How to clear udid_mismatch?
Created: 08 Nov. 2011 | Updated: 09 Décembre 2011 | 10 comments
Ce problème a été résolu. Voir la solution.
Dear experts,
could you advice me how to clear udid_mismatch from disks CFS 5.0?
I cloned a set of disks from a cluster in primary locality to another set of disks from a cluster in backup locality on disk array level.
I successfully imported DGs and mount file systems however problem is that status of cloned disks varies on servers in backup locality. Disks on CFS master are marked as "online shared" but they are marked "online shared udid_mismatch" on CFS slave.
I would like to clear udid_mismatch from disks on CFS slave as I suppose "online shared" is correct status.
Best regards,
Marek
Discussion Filed Under:
Commentaires Commentaires • Accéder au dernier commentaire
Hi,
Effectively when you cloned the disks, you also cloned a unique identifier (UUID) which has been detected on the system. Look up "vxdisk updateudid" in the document section below
https://sort.symantec.com/public/documents/sfha/5.1sp1/solaris/productguides/html/vxvm_admin/ch06s11.htm
Also you mention you have CFS 5.0 (no patches), so you may want to also investigate patching.
http://sort.symantec.com/patches
cheers
tony
Hi Tony,
thanks for advice.
If I'm right disk headers (UUID) were updated during DG import by "vxdg ... -o updateid import ..." and as all disks have correct status on CFS master I suppose that the update of disk headers passed successfully.
I need to propagate these changes to vxconfigd on CFS slave what is reverse operation to "vxdisk updateudid" - if I understand it well.
Regards,
Marek
Hi,
Can you share output or extract of "vxdisk -o alldgs list" from the slave system
cheers
tony
Hi Tony,
below are extracts of "vxdisk -o alldgs list" and "vxdg list" related to my case:
Regards,
Marek
Hi
You could try a VxVM device re-scan (i.e vxdisk scandisks) on the slave to see if that clears the udid_mismatch. Other than that, you may want to open up a Support case.
cheers
tony
Can you run this command, I let me now the output:
vxdg -o useclonedev=on -o updateid import dgapp
If this post has helped you, please vote or mark as solution. MBA. TC Venezuela Consultor Power
Hi all,
"vxdisk scandisks" did not solved my problem. Actually nothing has changed, udid_mismatch flags have not been cleared.
Unfortunately I cannot run "vxdg -o useclonedev=on -o updateid import dgapp" now as DG is already imported. I imported it on CFS master where all disks have correct status (see above) when I cloned disks few days ago. I used same command except option "-o useclonedev=on". However I assume it is not necessary to use this option as servers in backup locality see only cloned disks, not source disks.
Regards,
Marek
Hi,
I think we are now in the realm of opening a support case.
One area we could check is the master & slave view of the UDIDs for a particular disk, this can be done with "vxdisk -v list DiskAcess_name" on the later versions (not sure which exact version you are using)
i.e using dc2001 as an example
#On master
vxdisk -v list c10t4d0
#On Slave
vxdisk -v list c8t4d0
cheers
tony
Hi,
I had the same problem after some EMC Clones and used the following command :
# vxdisk -f -g disk_dg updateudid disk
Maybe you need to install the patch 140624-09 or higher (correct the udid flapping problem)
Regards.
Hi all,
first I would like to thank you for your advices.
Below are outputs from "vxdisk -v list c[9,7]t5d2" which is an identical disk (c9... is device available on CFS master and c7... is device available on CFS slave). You can see that there are only few differences and most of them are related to device names and the only real difference is udid_mismatch flag.
###CFS master Device: c9t5d2 devicetag: c9t5d2 type: auto clusterid: app_cluster disk: name=dc2d013 id=1292947259.104.dc2srv1 VxVM vxdisk ERROR V-5-1-0 dc2d013 - Record in multiple disk groups group: name=dgapp id=1260099189.120.dc2srv1 info: format=cdsdisk,privoffset=128 flags: online ready private autoconfig shared autoimport imported pubpaths: block=/dev/vx/dmp/c9t5d2 char=/dev/vx/rdmp/c9t5d2 guid: {8079aefc-1dd2-11b2-a026-001a4b08fee6} udid: HP%5F50%5F0EAFD%5F50%200EAFD0735 site: - version: 3.1 iosize: min=512 (bytes) max=256 (blocks) public: slice=0 offset=32896 len=67067264 disk_offset=0 private: slice=0 offset=128 len=32768 disk_offset=0 update: time=1318534322 seqno=0.9 ssb: actual_seqno=0.0 headers: 0 120 configs: count=1 len=24072 logs: count=1 len=3648 Defined regions: config priv 000024-000119[000096]: copy=01 offset=000000 disabled config priv 000128-024103[023976]: copy=01 offset=000096 disabled log priv 024104-027751[003648]: copy=01 offset=000000 disabled lockrgn priv 027752-027823[000072]: part=00 offset=000000 Annotations: tag udid_asl=HP%5F50%5F13131%5F50%20131310A6E Multipathing information: numpaths: 2 c9t5d2 state=enabled c5t5d2 state=enabled ###CFS slave VxVM vxdisk ERROR V-5-1-0 dc2d013 - Record in multiple disk groups Device: c7t5d2 devicetag: c7t5d2 type: auto clusterid: app_cluster disk: name=dc2d013 id=1292947259.104.dc2srv1 group: name=dgapp id=1260099189.120.dc2srv1 info: format=cdsdisk,privoffset=128 flags: online ready private autoconfig udid_mismatch shared autoimport imported pubpaths: block=/dev/vx/dmp/c7t5d2 char=/dev/vx/rdmp/c7t5d2 guid: {8079aefc-1dd2-11b2-a026-001a4b08fee6} udid: HP%5F50%5F0EAFD%5F50%200EAFD0735 site: - version: 3.1 iosize: min=512 (bytes) max=256 (blocks) public: slice=0 offset=32896 len=67067264 disk_offset=0 private: slice=0 offset=128 len=32768 disk_offset=0 update: time=1318534322 seqno=0.9 ssb: actual_seqno=0.0 headers: 0 120 configs: count=1 len=24072 logs: count=1 len=3648 Defined regions: config priv 000024-000119[000096]: copy=01 offset=000000 disabled config priv 000128-024103[023976]: copy=01 offset=000096 disabled log priv 024104-027751[003648]: copy=01 offset=000000 disabled lockrgn priv 027752-027823[000072]: part=00 offset=000000 Annotations: tag udid_asl=HP%5F50%5F13131%5F50%20131310A6E Multipathing information: numpaths: 2 c7t5d2 state=enabled c12t5d2 state=enabledI need to remove udid_mismatch from vxconfigd on CFS slave what is reverse operation to "vxdisk updateudid". If I understand it well "vxdisk updateudid" replaces the value that is stored in private region of the disk with the value that is stored in DDL property database (server memory/disk i.e. vxconfigd).
We are searching for current patches with our support guys which could solve our problem.
Regards,
Marek
aimez-vous répondre ?
Connectez-vous ou inscrivez-vous pour publier votre commentaire.