Sponsored Content
Operating Systems Solaris Cannot mount Veritas volumes after cluster patch. Post 302587599 by buggin on Thursday 5th of January 2012 10:10:43 AM
Old 01-05-2012
Cannot mount Veritas volumes after cluster patch.

I am working with a Sun StoredgeTek 6540 disk array connected to two Sun 490 servers. After taking one of the 490 nodes on the cluster down to single user mode I proceeded to install the latest cluster patch from Oracle. After the patch was completed the system rebooted, failed to rejoin the cluster, crashdumped and rebooted continuously. The second node which was not updated and was functioning fine beforehand is also no longer able to connect to all the drives or the main Veritas drive group. I was running Solaris 10 137137-09 Sun Cluster 3.2 and Veritas 5 on both 490s before the cluster patch. I have since removed all the patches from the cluster patch and now both systems can boot up and join the cluster but cannot access the main Veritas drive group.

I have tried vxconfigrestore which fails.

Here are some sample errors when booting:

Dec 22 10:59:22 server02 Cluster.Framework: stderr: VxVM vxdg WARNING V-5-1-560
Dec 22 10:59:22 server02 Cluster.Framework: stderr: Disk satadg08: Not found, last known location: FLEXLINE0_4
Dec 22 10:59:22 server02 VxVM vxdg WARNING V-5-1-560 Disk satadg31: Not found, last known location: FLEXLINE0_0
Dec 22 10:59:22 server02 VxVM vxdg WARNING V-5-1-560 Disk satadg10: Not found, last known location: FLEXLINE0_10
..............
Dec 22 10:55:35 server02 scsi: WARNING: /scsi_vhci/ssd@g500000e014adbd30 (ssd2):
Dec 22 10:55:35 server02 drive offline
Dec 22 10:56:18 server02 vxvm:vxconfigd: V-5-1-8646 Error in claiming /dev/rdsk/c6t600A0B8000470CC2000008A347E0DDD6d0s2: I/O error
Dec 22 10:56:18 server02 vxvm:vxconfigd: V-5-1-8646 Error in claiming /dev/rdsk/c6t600A0B80004702300000085147E0E215d0s2: I/O error
Dec 22 10:56:18 server02 vxvm:vxconfigd: V-5-1-8646 Error in claiming /dev/rdsk/c6t600A0B80004702300000082147E0D921d0s2: I/O error
..........

I have copies of vxprint and vxdisk output from when the system was functioning normally and have 3 months worth of full tape backups but do not know where to begin to try and restore the main disk group. Been doing tons of reading online and just cannot seem to find the identical situation to mine. Not yet familiar enough with Veritas and Sun Cluster to fix this but am trying to learn as much as possible from this experience before this server is later put into production.
 

8 More Discussions You Might Find Interesting

1. High Performance Computing

veritas cluster with kernel patch update

hi all, can someone please share their process on how to kernel patch a red hat ent 4 with veritas cluster 5? it's compose of a primary and backup node. the resources are db, disk, nic. it doesn't need to be detail, just give me the steps like: login to the backup node and update the... (0 Replies)
Discussion started by: itik
0 Replies

2. Linux

one nfs mount point for many logical volumes

hi i am wondering if it is possible to use one nfs mount point for several logical volumes. i have a top level directory /imaging with data1 - data50 below it. each dataX directory is a logical volume configured through LVM. if i mount them separately on the client (i.e. 50 lines in fstab... (1 Reply)
Discussion started by: user23
1 Replies

3. High Performance Computing

SUN Cluster Vs Veritas Cluster

Dear All, Can anyone explain about Pros and Cons of SUN and Veritas Cluster ? Any comparison chart is highly appreciated. Regards, RAA (4 Replies)
Discussion started by: RAA
4 Replies

4. UNIX for Dummies Questions & Answers

Move veritas filesystem space between Logical volumes

I have a veritas file system (fsA) that is nearing capacity. We have secondary file system (fsB). that is unused and would like to move some of it's disk space but I'm unsure as to how to do this with. /dev/vx/dsk/vg05/lvol05 25288704 21887258 3188911 88% ... (2 Replies)
Discussion started by: Steelysteel
2 Replies

5. OS X (Apple)

Mount a disk elsewhere /Volumes

When a new disk is connected to a Mac/OSX, it automatically mounts on /Volumes. Is it possible to manually mount it elsewhere? For example, on "/raid"? - m66 - (5 Replies)
Discussion started by: makrell66
5 Replies

6. Solaris

Veritas Cluster

How to create your cluster members routes and manual nat, and proxy. (2 Replies)
Discussion started by: padmaja@tech
2 Replies

7. Solaris

Sun cluster and Veritas cluster question.

Yesterday my customer told me to expect a vcs upgrade to happen in the future. He also plans to stop using HDS and move to EMC. Am thinking how to migrate to sun cluster setup instead. My plan as follows leave the existing vcs intact as a fallback plan. Then install and build suncluster on... (5 Replies)
Discussion started by: sparcguy
5 Replies

8. UNIX for Beginners Questions & Answers

How to extend a disk in veritas volume manager in veritas cluster?

Hi Experts, I wanted to extend a veritas file system which is running on veritas cluster and mounted on node2 system. #hastatus -sum -- System State Frozen A node1 running 0 A node2 running 0 -- Group State -- Group System Probed ... (1 Reply)
Discussion started by: Skmanojkum
1 Replies
vxsplitlines(1M)														  vxsplitlines(1M)

NAME
vxsplitlines - show disks with conflicting configuration copies in a cluster SYNOPSIS
vxsplitlines [-g diskgroup] [-c daname] DESCRIPTION
If you import portions of a disk group on different systems, this can lead to conflicting configuration copies on the disks of the disk group. If the configuration information in a disk group is ambiguous, it may not be possible for Veritas Volume Manager to determine which config- uration copy is most up-to-date. (This is usually termed a serial split brain (SSB) condition when it occurs in a cluster.) You cannot import a disk group in this state unless you specify which disk's configuration copy to use. You can use the vxsplitlines command to see which disks in a disk group have conflicting configuration copies, and use this information together with your knowledge of the history of the disk groups' usage to determine which configuration copy is most valid. The output from vxsplitlines displays the vxdg commands that you can run to import the disk group using the available configuration copies. The -o selectcp option of the vxdg import command is used to select the configuration copy to use for the import. OPTIONS
-c daname Display the SSB IDs for each disk that are stored in the configuration copy on the disk specified by its disk access name. Note: Although the SSB IDs for some disks may match, this does not necessarily mean that those disks' configuration copies have recorded all the configuration changes. When viewed from some other configuration copies, the SSB IDs of the same disks may not match. -g diskgroup Specifies the disk group. If a disk group is not specified, the default disk group is used as determined from the rules on the vxdg(1M) manual page. EXAMPLES
Display the disks on each side of the split in the disk group newdg: vxsplitlines -g newdg Display the SSB IDs stored in the configuration copy on disk c2t4d0: vxsplitlines -g newdg -c c2t4d0 NOTES
The vxsplitlines is primarily intended to be used with private disk groups, but it also works with shared disk groups. The version number of the disk group must be 110 or greater. SEE ALSO
vxdg(1M) Veritas Volume Manager Administrator's Guide VxVM 5.0.31.1 24 Mar 2008 vxsplitlines(1M)
All times are GMT -4. The time now is 09:57 PM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy