Locked up FreeNAS box

B

bdt

Executive Member
Joined
Jun 7, 2004
Messages
7,001
Reaction score
823
Location
Mama Randburg, middle of everywhere
Hi all, I'm in a bind and (so far) haven't found a way out of this. I've got a FreeNAS box setup with 2x "250"GB drives in RAID1 that was working perfectly well ..until there was a power cut (and yes, we now have a UPS to bring to the party!)

Anyhow, we're now locked out of the array I've got the following to work with:
Code: 
Status: Disks
Disk 	Size 	Description 	Status
ad4 	238475MB 	WDC WD2500JS-40MVB1/10.02E01  	ONLINE 
ad6 	238475MB 	WDC WD2500JS-19NCB1/10.02E01  	ONLINE 
main 	232 GB 	Software RAID volume  	down

Code: 
Manage RAID:
Volume Name  	Type  	Size  	Status  	
main 	1 	232 GB  	down

Code: 
Information
Software RAID information and status:
2 drives:
D disk_ad4              State: up	/dev/ad4s1a	A: 0/238475 MB (0%)
D disk_ad6              State: up	/dev/ad6s1a	A: 0/238475 MB (0%)

1 volume:
V main                  State: down	Plexes:       2	Size:        232 GB

2 plexes:
P main.p0             C State: down	Subdisks:     1	Size:        232 GB
P main.p1             C State: down	Subdisks:     1	Size:        232 GB

2 subdisks:
S main.p0.s0            State: stale	D: disk_ad4     Size:        232 GB
S main.p1.s0            State: stale	D: disk_ad6     Size:        232 GB

Code: 
System log
Jun 19 17:25:12 	kernel: g_vfs_done():gvinum/main[READ(offset=65536, length=8192)]error = 6
Jun 19 17:25:12 	kernel: g_vfs_done():gvinum/main[READ(offset=65536, length=8192)]error = 6
Jun 19 17:25:11 	kernel: GEOM_VINUM: subdisk main.p0.s0 state change: down -> stale
Jun 19 17:25:11 	kernel: Trying to mount root from ufs:/dev/md0
Jun 19 17:25:11 	kernel: ad6: 238475MB <WDC WD2500JS-19NCB1 10.02E01> at ata3-master SATA150
Jun 19 17:25:11 	kernel: ad4: 238475MB <WDC WD2500JS-40MVB1 10.02E01> at ata2-master SATA150
Jun 19 17:25:11 	kernel: acd0: CDRW <SONY CD-RW CRX220E1/6YS1> at ata1-master UDMA33
Jun 19 17:25:11 	kernel: ad0: 122MB <Hitachi XX.V.3.3.0.0 Rev 0.00> at ata0-master PIO4
Jun 19 17:25:11 	kernel: md0: Preloaded image </mfsroot> 33554432 bytes at 0xc0b24770
Jun 19 17:25:11 	kernel: Timecounters tick every 1.000 msec
Jun 19 17:25:11 	kernel: Timecounter "TSC" frequency 2527028158 Hz quality 800
Jun 19 17:25:11 	kernel: vga0: <Generic ISA VGA> at port 0x3c0-0x3df iomem 0xa0000-0xbffff on isa0

So the bottom line is the actual disks are fine are but the verdomte thing is refusing to give me my RAID volume back! And, if it counts for anything, I've taken the machine through a graceful reboot or two (to confirm on the box that things are actually as they say they are in the remote management) and still no data!

FreeNAS is built off m0n0wall so it's BSD we're talking about here ..anyone got any input for me here? (HUGE please!)
 
Oh dear.

Jun 19 17:25:12 kernel: g_vfs_done():gvinum/main[READ(offset=65536, length=8192)]error = 6
Jun 19 17:25:12 kernel: g_vfs_done():gvinum/main[READ(offset=65536, length=8192)]error = 6
Jun 19 17:25:11 kernel: GEOM_VINUM: subdisk main.p0.s0 state change: down -> stale

That is your problem.

Is there no recovery procedure or howto for dealing with a hosed gvinum install ???.
 
Thanks for your posts tibby. I've tried sending those commands in the two places that allow it and haven't got anywhere yet. I'll give that (somewhat long and intimidating!) .txt a read to see if I can make anything out of it, but my ultimate failure mode position is to rip the drive out, stick it into a 'doze box and hope like hell something like ufstools, UFS Explorer or the like can pull the data off ..at which time I can hose the install and rebuild!
 
AAAARRRGGH! You can't begin to imagine the frustration between the relief of the verk4kte thing working again ..but for no good reason! :mad: I've been up, down & around the 'net, tried reading one of the drives in two other machines ..hell, I've even tried doing Linux liveCD action that all got me nowhere.

So when I finally put it back together in despair (and mentally steel myself for some BigTime(tm) grovelling/emptying of bank account) - it up and just WORKS when I restart! ... *sob*, it's all just too much! :rolleyes: (sometimes I REALLY hate ICT hardware!)

OK, so now the system log is bitching thusly:
Code: 
Jun 20 04:27:17 kernel: WARNING: /mnt/docs was not properly dismounted 
Jun 20 04:27:16 kernel: Trying to mount root from ufs:/dev/md0 
Jun 20 04:27:16 kernel: ad6: 238475MB <WDC WD2500JS-19NCB1 10.02E01> at ata3-master SATA150 
Jun 20 04:27:16 kernel: ad4: 238475MB <WDC WD2500JS-40MVB1 10.02E01> at ata2-master SATA150 
Jun 20 04:27:16 kernel: acd0: CDRW <SONY CD-RW CRX220E1/6YS1> at ata1-master UDMA33 
Jun 20 04:27:16 kernel: ad0: 122MB <Hitachi XX.V.3.3.0.0 Rev 0.00> at ata0-master PIO4 
Jun 20 04:27:16 kernel: md0: Preloaded image </mfsroot> 33554432 bytes at 0xc0b24770
..any sagacious words on how to appease the beast? (as in make it happy about the state of the filesystem again)
 
Last edited:
Ack! (and no, not Ack!Ack!..ACK!, that's something entirely different!) :cool:

Going by:
Code: 
Software RAID status:
1 drive:
D disk_ad4              State: up	/dev/ad4s1a	A: 0/238475 MB (0%)
1 volume:
V main                  State: up	Plexes:       1	Size:        232 GB
2 plexes:
[B]P main.p1           C State: down[/B]   Subdisks:     0 Size:        0  B
P main.p0             C State: up	Subdisks:     1	Size:        232 GB2
subdisks:
[B]S main.p1.s0         State: stale[/B]   D: disk_ad6     Size:        232 GB
S main.p0.s0            State: up	D: disk_ad4     Size:        232 GB
it would appear I'm not out of the woods yet, any BSD gurus/BSD hotshots out there with any help?
 
Hey bdt,

Can you take the stale drive out of service and replace it - it's an old trick on the linux version (raidtools) which I am sure will work the same here...

You need those to re-sync :)

laterz

bdt said:
Ack! (and no, not Ack!Ack!..ACK!, that's something entirely different!) :cool:

Going by:
Code: 
Software RAID status:
1 drive:
D disk_ad4              State: up	/dev/ad4s1a	A: 0/238475 MB (0%)
1 volume:
V main                  State: up	Plexes:       1	Size:        232 GB
2 plexes:
[B]P main.p1           C State: down[/B]   Subdisks:     0 Size:        0  B
P main.p0             C State: up	Subdisks:     1	Size:        232 GB2
subdisks:
[B]S main.p1.s0         State: stale[/B]   D: disk_ad6     Size:        232 GB
S main.p0.s0            State: up	D: disk_ad4     Size:        232 GB
it would appear I'm not out of the woods yet, any BSD gurus/BSD hotshots out there with any help?
Click to expand...
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X