Sun System Handbook - ISO 3.4 June 2011 Internal/Partner Edition | |||
|
|
Solution Type Problem Resolution Sure Solution 1018111.1 : VERITAS Volume Manager 3.x and Sun Enterprise Volume Manager[TM] 2.x: How to Repair the Private Region When "vxdisk list" Shows "online altused"
PreviouslyPublishedAs 229433 Symptoms This document applies to Sun Enterprise Volume Manager[TM] 2.x and Veritas Volume Manager (VxVM) 3.x. There are several scenarios where a disk under the control of Volume Manager software can get into an "online altused" state. Volume Manager software stores the configuration of the individual disk in the private region in two places for redundancy. If one of these "configuration copies" gets corrupted because of a bad block on the disk, the disk will be marked as "online altused" to notify the administrator that one of the copies is disabled. Online altused can also be generated after moving an A3x00 array from one host to another. The private regions are not corrupted but have sequence numbers out of sync. In either case, the following symptoms will occur: # vxdisk list DEVICE TYPE DISK GROUP STATUS c0t8d0s2 sliced rootdisk rootdg online c0t9d0s2 sliced disk02 rootdg online c1t0d0s2 sliced disk01 rootdg online c2t4d1s2 sliced - - online altused c2t5d0s2 sliced - - online altused When attempting to import the disk group the following error is produced: # vxdg -C import datadg vxvm:vxconfigd:ERROR error in disk group configuration copies. No valid disk found containing disk group. Resolution To fix the private region, you can perform the following procedure. Sometimes, this procedure does not work because of an actual hardware problem on the physical disk. If dd'ing to the private region produces errors, or does not change the disk to an online state, it might need to be replaced. 1) From a terminal window enter the following: # vxdisk list The following is an excerpt from the vxdisk list: DEVICE TYPE DISK GROUP STATUS c0t8d0s2 sliced rootdisk rootdg online c0t9d0s2 sliced disk02 rootdg online c1t0d0s2 sliced disk01 rootdg online c2t4d1s2 sliced - - online altused c2t5d0s2 sliced - - online altused 2) Perform the following steps: a) Determine which slice is being used for the private region: # prtvtoc /dev/rdsk/cXtXdXs2 b) Use the vxdisk list to determine which bad disk to review, that is: # prtvtoc /dev/rdsk/c2t4d1s2 Output appears that is similar to the following example: * /dev/rdsk/c2t4d1s2 partition map * * Dimensions: * 512 bytes/sector * 16 sectors/track * 2 tracks/cylinder * 32 sectors/cylinder * 63 cylinders * 61 accessible cylinders * * Flags: * 1: unmountable * 10: read-only * * First Sector Last * Partition Tag Flags Sector Count Sector Mount Directory 2 5 01 0 8907450 8907449 3 15 01 0 1290 1289 4 14 01 1290 8906160 8907449 c) In the Tag column in the example output, find the line for Tag 15 (the tag for the private region) and note the partition that corresponds to it. In the example output, the private region is on partition 3. Record the partition number for the disk for use in future steps. 3) Repeat step 2 for each corrupted disk. 4) dd each disk's private region to a file (naming the file accordingly so that In the following, c2t4d1s2 is used as an example disk and c2t4d1s2_privreg as the filename. Replace these labels with the actual labels for the disk and and the correct private region slice that you are attempting to repair. The filename can be anything that will help you remember the file: # dd if=/dev/rdsk/c2t4d1s3 of=/tmp/c2t4d1s3_privreg 5) Run the following script on the file you created in step 4. (If you save the diskid: 86251073.30247.cradev group: name=ora_dg id=881621654.31353.cradev flags: private autoimport ... ...... logs: count=1 len=167 tocblks: 0 tocs: 2/1533 Defined regions: config priv 000017-000247[000231]: copy=01 offset=000000 enabled config priv 000249-001119[000871]: copy=01 offset=000231 enabled log priv 001120-001286[000167]: copy=01 offset=000000 enabled Note: Look for the line marked "flags." An altused flag set should not appear. If the output is not good, you might also get an error message that is similar to the following: vxvm:vxprivutil: ERROR: join operation failed: Format error in disk private region 6) If the output from the script looks good, dd data from the good file back to # dd if=/tmp/c2t4d1s3_privreg.good of=/dev/rdsk/c2t4d1s3 7) Repeat steps 4 through 6 for each bad disk. 8) Do a "vxdctl enable" and then verify success: vxdctl enable vxdisk list DEVICE TYPE DISK GROUP STATUS c0t8d0s2 sliced rootdisk rootdg online c0t9d0s2 sliced disk02 rootdg online c1t0d0s2 sliced disk01 rootdg online c2t4d1s2 sliced data01 datadg online c2t5d0s2 sliced data02 datadg online 9) If the procedure does not give the expected result, the diskgroup should be deported then imported. If that is not possible, the server should be rebooted. Product VERITAS Volume Manager 3.2 Software Sun Enterprise Volume Manager 2.4 Sun StorageTek A3500 FC Array Sun StorageTek A3500 Array vxdisk list, online altused, online, altused Previously Published As 20907 Change History Date: 2006-05-11 User Name: 29589 Action: Approved Comment: One additional step added to procedure. No other changes. Reviewed. Republish. Version: 6 Date: 2006-05-11 User Name: 29589 Action: Accept Comment: Version: 0 Date: 2006-05-11 User Name: 78962 Action: Approved Comment: The update is fine. Looks good. Version: 0 Attachments This solution has no attachment |
||||||||||||
|