I know of not assuming everyone on the planet uses Sun JREs, and he's signed off on this, I'll defer.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149584#comment-13149584 ]
Robert Muir commented on LUCENE-3235:
-------------------------------------
I like Uwe's idea: not-the-best-performance is far preferable to a hang/deadlock!!!!!
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Michael McCandless (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150775#comment-13150775 ]
Michael McCandless commented on LUCENE-3235:
--------------------------------------------
bq. In my opinion we should not change our code to work around that issue.
In general, I think we should change our code to work around awful JVM
bugs, as long as 1) it's not so much effort for us to to do so (and as
always a volunteer steps up to the task), and 2) the change has
negligible cost to "lucky" users (who use a JVM / the right flags that
would not have hit the JVM bug).
I think the last patch fits these criteria, since it's a tiny change
and it scopes the workaround?
We've done this many times in the past; if the cost to "lucky" users
is negligible and the benefit to "unlucky" users (unknowingly using
the affected JVMs) is immense (not hitting horrific bug), I think the
tradeoff is worthwhile? Otherwise users will conclude Lucene (or
whatever software is embedding it) is buggy.
bq. This testcase fails, but we are using concurrent also in ParallelMultiSearcher (die, die, die) and other places (even the indexer was partly upgraded to use ConcurrentLock).
Right, we use concurrent* elsewhere, but terms dict is the big
user... very few apps actually use PMS.
bq. It brings a false security and slows down VMs that work correctly.
Well, we already have "false security" that Lucene won't hang on any
JVM... we don't claim this patch will fully work around the bug, but
at least it should reduce it.
How are we slowing down other VMs...? We scope the workaround?
I'm not saying we should go crazy here, making a big patch to avoid
concurrent* everywhere, but the current patch is minimal, addresses
the big usage of concurrent* in 3.x, is scoped down well.
It will avoid hangs for some number unlucky users out there... so why
not commit it?
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Michael McCandless (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13054362#comment-13054362 ]
Michael McCandless commented on LUCENE-3235:
--------------------------------------------
Yes the stack looks just like the stack overflow link I posted -- several threads stuck in sun.misc.Unsafe.park ;)
java -Xint definitely does not hang... ran for like 4200 iterations.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Dawid Weiss (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13054353#comment-13054353 ]
Dawid Weiss commented on LUCENE-3235:
-------------------------------------
I don't think you can force -client if it's a 64 bit release and you have tons of memory, can you? You can check by running java -client -version -- this is what it tells me, for example:
{noformat}
dweiss@dweiss-linux:~/work/lucene/lucene-trunk$ java -client -version
java version "1.6.0_16"
Java(TM) SE Runtime Environment (build 1.6.0_16-b01)
Java HotSpot(TM) 64-Bit Server VM (build 14.2-b01, mixed mode)
{noformat}
Can you do a remote stack of all the VM (or run it from the console and send it a signal to dump all threads)?
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Uwe Schindler (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149701#comment-13149701 ]
Uwe Schindler commented on LUCENE-3235:
---------------------------------------
I wait until tomorrow before I commit this "safe-but-slow" fix.
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Robert Muir (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13151001#comment-13151001 ]
Robert Muir commented on LUCENE-3235:
-------------------------------------
we can also offer an option... fasterButMoreHangs...
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (LUCENE-3235) TestDoubleBarrelLRUCache hangs
under Java 1.5, 3.x and trunk, likely JVM bug
Posted by "Simon Willnauer (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/LUCENE-3235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150355#comment-13150355 ]
Simon Willnauer commented on LUCENE-3235:
-----------------------------------------
bq. Thanks Simon, I think we should list this bug and its workaround in the wiki page and close this report.
+1 this is not our problem. if we go and fix all java.util.concurrent uses in solr & lucene we gonna end up in a big mess. According to the oracle blog this is also in 1.6 jvms and you will be vulnderable if you use any CHM like classes in your own code...
> TestDoubleBarrelLRUCache hangs under Java 1.5, 3.x and trunk, likely JVM bug
> ----------------------------------------------------------------------------
>
> Key: LUCENE-3235
> URL: https://issues.apache.org/jira/browse/LUCENE-3235
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 3.0, 3.1, 3.2, 3.3, 3.4
> Reporter: Michael McCandless
> Fix For: 3.5
>
> Attachments: LUCENE-3235.patch, LUCENE-3235.patch, LUCENE-3235.patch
>
>
> Not sure what's going on yet... but under Java 1.6 it seems not to hang bug under Java 1.5 hangs fairly easily, on Linux. Java is 1.5.0_22.
> I suspect this is relevant: http://stackoverflow.com/questions/3292577/is-it-possible-for-concurrenthashmap-to-deadlock which refers to this JVM bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6865591 which then refers to this one http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6822370
> It looks like that last bug was fixed in Java 1.6 but not 1.5.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org