I recently corrupted my linux distro. I then reinstalled Ubuntu 16.04.3 and after setting it up, I have been unable to mount my Raid 5 setup after extensive troubleshooting.

Note even after the original system was corrupted I still had access (the windows explorer still worked) and I know I was still able to access the Raid.

When I try to mount the drive I get the following error:

mount: wrong fs type, bad option, bad superblock on /dev/mapper/isw_cidjefdfai_Volume0,
       missing codepage or helper program, or other error

       In some cases useful info is found in syslog - try
       dmesg | tail or so.

I then did what it instructed:

dmesg | tail

... Lots of other output, then:

[63460.380362] EXT4-fs (dm-0): bad geometry: block count 2783759360 exceeds size of device (1173146784 blocks)

After searching solutions to the bad geometry problem I came across articles saying to run fsck. Here is the output from fsck:

$ sudo fsck /dev/mapper/isw_cidjefdfai_Volume0
fsck from util-linux 2.27.1
e2fsck 1.42.13 (17-May-2015)
Error reading block 1391493120 (Invalid argument).  Ignore error<y>? yes
Force rewrite<y>? no
Superblock has an invalid journal (inode 8).
Clear<y>? no
fsck.ext4: Illegal inode number while checking ext3 journal for DriveName

DriveName: ********** WARNING: Filesystem still has errors **********

Most articles I come across have similar problems after resizing, however I did not resize the drive. Additional info that may be helpful, This raid setup was created with Intel PCH RSTe on a SuperMicro board, which I found is typically referred to as fakeRaid. However when I boot the machine, Intel RSTe finds the raid volume (as well as the 4 drives that make it up) and says the status is "Normal" as well as all 4 drives as members of the Raid volume (picture attached).Bios screen from Intel RSTe I then found that these are usually managed on linux with dmraid so I then ran the following:

$ sudo dmraid -s
*** Group superset isw_cidjefdfai
--> Active Subset
name   : isw_cidjefdfai_Volume0
size   : 9385174272
stride : 128
type   : raid5_la
status : ok
subsets: 0
devs   : 4
spares : 0
$ sudo dmraid -r
/dev/sdc: isw, "isw_cidjefdfai", GROUP, ok, 7814037166 sectors, data@ 0
/dev/sde: isw, "isw_cidjefdfai", GROUP, ok, 7814037166 sectors, data@ 0
/dev/sdd: isw, "isw_cidjefdfai", GROUP, ok, 7814037166 sectors, data@ 0
/dev/sdf: isw, "isw_cidjefdfai", GROUP, ok, 7814037166 sectors, data@ 0

I am going to try and use dd to copy all the information from the drives before attempting something like a rebuild. But I find it odd that both dmraid and Intel think the drive is perfectly fine, but I cannot mount it.

  • You're confusing slightly two unrelated things, one is the physical devices the bios sees, or uefi, the other is partitions with file systems on those devices. Your problem appears to be with the latter. Again, it seems to me that there's no confusion, the issue is with the file system on the raid, not the raid, so intel thinks it's fine. – Lizardx Jan 26 '18 at 1:30
  • @Lizardx, I agree, I think the raid itself is fine, but somehow it thinks the filesystem is more than double the size of the drive (or that the drive is half the size of the filesystem) either way I am not sure how to remedy this. – fjames003 Jan 26 '18 at 1:40
  • I'm surprised nobody has jumped on this one, you constructed your question as well as could be asked, and the problem is obviously real. I added an upvote to the question in hopes of attracting some more attention. I'm not familiar enough with how you did this to help, my understanding of the raid issue is that you let the hardware treat the drives as drives, and you let dmraid create and handle the actual raid activity, but I've never seen the case where the hardware even knows about the raid, and I can't say if that is even correct or incorrect. I would not have done it that way. – Lizardx Jan 26 '18 at 22:23
  • @Lizardx I appreciate you taking the time to look through this. – fjames003 Jan 26 '18 at 23:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.