Scsi reservation reset?

Please have a look at primus emc81931

There are at least 2 different SCSI reservation .. Both can be managed from either sides.

A SCSI reservation is a lock you put on your device. A lock preventing someone to dig or simply change content of your drive. Usually clusters uses reservation to obtain exclusive access to shared resources.

But if your host crashes (just an example), the disk will hold its lock and you need to remove it before being able to use again the disk. That's why you have to reset scsi reservation from time to time.

While removing a "plain" SCSI reservation requires a single symcli command, removing a SCSI3 PGR (persistent group reservation) requires either a PSE and a SR (from storage point of view) or host-specific commands (ask your beloved sysadmins).

from UNIX/host side use #man scsictl


scsictl command can be used to clear persistent reservation from a
device, as shown in the second form above, using the -p option.

STK can you please be more specific ?? Hpux offers a scsictl command but doesn't have "-p" option... And I don't know if other unixes have scsictl command...

Do u HP-UX box to run "man scsictl".. Even though i haven't used that option myself, it should support -p if listed in man page.

http://docs.hp.com/en/B2355-90692/scsictl.1M.html

"-p" does not appear to be a valid option...

Went through this explanation, might be of some help :

¿SCSI-3 PR, which stands for Persistent Reservation, supports multiple nodes accessing a device while at the same time blocking access to other nodes.
SCSI-3 PR reservations are persistent across SCSI bus resets or node reboots and also support multiple paths from host to disk.
For SCSI-2 disks, reservations are not persistent which means they do not survive node reboots.

SCSI-3 PR uses a concept of registration and reservation. Systems that participate, register a key with SCSI-3 device. Each system registers its own key.
Then registered systems can establish a reservation. With this method, blocking write access is as simple as removing registration from a device.
A system wishing to eject another system issues a pre-empt and abort command and that ejects another node.
Once a node is ejected, it has no key registered so that it cannot eject others. This method effectively avoids the split-brain condition. ¿

On the Symmetrix control hosts. do the following

This is done on server side(by Storage Engineer) make a text file SCSI3 in that enter.

set dev 098 attribute=SCSI3_persist_reserve;
set dev 099 attribute=SCSI3_persist_reserve;

(do this for each deivce you want to set SCSI3 flag)

And run the symconfigure command

symconfigure -sid 1956 -f SCSI3 preview

symconfigure -sid 1956 -f SCSI3 prepare

symconfigure -sid 68 -f SCSI3 commit

And if you want to reset scsi do the following:

edit the text file SCSI3

set dev 098 attribute=SCSI3_persist_no_reserve;
set dev 099 attribute=SCSI3_persist_no_reserve;

Run the symconfigure command as above.



This link from Ibm may help you to understand the concept more

http://publib.boulder.ibm.com/infocenter/clresctr/vxrx/index.jsp?topic=/com.ibm.cluster.rsct_5300_06.vsd.doc/bl506_undperr.html

http://seer.entsupport.symantec.com/docs/261824.htm

If you feel I have answered your question Please assign points.

Thanks a lot

sorry for the typo...and misses.

The text file name is SCSI3 on the symconfigure commands.

and the third commit symconfigure command
read it as

symconfigure -sid 1956 -f SCSI3 commit ( -sid of your symmetrix)

Thanks