Failed Disk multiple PD's referencing same Mag location
Posted: Wed Sep 10, 2014 1:08 pm
Trying to figure out how to recover from this, first time I see this happen.
Had a disk fail PD91
ran servicemag start etc... all well
Replaced disk and ran resume.
Resume failed due to replaced disk failed in when relocating chunklets from spare space.
Now I have two PD's with chunklets reference same location 2:2:3
I have PD91 and the new PD141 that was to replace PD91. I can't dismisspd 91 since when I do I get an error that a chunklet is referenced
If you try rerunning sericemag start you get an error, thinks to disks on a sled cause PD91 and 141 are on POS 3 on MAG.
servicemag start -log -wait -pdid 141
Are you sure you want to run servicemag?
select q=quit y=yes n=no: y
servicemag start -log -wait -pdid 141
... servicing disks in mag: 2 2
... normal disks: WWN [5000C50019B03CD7] Id [89] diskpos [1]
.................... WWN [5000C50010C719A3] Id [90] diskpos [2]
.................... WWN [5000C50010C8DA18] Id [112] diskpos [0]
... not normal disks: WWN [5000C50019B74022] Id [91]
.................... WWN [5000C5001997DB2F] Id [141] diskpos [3]
... relocating chunklets to spare space...
... relocating chunklets from degraded raid sets to spare space
... logging chunklets from pd WWN [5000C50019B03CD7] Id [89]
... logging chunklets from pd WWN [5000C50010C719A3] Id [90]
... logging chunklets from pd WWN [5000C50010C8DA18] Id [112]
2014-09-09 18:01:14.00 CDT {32107} Assertion failed: m->psvc_count > 0 && m->psvc_count <= MAX_DISKS
_PER_SLED
PANIC [pid: 32107, thread: 32107]: Assertion point: file: cmd_servicesled.c, line: 3032
PD91 shows no chunklets but will error when trying to dismisspd
showpdch 91
Pdid Chnk LdName LdCh State Usage Media Sp Cl From To
-----------------------------------------------------
Total chunklets: 0
dismisspd 91
Error : Pd id 91 is referenced by chunklet 0:124
Any Ideas ?
Had a disk fail PD91
ran servicemag start etc... all well
Replaced disk and ran resume.
Resume failed due to replaced disk failed in when relocating chunklets from spare space.
Now I have two PD's with chunklets reference same location 2:2:3
I have PD91 and the new PD141 that was to replace PD91. I can't dismisspd 91 since when I do I get an error that a chunklet is referenced
If you try rerunning sericemag start you get an error, thinks to disks on a sled cause PD91 and 141 are on POS 3 on MAG.
servicemag start -log -wait -pdid 141
Are you sure you want to run servicemag?
select q=quit y=yes n=no: y
servicemag start -log -wait -pdid 141
... servicing disks in mag: 2 2
... normal disks: WWN [5000C50019B03CD7] Id [89] diskpos [1]
.................... WWN [5000C50010C719A3] Id [90] diskpos [2]
.................... WWN [5000C50010C8DA18] Id [112] diskpos [0]
... not normal disks: WWN [5000C50019B74022] Id [91]
.................... WWN [5000C5001997DB2F] Id [141] diskpos [3]
... relocating chunklets to spare space...
... relocating chunklets from degraded raid sets to spare space
... logging chunklets from pd WWN [5000C50019B03CD7] Id [89]
... logging chunklets from pd WWN [5000C50010C719A3] Id [90]
... logging chunklets from pd WWN [5000C50010C8DA18] Id [112]
2014-09-09 18:01:14.00 CDT {32107} Assertion failed: m->psvc_count > 0 && m->psvc_count <= MAX_DISKS
_PER_SLED
PANIC [pid: 32107, thread: 32107]: Assertion point: file: cmd_servicesled.c, line: 3032
PD91 shows no chunklets but will error when trying to dismisspd
showpdch 91
Pdid Chnk LdName LdCh State Usage Media Sp Cl From To
-----------------------------------------------------
Total chunklets: 0
dismisspd 91
Error : Pd id 91 is referenced by chunklet 0:124
Any Ideas ?