SAN relocation task

From wiki
Revision as of 15:44, 25 January 2018 by Rf (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Introduction

During Dec 2017 and Jan 2018, Ally Kirk rang up to say IT Services were doing some relocation on their SAN which is where marvin's STORAGE volume is located. The whole process is transparent, but it would be useful to have been supervise thigns operationally, just in case anything went wrong.

Tasks

Pretty similar to any changes that may occur on this volume:

  • scan SCSI
  • use multipath -ll to see what need device some up.

Read-outs prior to change-over

root@marvin:/storage/home/users/ramon > multipath -ll
mpathd (36000d3100051f800000000000000016e) dm-6 COMPELNT,Compellent Vol
size=62T features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=1 status=active
 |- 6:0:0:1 sdd 8:48  active ready running
 |- 7:0:6:1 sdf 8:80  active ready running
 |- 6:0:1:1 sde 8:64  active ready running
 |- 7:0:7:1 sdg 8:96  active ready running
 |- 6:0:4:1 sdh 8:112 active ready running
 |- 7:0:1:1 sdk 8:160 active ready running
 |- 6:0:5:1 sdi 8:128 active ready running
 `- 7:0:0:1 sdj 8:144 active ready running
root@marvin:/storage/home/users/ramon > multipath -l
mpathd (36000d3100051f800000000000000016e) dm-6 COMPELNT,Compellent Vol
size=62T features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=0 status=active
 |- 6:0:0:1 sdd 8:48  active undef unknown
 |- 7:0:6:1 sdf 8:80  active undef unknown
 |- 6:0:1:1 sde 8:64  active undef unknown
 |- 7:0:7:1 sdg 8:96  active undef unknown
 |- 6:0:4:1 sdh 8:112 active undef unknown
 |- 7:0:1:1 sdk 8:160 active undef unknown
 |- 6:0:5:1 sdi 8:128 active undef unknown
 `- 7:0:0:1 sdj 8:144 active undef unknown

Things to check

How are the devices looking? Go to

/sys/block/sdg/device

for sdg (similarly for sdd, sde, sdf)

However, this exercise didn't really reveal anything

After the change-over

mpathd (36000d3100051f800000000000000016e) dm-6 COMPELNT,Compellent Vol
size=62T features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=1 status=active
 |- 6:0:0:1 sdd 8:48  failed faulty running
 |- 7:0:6:1 sdf 8:80  failed faulty running
 |- 6:0:1:1 sde 8:64  failed faulty running
 |- 7:0:7:1 sdg 8:96  failed faulty running
 |- 6:0:4:1 sdh 8:112 active ready running
 |- 7:0:1:1 sdk 8:160 active ready running
 |- 6:0:5:1 sdi 8:128 active ready running
 `- 7:0:0:1 sdj 8:144 active ready running

Because all eight of these devices were

active ready running

previously, I wrote to Ally with my understanding of the process:

"SCSI id numbers coincide with previous scan output, new device names. Though a bit clueless about the process, it looks to me as though both sets of paths (4 devices each) were working simultaneously, and what happened was, in the changeover, the first set was pulled and we're seeing the second set stand on its own two feet (so to speak)."

He verified this, so that "failed faulty running" is entirely expected, as the old SAN multipaths are withdrawn.