Hi, I did last weekend yum updates on 7 RHEL ( Enterprise Server) 2.6.18-238.12.1.el5 to version 2.6.18-308.8.2.el5 . All of this servers has Veritas 5.6 and Veritas Cluster 5.1 on top. The problem we run in was on the vrts level. For example , a simple vxdisk list command takes 5 to 10 minutes to come with his output. From the Symantec SORT site , we've found the patches for the version 5.6 and installed this. The patch VRTSvxvm-126.96.36.1990-SP1RP2P3_RHEL5; VRTSvxfen-188.8.131.52-SP1RP2P2_RHEL5; VRTSvxfs-5.1.101.000-RP1_RHEL5. But with no results.
At the END of the night, we've booted the RHEL servers with the Previous kernel 2.6.18-238.12.1.el5 . Than, the Veritas disk commands runs as before .
Is the kernel in the Compatibility Matrix or has someone a brilliant idea?
(I need a solution is more as I would like to have a solution(°_°))
So it sounds that after the RHEL kernel upgrade, some of the VxVM commands experienced delays in their execution.
Some Questions if I may
- Did all seven systems have the issue ?
- What storage is being used ?
- Any thing "odd" in syslog ?
- Did any other storage related commands (fdisk etc) experience a problem
1) yes, all the seven severs
2) clariion CX500 flare 26 , update RHEL servers needed for migration to Fujitsu Eternus DX410
3)no errors in the messages
4) not tryed
Is there a Matrix on the Symantec with RHEL kernel vesion versus the VRTS version?
We are running a 2 node RHEL 5.8 OS with SFHA 5.1 SP1RP2 installed.
The patch Veritas Cluster Server 5.1 SP1 RP2 P-patch 2, is also installed due to some fencing issues.
When upgrading RHEL patches a new kernel was also installed.
After this we faced problems with importing disk groups from VCS (cluster).
It takes too long time to import the diskgroups so some timeouts in VCS activates and the cluster fails.
We are using HP EVA 4400 disk system.
If you load the old kernel everything works OK again.
Old working kernel: 2.6.18-274.17.1.el5
New failing kernel: 2.6.18-308.13.1.el5