> On 2010-12-01 06:41:30, Kenneth Giusti wrote:
> > Alan, I think I found a bug in the part of the code I contributed to this
> > patch:
> >
> > In ManagementAgent::exportDeletedObject() - we should add a call to
> > moveNewObjectsLH() right before entering the main loop. Specifically:
> >
> >
> > =====
> >
> > void ManagementAgent::exportDeletedObjects(DeletedObjectList& outList)
> > {
> > sys::Mutex::ScopedLock lock (userLock);
> > list<pair<ObjectId, ManagementObject*> > deleteList;
> >
> > +
> > + moveNewObjectsLH();
> >
> > for (ManagementObjectMap::iterator iter = managementObjects.begin();
> > iter != managementObjects.end();
> > iter++) {
> >
> > ======
> >
> > Without this, there is a small chance that objects added, then deleted
> > *before* the next poll would not make it into the exported list.
Thanks, I made that change.
- Alan
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/130/#review48
-----------------------------------------------------------
On 2010-11-30 12:08:32, Alan Conway wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/130/
> -----------------------------------------------------------
>
> (Updated 2010-11-30 12:08:32)
>
>
> Review request for qpid.
>
>
> Summary
> -------
>
> This patch is a roll up of the following 3 patches:
>
> commit 5b28517ab3890a7f5316f2c89964343b9cc0b0dd
> Author: Alan Conway <[email protected]>
> Date: Mon Nov 29 14:39:46 2010 -0500
>
> Modified cluster_tests causes broker shut down with invalid-argument
> error.
>
> Described in https://bugzilla.redhat.com/show_bug.cgi?id=655078. The
> management agent's deleted-object list was not being replicated to new
> members joining the cluster, so management generated fewer deleted
> object notifications on the newer member, causing it to fail with an
> invalid-argument error. The list is now being replicated correctly.
>
> commit 733262b4d6cec8b0d30db949bfa93f11dc07773f
> Author: Alan Conway <[email protected]>
> Date: Tue Nov 23 16:35:24 2010 -0500
>
> Add missing call to Message::setTimestamp in
> ManagementAgent::sendBufferLH.
>
> commit 32ed120146db5f756d8ea4f9a7e0330bf5716f9d
> Author: Alan Conway <[email protected]>
> Date: Tue Nov 23 15:50:00 2010 -0500
>
> Enable cluster-safe assertions on transition to CATCHUP
>
> Delaying until READY was causing multiple clientConnect management
> events to be raised, because broker::Connection::setUserId relies on
> sys::isCluster to avoid producing duplicate events with
>
>
> Diffs
> -----
>
> /trunk/qpid/cpp/src/cluster.mk 1040689
> /trunk/qpid/cpp/src/qpid/cluster/Cluster.cpp 1040689
> /trunk/qpid/cpp/src/qpid/cluster/Connection.h 1040689
> /trunk/qpid/cpp/src/qpid/cluster/Connection.cpp 1040689
> /trunk/qpid/cpp/src/qpid/cluster/UpdateClient.cpp 1040689
> /trunk/qpid/cpp/src/qpid/management/ManagementAgent.h 1040689
> /trunk/qpid/cpp/src/qpid/management/ManagementAgent.cpp 1040689
> /trunk/qpid/cpp/src/tests/cluster_tests.py 1040689
> /trunk/qpid/cpp/xml/cluster.xml 1040689
>
> Diff: https://reviews.apache.org/r/130/diff
>
>
> Testing
> -------
>
> Passes make check, make check-long. Currently running make check-long in a
> loop.
>
>
> Thanks,
>
> Alan
>
>
