SCSI disk (#9) not detected

obelix_a · 1. Februar 2016

Hi

not quite sure if it fits in here best, but as a starter I try. Thread may be moved if totally wrong...

I had 8 disks for OMV which I tied together as a RAID5. Some days ago I decided to move to RAID6 by adding another disk. Today the additional disk arrived, I mounted it in the server and hoped to get it recognized. My "outer" OS (windows 10 with HyperV running the OMV) detects the disk with no problems. But OMV cannot see the 9th disk. It only sees the former 8 disks.

To prevent comments like "do not use virtual machine" or something: If it WOULD have to do with it, why are the 8 disks ok and why can I hand in the 9th but it is not detected by OMV?

I know it is more something for debian itself but OMV - but the greatest geeks live around here
thanks
Obelix

ryecoaaron · 1. Februar 2016

I would never say don't use a virtual machine but I would say don't use windows

Are you using a hardware raid controller, windows software raid, or OMV software raid?
If raid controller or windows software raid, you need to grow the filesystem. It is easiest to boot gparted-live and expand the filesystem.
If OMV software raid, what is the output of: fdisk -l and cat /proc/mdstat

obelix_a · 1. Februar 2016

HI

Windows is mandatory because I am running tuners on the bare metal with ARGUS-TV. If argus is someday portable to linux I will switch ASAP
Ok, sorry, very less information in my first post.
I am running OMV software raid, so under physical disks I can see the 8 disks (the new one missing) and in the RAID section I see my /dev/md0 with 8 disks.

fdisk -l gives

Code

Disk /dev/sda: 21.5 GB, 21474836480 bytes
255 heads, 63 sectors/track, 2610 cylinders, total 41943040 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x0001c11e


   Device Boot      Start         End      Blocks   Id  System
/dev/sda1   *        2048    41236479    20617216   83  Linux
/dev/sda2        41236480    41943039      353280    5  Extended
/dev/sda5        41238528    41943039      352256   82  Linux swap / Solaris


Disk /dev/sdb: 6001.2 GB, 6001175126016 bytes
256 heads, 63 sectors/track, 726751 cylinders, total 11721045168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x00000000


   Device Boot      Start         End      Blocks   Id  System
/dev/sdb1               1  4294967295  2147483647+  ee  GPT
Partition 1 does not start on physical sector boundary.


Disk /dev/sdc: 6001.2 GB, 6001175126016 bytes
256 heads, 63 sectors/track, 726751 cylinders, total 11721045168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x00000000


   Device Boot      Start         End      Blocks   Id  System
/dev/sdc1               1  4294967295  2147483647+  ee  GPT
Partition 1 does not start on physical sector boundary.


Disk /dev/sdf: 6001.2 GB, 6001175126016 bytes
256 heads, 63 sectors/track, 726751 cylinders, total 11721045168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x00000000


   Device Boot      Start         End      Blocks   Id  System
/dev/sdf1               1  4294967295  2147483647+  ee  GPT
Partition 1 does not start on physical sector boundary.


Disk /dev/sdd: 6001.2 GB, 6001175126016 bytes
256 heads, 63 sectors/track, 726751 cylinders, total 11721045168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x00000000


   Device Boot      Start         End      Blocks   Id  System
/dev/sdd1               1  4294967295  2147483647+  ee  GPT
Partition 1 does not start on physical sector boundary.


Disk /dev/sdh: 6001.2 GB, 6001175126016 bytes
256 heads, 63 sectors/track, 726751 cylinders, total 11721045168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x00000000


   Device Boot      Start         End      Blocks   Id  System
/dev/sdh1               1  4294967295  2147483647+  ee  GPT
Partition 1 does not start on physical sector boundary.


Disk /dev/sdg: 6001.2 GB, 6001175126016 bytes
256 heads, 63 sectors/track, 726751 cylinders, total 11721045168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x00000000


   Device Boot      Start         End      Blocks   Id  System
/dev/sdg1               1  4294967295  2147483647+  ee  GPT
Partition 1 does not start on physical sector boundary.


Disk /dev/sde: 6001.2 GB, 6001175126016 bytes
256 heads, 63 sectors/track, 726751 cylinders, total 11721045168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x00000000


   Device Boot      Start         End      Blocks   Id  System
/dev/sde1               1  4294967295  2147483647+  ee  GPT
Partition 1 does not start on physical sector boundary.


Disk /dev/sdi: 6001.2 GB, 6001175126016 bytes
256 heads, 63 sectors/track, 726751 cylinders, total 11721045168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
Disk /dev/md0 doesn't contain a valid partition table
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x00000000


   Device Boot      Start         End      Blocks   Id  System
/dev/sdi1               1  4294967295  2147483647+  ee  GPT
Partition 1 does not start on physical sector boundary.


Disk /dev/md0: 42007.3 GB, 42007282057216 bytes
2 heads, 4 sectors/track, -1 cylinders, total 82045472768 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 524288 bytes / 3670016 bytes
Disk identifier: 0x00000000

Alles anzeigen

and cat /proc/mdstat

Code

Personalities : [raid6] [raid5] [raid4]
md0 : active raid5 sdb[0] sdi[7] sdh[6] sdg[5] sdf[4] sde[3] sdd[2] sdc[1]
      41022736384 blocks super 1.2 level 5, 512k chunk, algorithm 2 [8/8] [UUUUUUUU]


unused devices: <none>

ryecoaaron · 1. Februar 2016

Linux (OMV) definitely does not see the drive. This seems like something that needs be changed in hyper-v. Do you have to pass the drive to the VM? And just to warn you, you cannot change from raid-5 to raid-6 in the OMV web interface.

obelix_a · 1. Februar 2016

the change via mdadm should not be too challenging
I hand the disk "in" the same way as all others (if you know HyperV I set the disk offline, add another disk in machine settings and use the physical disk no 9)
On HyperV side ther is nothing I can do in other manner as far as I can see.
I could also not see the disk while booting OMV/debian - so I ask...
any good ideas (TeamViewer possible)

ryecoaaron · 1. Februar 2016

Funny, I actually used to run an OMV fileserver on hyper-v. But it used .vhd files since the host had a hardware raid controller.

Have you powered off the VM? Is OMV running the 3.16 backports kernel? I would try booting systemrescuecd on the VM to see if it recognizes the 9 drives. If it doesn't, I would guess it is a hyper-v config issue somewhere.

obelix_a · 1. Februar 2016

Well, booting gparted made me believe

It seems like the SCSI controller emulated by Windows handles more than 8 drives in the GUI but does not map them into the machine.
Solution:
- add another SCSI controller
- add the 9th disk on the new controller
- power on OMV
this will lead to many messages concerning /dev/sd(X) with letters you would never expect, but I could not find the messages again to have a look at them.

now for the interested ones, here comes how to manually convert RAID5 to RAID6:

to be sure the RAID5 is clean

Code

cat /proc/mdstat

to see all disks are ok

Code

mdadm --detail /dev/md0

add the new disk to the RAID in /dev/md0

Code

mdadm --add /dev/md0 /dev/sdj

be sure the new disk is in and ok

Code

mdadm --detail /dev/md0

and now change RAID level and tell mdadm now to use 9 disks and also make a backup of the RAID5 config

Code

mdadm --grow /dev/md0 --level=6 --raid-devices=9 --backup-file=/root/raid5backup

and now... wait.... wait....... wait................

ryecoaaron · 1. Februar 2016

Glad it is working.

Zitat von obelix_a

this will lead to many messages concerning /dev/sd(X) with letters you would never expect, but I could not find the messages again to have a look at them.

You can view those messages with dmesg.

obelix_a · 1. Februar 2016

argh... I had a look into it but did not find instantly, so I thought its gone...

BUT taking more than 5 seconds to search leads to results

any idea what this means?

Code

[    0.965030] sd 4:0:0:0: [sdj] Very big device. Trying to use READ CAPACITY(16).
[    0.965035] sd 4:0:0:1: [sdk] Sector size 0 reported, assuming 512.
[    0.965037] sd 4:0:0:1: [sdk] 1 512-byte logical blocks: (512 B/512 B)
[    0.965038] sd 4:0:0:1: [sdk] 0-byte physical blocks
[    0.965041] sd 4:0:0:2: [sdl] Sector size 0 reported, assuming 512.
[    0.965043] sd 4:0:0:2: [sdl] 1 512-byte logical blocks: (512 B/512 B)
[    0.965044] sd 4:0:0:2: [sdl] 0-byte physical blocks
[    0.965046] sd 4:0:0:3: [sdm] Sector size 0 reported, assuming 512.
[    0.965048] sd 4:0:0:3: [sdm] 1 512-byte logical blocks: (512 B/512 B)
[    0.965049] sd 4:0:0:3: [sdm] 0-byte physical blocks
[    0.965051] sd 4:0:0:4: [sdn] Sector size 0 reported, assuming 512.
[    0.965052] sd 4:0:0:4: [sdn] 1 512-byte logical blocks: (512 B/512 B)
[    0.965053] sd 4:0:0:4: [sdn] 0-byte physical blocks
[    0.965056] sd 4:0:0:5: [sdo] Sector size 0 reported, assuming 512.
[    0.965057] sd 4:0:0:5: [sdo] 1 512-byte logical blocks: (512 B/512 B)
[    0.965058] sd 4:0:0:5: [sdo] 0-byte physical blocks
[    0.965061] sd 4:0:0:6: [sdp] Sector size 0 reported, assuming 512.
[    0.965062] sd 4:0:0:6: [sdp] 1 512-byte logical blocks: (512 B/512 B)
[    0.965063] sd 4:0:0:6: [sdp] 0-byte physical blocks
[    0.965065] sd 4:0:0:7: [sdq] Sector size 0 reported, assuming 512.
[    0.965067] sd 4:0:0:7: [sdq] 1 512-byte logical blocks: (512 B/512 B)
[    0.965068] sd 4:0:0:7: [sdq] 0-byte physical blocks
[    0.965278] sd 4:0:0:0: [sdj] 11721045168 512-byte logical blocks: (6.00 TB/5.45 TiB)
[    0.965280] sd 4:0:0:0: [sdj] 4096-byte physical blocks
[    0.965285] sd 4:0:0:1: [sdk] Write Protect is off
[    0.965286] sd 4:0:0:1: [sdk] Mode Sense: 00 00 00 00
[    0.965289] sd 4:0:0:2: [sdl] Write Protect is off
[    0.965290] sd 4:0:0:2: [sdl] Mode Sense: 00 00 00 00
[    0.965293] sd 4:0:0:3: [sdm] Write Protect is off
[    0.965294] sd 4:0:0:3: [sdm] Mode Sense: 00 00 00 00
[    0.965296] sd 4:0:0:4: [sdn] Write Protect is off
[    0.965297] sd 4:0:0:4: [sdn] Mode Sense: 00 00 00 00
[    0.965299] sd 4:0:0:5: [sdo] Write Protect is off
[    0.965300] sd 4:0:0:5: [sdo] Mode Sense: 00 00 00 00
[    0.965303] sd 4:0:0:6: [sdp] Write Protect is off
[    0.965304] sd 4:0:0:6: [sdp] Mode Sense: 00 00 00 00
[    0.965306] sd 4:0:0:7: [sdq] Write Protect is off
[    0.965307] sd 4:0:0:7: [sdq] Mode Sense: 00 00 00 00
[    0.965502] sd 3:0:0:3: [sde] Very big device. Trying to use READ CAPACITY(16).
[    0.966730] sd 3:0:0:2: [sdd] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[    0.966888] sd 4:0:0:6: [sdp] Asking for cache data failed
[    0.968103] sd 4:0:0:6: [sdp] Assuming drive cache: write through
[    0.968548] sd 4:0:0:7: [sdq] Asking for cache data failed
[    0.968753] sd 4:0:0:7: [sdq] Assuming drive cache: write through
[    0.968898] sd 4:0:0:5: [sdo] Asking for cache data failed
[    0.970024] sd 4:0:0:5: [sdo] Assuming drive cache: write through
[    0.970328] sd 3:0:0:7: [sdi] Very big device. Trying to use READ CAPACITY(16).
[    0.970332] sd 3:0:0:6: [sdh] Very big device. Trying to use READ CAPACITY(16).
[    0.970335] sd 3:0:0:5: [sdg] Very big device. Trying to use READ CAPACITY(16).
[    0.970339] sd 3:0:0:4: [sdf] Very big device. Trying to use READ CAPACITY(16).
[    0.970343] sd 4:0:0:4: [sdn] Asking for cache data failed
[    0.970490] sd 4:0:0:4: [sdn] Assuming drive cache: write through
[    0.970652] sd 4:0:0:3: [sdm] Asking for cache data failed
[    0.970785] sd 4:0:0:3: [sdm] Assuming drive cache: write through
[    0.970985] sd 4:0:0:2: [sdl] Asking for cache data failed
[    0.971313] sd 4:0:0:2: [sdl] Assuming drive cache: write through
[    0.971657] sd 4:0:0:1: [sdk] Asking for cache data failed
[    0.971990] sd 4:0:0:1: [sdk] Assuming drive cache: write through
[    0.972384] sd 3:0:0:3: [sde] 11721045168 512-byte logical blocks: (6.00 TB/5.45 TiB)
[    0.972386] sd 3:0:0:3: [sde] 4096-byte physical blocks
[    0.972399] sd 3:0:0:5: [sdg] 11721045168 512-byte logical blocks: (6.00 TB/5.45 TiB)

Alles anzeigen

ryecoaaron · 1. Februar 2016

I don't see any issues other then maybe performance. Hyper-v must pass all drives to the VM as 512 byte sectors instead of 4k sectors. I would definitely upgrade to the backports 3.16 kernel if you haven't already.

obelix_a · 1. Februar 2016

...on my way to backports kernel...

should I restart the machine while RAID is reshaping?

ryecoaaron · 2. Februar 2016

Zitat von obelix_a

should I restart the machine while RAID is reshaping?

Nope.

Jetzt mitmachen!