Closed openhpi2 closed 8 years ago
Original comment by: taclark
Deleted the description and reposted
Original comment by: dr_mohan
Hi,
As mentioned in the bug description, I upgraded the OpenHPI to 3.4.0 followed by a reboot, but do not see any core dump.
Any steps in order to reproduce the issue would be helpful for me.
Could you please do let us know that exactly what operation was being performed at the time of core observed?
Here what is the meaning of client application upgrade? What exactly do you do over here?
Could you please let us know the below: OA version OS version SSL version Glib version
Thanks! Praveen
Original comment by: openhpi2
If possible could you please try your steps with OpenHPI 3.6.0 and let us know the outcome?
Original comment by: openhpi2
Praveen,
We've never had a reproducer for this crash. It hasn't been seen by us or by our customer since I reported it here.
But to answer your questions, the OS where the problem was originally seen was Oracle Linux 6.5, based on RHEL 6.5, with openssl 1.0.1e and glibc 2.12. The OA hardware and firmware versions are not known. Our application upgrade amounts to moving the system to a runlevel that stops openhpid and the application using it, replaces files with new versions using RPM upgrades, and reboots.
As for OpenHPI 3.6.0, back in late January we updated from 3.4.0 to 3.6.1 in our development builds. Our post-build testing includes various upgrades, and we haven't seen the problem there either.
So it's all good news so far. . . .
Tim Clark
Original comment by: taclark
Thanks Tim for your update.
As we too are unable to re-create this issue with OpenHPI 3.6.1, shall I go ahead and close this issue as of now.
In future, in case this occurs then we can open it.
Thanks! Praveen
Original comment by: openhpi2
Original comment by: openhpi2
We had two cores of openhpid 3.4.0 on one system, one day apart. This was openhpid 3.4.0 + oa_soap plugin on a Red Hat based Linux system managing HP c-Class equipment.
I've included both cores in this ticket. Both were segfaults that happened in the process of dealing with power state. The first one occurred shortly after an upgrade of openhpid and our client application; the second one about a day later.
Core #1, from Sep 8 2015 07:48:49
This first core occurred after a reboot of the server on which the openhpid client application was running. The reboot followed an upgrade of the client application, in which openhpid was upgraded from 2.14.1 to 3.4.0.
Core #2, from Sep 9 2015 05:05:33
The second core occurred the next day. As far as I can tell, nothing unusual was happening on the system at the time.
Reported by: taclark