I did a little debugging and it seems like the first stack trace I
posted was wrong. I was using the wrong libtool binary I guess. Here
is what gdb reports on the correct binary.
I'll try and svn up and retest again tomorrow, but if anyone knows of
any changes in this codepath recently let me
On Thu, Oct 28, 2010 at 8:16 AM, Andrew Stitcher astitc...@redhat.com wrote:
On Wed, 2010-10-27 at 15:31 -0700, Aaron Fabbri wrote:
I recently updated to svn trunk latest and now my Rdma broker is crashing.
1. Is anyone else seeing this?
Is this happening before connections, during, after
Yep you got it. Thanks for the hint. Adding --no-module-dir (also
suggested by Andrew Stitcher) fixed it. I'm going to file a bug and
you guys can decide what (if anything) to do about it.
On Thu, Oct 28, 2010 at 2:27 AM, Gordon Sim g...@redhat.com wrote:
On 10/28/2010 08:18 AM, Aaron Fabbri
Affects Versions: 0.7
Environment: RHEL 5.5
Reporter: aaron fabbri
On svn trunk as of about Oct 25th, my qpidd broker was crashing due to rdma.so
being loaded twice (or due to one of them being old and incompatible). There
should be some warning to the user what they are doing
I recently updated to svn trunk latest and now my Rdma broker is crashing.
1. Is anyone else seeing this?
2. Should I open a bug.. and if so please give pointer to url and
anything else I'd need to know.
I'm running from gdb with this script
# cat gdbscript
file
I whipped up a quick state diagram for Andrew's latest RdmaIO stuff.
I'll paste the dot source here and attach the png.
To generate (assuming you have graphviz installed):
linux$ dot asynch_io_state_machine.dot -Tpng -o asynch_io_state_machine.png
Here's the source:
digraph asynchio_fsm {
Figures.. found a missing edge right after sending this out. This is better:
linux$ dot dot asynch_io_state_machine.dot -Tpng -o asynch_io_state_machine.png
---
digraph asynchio_fsm {
node [shape = doublecircle] IDLE;
node [fontsize=9] NOTIFY_PENDING;
node [shape =
On Tue, Oct 12, 2010 at 10:22 AM, Andrew Stitcher astitc...@redhat.com wrote:
For those interested in the Rdma implementation:
I've been doing a lot of stability work, stressing the rdma code in odd
corner cases (unexpected disconnects mostly). While on this trail I
reailised I could simplify
Just noticed this when reading the Rdma transport unit test code:
Index: RdmaServer.cpp
===
--- RdmaServer.cpp (revision 1004875)
+++ RdmaServer.cpp (working copy)
@@ -118,18 +118,18 @@
ConRec* cr =
Actually, use this patch.. it fixes an additional use-after-free.
Index: RdmaServer.cpp
===
--- RdmaServer.cpp (revision 1004875)
+++ RdmaServer.cpp (working copy)
@@ -118,18 +118,18 @@
ConRec* cr =
Thanks for the comments Andrew... this thread piqued my interest.. see
inline comments below.
On Tue, Sep 14, 2010 at 7:47 AM, Andrew Stitcher astitc...@redhat.com wrote:
From the description it seems that you are asking about the RDMA
AsynchIO state transition. You should have said that as
On Tue, Aug 31, 2010 at 7:00 PM, Alan Conway acon...@redhat.com wrote:
On 08/31/2010 01:15 AM, a fabbri wrote:
On Mon, Aug 30, 2010 at 6:38 AM, Alan Conwayacon...@redhat.com wrote:
On 08/27/2010 07:55 PM, a fabbri wrote:
...
I build from SVN checkout. I prefer to use a single NFS-mounted
cc:ing the list, as I think this may be interesting for others.
On 8/27/10 7:18 AM, Andrew Stitcher astitc...@redhat.com wrote:
On Thu, 2010-08-26 at 16:13 -0700, aafabbri wrote:
Hi Andrew,
Just curious if you observed any performance gains from your commit (below).
Chances are you haven't
On Fri, Aug 27, 2010 at 12:46 PM, Andrew Stitcher astitc...@redhat.com wrote:
On Fri, 2010-08-27 at 12:32 -0700, a fabbri wrote:
...
I've not tested, but I'd be very surprised.
IIRC, my strace of qpidd and client showed a ton of setsigprocmask()
syscalls. These are not cheap. I haven't
Does anyone regularly build the linux C++ stuff outside of the source tree?
That is, you're supposed to be able to run configure and make from a
build directory outside of qpid/cpp. I seem to have more build issues
this way though.
Hi,
I'm able to run the C++ broker w/ the RDMA transport by using the
--load-module option, e.g.:
qpidd --load-module rdma.so --transport rdma ...
Next I try to use latencytest with -P rdma, but it complains (as the
protocol is not registered).
How should I make latencytest load the transport
, a fabbri ajfab...@gmail.com wrote:
Hi,
I'm able to run the C++ broker w/ the RDMA transport by using the
--load-module option, e.g.:
qpidd --load-module rdma.so --transport rdma ...
Next I try to use latencytest with -P rdma, but it complains (as the
protocol is not registered).
How
17 matches
Mail list logo
