5:05 p.m.
Thomas Fromm created ISPN-2163:
----------------------------------
Summary: LowMemoryEvent in case of internal OoM Exceptions
Key: ISPN-2163
URL: https://issues.jboss.org/browse/ISPN-2163
Project: Infinispan
Issue Type: Feature Request
Reporter: Thomas Fromm
Assignee: Manik Surtani
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
5:12 p.m.
[
https://issues.jboss.org/browse/ISPN-2163?page=com.atlassian.jira.plugin....
]
Thomas Fromm updated ISPN-2163:
-------------------------------
Description:
Problem:
During operations like put() on node A and distribution of the data to B it could happen,
that an OutOfMemoryException appears at B and the transaction fails. At node B I don't
get notified (beside log) about the underlying exception, cause the tx is invoked
remotly.
I'd like to get informed at node B, if such execptions happen. Maybe LowMemoryEvent in
CacheManager?
With such event the application on B can decide cleanup stuff, shutdown or whatever.
LowMemoryEvent in case of internal OoM Exceptions
-------------------------------------------------
Key: ISPN-2163
URL: https://issues.jboss.org/browse/ISPN-2163
Project: Infinispan
Issue Type: Feature Request
Reporter: Thomas Fromm
Assignee: Manik Surtani
Problem:
During operations like put() on node A and distribution of the data to B it could happen,
that an OutOfMemoryException appears at B and the transaction fails. At node B I don't
get notified (beside log) about the underlying exception, cause the tx is invoked
remotly.
I'd like to get informed at node B, if such execptions happen. Maybe LowMemoryEvent
in CacheManager?
With such event the application on B can decide cleanup stuff, shutdown or whatever.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
11:43 a.m.
[
https://issues.jboss.org/browse/ISPN-2163?page=com.atlassian.jira.plugin....
]
Galder Zamarreño commented on ISPN-2163:
----------------------------------------
The problem is that there's no guarantees that a JVM will be able to do anything when
an OOME is thrown.
LowMemoryEvent in case of internal OoM Exceptions
-------------------------------------------------
Key: ISPN-2163
URL: https://issues.jboss.org/browse/ISPN-2163
Project: Infinispan
Issue Type: Feature Request
Reporter: Thomas Fromm
Assignee: Manik Surtani
Problem:
During operations like put() on node A and distribution of the data to B it could happen,
that an OutOfMemoryException appears at B and the transaction fails. At node B I don't
get notified (beside log) about the underlying exception, cause the tx is invoked
remotly.
I'd like to get informed at node B, if such execptions happen. Maybe LowMemoryEvent
in CacheManager?
With such event the application on B can decide cleanup stuff, shutdown or whatever.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
11:53 a.m.
[
https://issues.jboss.org/browse/ISPN-2163?page=com.atlassian.jira.plugin....
]
Thomas Fromm commented on ISPN-2163:
------------------------------------
I know, but what could be a workaround? Parsing Tx-Fail reason, maybe its possible to
determine the failing node and trying to kick node out of cluster using jgroups?
LowMemoryEvent in case of internal OoM Exceptions
-------------------------------------------------
Key: ISPN-2163
URL: https://issues.jboss.org/browse/ISPN-2163
Project: Infinispan
Issue Type: Feature Request
Reporter: Thomas Fromm
Assignee: Manik Surtani
Problem:
During operations like put() on node A and distribution of the data to B it could happen,
that an OutOfMemoryException appears at B and the transaction fails. At node B I don't
get notified (beside log) about the underlying exception, cause the tx is invoked
remotly.
I'd like to get informed at node B, if such execptions happen. Maybe LowMemoryEvent
in CacheManager?
With such event the application on B can decide cleanup stuff, shutdown or whatever.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
10:59 a.m.
[
https://issues.jboss.org/browse/ISPN-2163?page=com.atlassian.jira.plugin....
]
Galder Zamarreño commented on ISPN-2163:
----------------------------------------
What Thomas is trying to do reminds me to those systems that monitor a
jvm/process/component for particular failures and are able to restart them if necessary
(or in this case, make them rejoin the cluster). It's maybe a bit heavy handed here,
but it's worth keeping it in mind.
LowMemoryEvent in case of internal OoM Exceptions
-------------------------------------------------
Key: ISPN-2163
URL: https://issues.jboss.org/browse/ISPN-2163
Project: Infinispan
Issue Type: Feature Request
Reporter: Thomas Fromm
Assignee: Manik Surtani
Problem:
During operations like put() on node A and distribution of the data to B it could happen,
that an OutOfMemoryException appears at B and the transaction fails. At node B I don't
get notified (beside log) about the underlying exception, cause the tx is invoked
remotly.
I'd like to get informed at node B, if such execptions happen. Maybe LowMemoryEvent
in CacheManager?
With such event the application on B can decide cleanup stuff, shutdown or whatever.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
1:32 p.m.
[
https://issues.jboss.org/browse/ISPN-2163?page=com.atlassian.jira.plugin....
]
Thomas Fromm commented on ISPN-2163:
------------------------------------
Yes, at the end some kind of disrupter detection and removal.
I'LL close this ticket. I'll count how many TX fails a node caused in a replicated
cache and a scheduled thread on every node check this value. If size gets to high during
an amount of time, the node shuts down.
LowMemoryEvent in case of internal OoM Exceptions
-------------------------------------------------
Key: ISPN-2163
URL: https://issues.jboss.org/browse/ISPN-2163
Project: Infinispan
Issue Type: Feature Request
Reporter: Thomas Fromm
Assignee: Manik Surtani
Problem:
During operations like put() on node A and distribution of the data to B it could happen,
that an OutOfMemoryException appears at B and the transaction fails. At node B I don't
get notified (beside log) about the underlying exception, cause the tx is invoked
remotly.
I'd like to get informed at node B, if such execptions happen. Maybe LowMemoryEvent
in CacheManager?
With such event the application on B can decide cleanup stuff, shutdown or whatever.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
1:32 p.m.
[
https://issues.jboss.org/browse/ISPN-2163?page=com.atlassian.jira.plugin....
]
Thomas Fromm closed ISPN-2163.
------------------------------
Resolution: Rejected
LowMemoryEvent in case of internal OoM Exceptions
-------------------------------------------------
Key: ISPN-2163
URL: https://issues.jboss.org/browse/ISPN-2163
Project: Infinispan
Issue Type: Feature Request
Reporter: Thomas Fromm
Assignee: Manik Surtani
Problem:
During operations like put() on node A and distribution of the data to B it could happen,
that an OutOfMemoryException appears at B and the transaction fails. At node B I don't
get notified (beside log) about the underlying exception, cause the tx is invoked
remotly.
I'd like to get informed at node B, if such execptions happen. Maybe LowMemoryEvent
in CacheManager?
With such event the application on B can decide cleanup stuff, shutdown or whatever.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
4295
days inactive
4299
days old
infinispan-issues@lists.jboss.org
6 comments
2 participants
participants (2)
-
Galder Zamarreño (JIRA) -
Thomas Fromm (JIRA)