[JBoss JIRA] (WFLY-2891) Not authorised write operation does not get audit logged if log-read-only="false"
by Kabir Khan (JIRA)
Kabir Khan created WFLY-2891:
--------------------------------
Summary: Not authorised write operation does not get audit logged if log-read-only="false"
Key: WFLY-2891
URL: https://issues.jboss.org/browse/WFLY-2891
Project: WildFly
Issue Type: Bug
Security Level: Public (Everyone can see)
Affects Versions: 8.0.0.CR1
Reporter: Kabir Khan
Fix For: 8.0.1.Final
This is because audit logging uses the controller lock to find out if the model was a write operation. If rbac is enabled and an operation is not authorised, the error happens before the controller lock is taken. So if audit log log-read-only="false" the failed operation does not get logged.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 5 months
[JBoss JIRA] (WFLY-1011) there should be some notification when modcluster is available but not working because Advertise is silently failing.
by Radoslav Husar (JIRA)
[ https://issues.jboss.org/browse/WFLY-1011?page=com.atlassian.jira.plugin.... ]
Radoslav Husar resolved WFLY-1011.
----------------------------------
Labels: rhq usability (was: rhq)
Fix Version/s: 8.0.0.Final
Resolution: Done
Fixed. I have added trace logging to help debug this. There is a log of adv. messages coming in, we dont want this on a high level. Its on trace logging of mod_cluster package.
If you turn on logging for
{code}
<logger category="org.jboss.modcluster">
<level name="TRACE"/>
</logger>
{code}
you will get:
{code}
13:33:16,694 TRACE [org.jboss.modcluster.advertise.impl.AdvertiseListenerImpl] (pool-1-thread-1) Advertise message digest verification passed for server 0c5d65e6-824d-4f84-9fe5-e6a47d5373da
{code}
or
{code}
13:35:09,671 TRACE [org.jboss.modcluster.advertise.impl.AdvertiseListenerImpl] (pool-1-thread-1) Advertise message digest verification failed for server a024799d-629d-4db6-9295-de3b84201d8d
{code}
> there should be some notification when modcluster is available but not working because Advertise is silently failing.
> -----------------------------------------------------------------------------------------------------------------------
>
> Key: WFLY-1011
> URL: https://issues.jboss.org/browse/WFLY-1011
> Project: WildFly
> Issue Type: Enhancement
> Security Level: Public(Everyone can see)
> Components: Clustering
> Environment: standalone server with modcluster and simple load balance test.
> Reporter: Simeon Pinder
> Assignee: Radoslav Husar
> Labels: rhq, usability
> Fix For: 8.0.0.Final
>
>
> Run standalone server with -c standalone-ha.xml and mod_cluster configured on the same box for a simple load balancing test. Modcluster is not able to detect the as7 instance without explicitly setting the Profiles > ModCluster > Proxies> ProxyList property in console to the ServerAdvertise value from httpd.conf. There should be some sort of indication in the UI or in the server logs when Advertise functionality is not working or fails to work.
> Additionally the documentation for modcluster setup and FAQ should be updated to include this note for those troubleshooting their modcluster setups.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 5 months
[JBoss JIRA] (JBJCA-1138) TransactionSynchronizer leaks memory and causes deadlocks in jacorb
by Andrey Smirnov (JIRA)
[ https://issues.jboss.org/browse/JBJCA-1138?page=com.atlassian.jira.plugin... ]
Andrey Smirnov updated JBJCA-1138:
----------------------------------
Description:
Seems like TransactionSynchronizer has been rolled back to state it was in August. I think this was caused by little typo in commit 469b4c303dd0d5c39f3476a5972556d8b9537b44
296: Record record = records.get(tx);
But 'records' map uses TransactionKey as key and above line should be written as
296:TransactionKey key = new TransactionKey(tx);
297: Record record = records.get(key);
I've checked fixed 1.0.22 and mem-leaks and deadlocks are all gone
was:
Seems like TransactionSynchronizer has been rolled back to state it was in August. I think this was caused by little typo in commit 469b4c303dd0d5c39f3476a5972556d8b9537b44
296: Record record = records.get(tx);
But 'records' map uses TransactionKey as key and above line should be written as
296:TransactionKey key = new TransactionKey(tx);
297: Record record = records.get(key);
> TransactionSynchronizer leaks memory and causes deadlocks in jacorb
> -------------------------------------------------------------------
>
> Key: JBJCA-1138
> URL: https://issues.jboss.org/browse/JBJCA-1138
> Project: IronJacamar
> Issue Type: Bug
> Components: Core
> Affects Versions: 1.0.23.Final
> Reporter: Andrey Smirnov
> Assignee: Jesper Pedersen
> Priority: Blocker
> Fix For: 1.0.24.Final
>
>
> Seems like TransactionSynchronizer has been rolled back to state it was in August. I think this was caused by little typo in commit 469b4c303dd0d5c39f3476a5972556d8b9537b44
> 296: Record record = records.get(tx);
> But 'records' map uses TransactionKey as key and above line should be written as
> 296:TransactionKey key = new TransactionKey(tx);
> 297: Record record = records.get(key);
> I've checked fixed 1.0.22 and mem-leaks and deadlocks are all gone
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 5 months