I do not understand the situation arose

[taesun]

Hi Friends

I do not understand the situation arose

All i/o channels active stated.

san-switch device firmware upgrades.

Was a sudden san-switch reboot .

the san-swtich of the i/o channels disable.

the normal state of other channels

For a while, but the service was not 30 second.

I wonder why this happened.

I think, 30 seconds as the service stops, the dmp reforming.

Tell me, what was the other possibility.

thanks

Operating system: HPUX

Operating system version: 11.31

Architecture: ia64

Server model: ia64 hp superdome server SD32B

veritas version : CFS5.0 (vxvm 5.0RP6, vxfs5.0RP6HF1)

Switch : BROCADE SilkWorm 48000

storage : EMC dmx series

syslog messages.

ul 11 14:41:28 eaiap1p vmunix: 0/0/14/1/0: Fibre Channel Driver received Link Dead Notification.

Jul 11 14:41:28 eaiap1p vmunix: 

Jul 11 14:41:28 eaiap1p vmunix: 2/0/14/1/0: Fibre Channel Driver received Link Dead Notification.

Jul 11 14:41:28 eaiap1p vmunix: class : tgtpath, instance 5

Jul 11 14:41:28 eaiap1p vmunix: Target path (class=tgtpath, instance=5) has gone offline.  The target path h/w path is 0/0/14/1/0.0

x5006048449af3675

Jul 11 14:41:28 eaiap1p vmunix: class : tgtpath, instance 4

Jul 11 14:41:28 eaiap1p vmunix: Target path (class=tgtpath, instance=4) has gone offline.  The target path h/w path is 2/0/14/1/0.0

x5006048449af3676

....skip

Jul 11 14:41:31 eaiap1p vmunix: NOTICE: VxVM vxdmp V-5-0-112 disabled path 255/0x1c0 belonging to the dmpnode 5/0xc0

Jul 11 14:41:31 eaiap1p vmunix: NOTICE: VxVM vxdmp V-5-0-112 disabled path 255/0x1d0 belonging to the dmpnode 5/0x90

Jul 11 14:41:31 eaiap1p vmunix: NOTICE: VxVM vxdmp V-5-0-112 disabled path 255/0x190 belonging to the dmpnode 5/0x200

Jul 11 14:41:31 eaiap1p vmunix: NOTICE: VxVM vxdmp V-5-0-112 disabled path 255/0x290 belonging to the dmpnode 5/0x80

Jul 11 14:41:31 eaiap1p vmunix: NOTICE: VxVM vxdmp V-5-0-112 disabled path 255/0x260 belonging to the dmpnode 5/0x1a0

....skip

Jul 11 14:44:57 eaiap1p vmunix: Target path (class=tgtpath, instance=4) has gone online.  The target path h/w path is 2/0/14/1/0.0x

5006048449af3676

Jul 11 14:45:17 eaiap1p vmunix: class : tgtpath, instance 5

Jul 11 14:44:57 eaiap1p vmunix: 

Jul 11 14:45:17 eaiap1p  above message repeats 11 times

Jul 11 14:45:17 eaiap1p vmunix: Target path (class=tgtpath, instance=5) has gone online.  The target path h/w path is 0/0/14/1/0.0x

5006048449af3675

[Marianne · Accepted solution]

I'm not sure if I understand correctly what your question is... Do you have a problem with the length of time that it took to mark the paths as online?

This is normal behaviour and is described in the VxVM Admin Guide. See p. 141 under the topic:
How DMP monitors I/O on paths

(4th paragraph):
The restore kernel thread is woken periodically (typically every 5 minutes) to check the health of the paths, and to resume I/O on paths that have been restored.
As some paths may suffer from intermittent failure, I/O is only resumed on a path if has remained healthy for a given period of time (by default, 5 minutes). DMP can be configured with different policies for checking the paths.

See “Configuring DMP path restoration policies” on page 189.

[Gaurav_S]

I agree with Marianne.... please elaborate on your question... As Marianne said above, if you are looking for time VxDMP took to restore the paths ... then "restored" is the daemon what looks for restoring the paths... Default time assigned for restored to restore the paths is 5 Min .. i.e 300 Sec... you can check the same with.. # vxdmpadm stat restored & default path type to check is "check_disabled" (restored will check health of all disabled paths every 5 mins) This parameter is tuneable however default is recommended since if you lower the time to scan paths, it might increase a bit of load to server.. Gaurav