| Random crashes / lockups - No errors, no video [message #12424] |
Fri, 27 April 2007 17:15  |
tutt
Messages: 19
Registered: April 2007
|
Junior Member |
|
|
I've got a customer with OpenVZ and HyperVM. Since they have been hosting with us, they have been riddled with problems. We went through 2 VERY expensive machines and then replaced them with 4 new VERY expensive machines to rule out hardware problems. The machines are 4 each of:
Dual 5130 Woodcrest Dual CPU
X7DVL-E SuperMicro Motherboard
1U Rackmount Chassis
4x 146GB Seagate 10Krpm HDDs
6GB FBDIMM 667MHz RAM
Hardware RAID 10
They are running CentOS 4.4 64bit with the latest OpenVZ kernel:
2.6.9-023stab043.2-smp #1 SMP Fri Mar 9 11:50:20 MSK 2007 x86_64 x86_64 x86_64 GNU/Linux
One of the 4 machines has a different SuperMicro motherboard and a SuperMicro RAID controller while the other 3 have the configurations listed above.
3 out of 4 of these machines (including the one that has slightly different hardware including the main board and RAID controller) have locked up on a completely random basis. As far as we know, there are no related errors in /var/log/messages. When one of our techs brings up a console to the machine, there is no video whatsoever. Some of these machines will run for a week or two and then have this happen. Others will go through spurts of locking up every few days. Every once in a while, massive filesystem corruption causes additional headaches.
Before you say that it must be hardware, we have burn tested each of these machines. Memory, disk, CPU all check out OK. These 4 already replace 2 even more powerful machines. These are brand new SuperMicro machines. I would think the chance of 5 SuperMicro machines all with sporadic hardware issues has to be equivelent to the chance of 10 meteors falling on my head as I type this.
I am thinking that someone out there has a similar hardware and software config as my customer and that maybe someone else is experiencing the same issues? Does ANYONE have any ideas as this has gotten to the point of frustrating me and my customer to no end!
[Updated on: Fri, 27 April 2007 17:16] Report message to a moderator |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Re: Random crashes / lockups - No errors, no video [message #12989 is a reply to message #12865] |
Thu, 17 May 2007 02:54  |
tutt
Messages: 19
Registered: April 2007
|
Junior Member |
|
|
|
We tried re-installing 32bit CentOS on one of the machines with no luck. What kernel version are you running now that these random lockups have disappeared? Other than switching to 32-bit CentOS, was there ANYTHING else you did that may have played a role in fixing the problem?
|
|
|
|
OpenVZ Forum
Members
Search
Help
Register
Login
Home
