fsck.reiserfs run manually
Hi all, discovered a bad hard disk, and now the system won't boot fully. Instead I get message : fsck.reiserf /dev/hdb3 failed (status 0x8) run manually! for each of the hard drive partitions (/dev/hdb1, /dev/hdb2 and /dev/hdb3) and a message at the end saying fsck failed. Repair manually and reboot. The root filesystem is currently mounted read-only. ... Anyways, I try mounting the partitions, but get /dev/hdb1 is not a valid block device. I saw something about vgcfgrestore, which suggested that there should be some files about partitions under /etc/lvmconf, but don't see that directory in my Suse 9.0 system. Any suggestions? Bruce
On Sunday 14 November 2004 11:31, bruce@scharlau.co.uk wrote:
Hi all,
discovered a bad hard disk, and now the system won't boot fully. Instead I get message :
fsck.reiserf /dev/hdb3 failed (status 0x8) run manually!
for each of the hard drive partitions (/dev/hdb1, /dev/hdb2 and /dev/hdb3) and a message at the end saying
fsck failed. Repair manually and reboot. The root filesystem is currently mounted read-only. ...
Anyways, I try mounting the partitions, but get /dev/hdb1 is not a valid block device.
I saw something about vgcfgrestore, which suggested that there should be some files about partitions under /etc/lvmconf, but don't see that directory in my Suse 9.0 system.
Any suggestions?
Bruce
Try: fsck.reiserf /dev/hdb1 and so forth for each hdb 2 and hdb3. Here's some details: I had a similar problem. I'm not sure this is exactly what you're describing but the instruction to run fsck manually is the same. Here's what I did to recover: I went through everything carefully and repeatedly to try to repair the problem to see what was wrong. I went through all the options using the DVD install disk. Repair could not restore ext2 in my situation and said so advising a manual repair without the -p or -a options. Root address was corrupted, and I could to get to it. I used the installation disk to boot initially then used its boot from the hard disk option. This put me into root ready to enter fsck manually which seems to be where you are above according to the message you got. I then entered: fsck.ext2 /dev/hda4 (for my configuration) I then just answered "yes" to each question, and it repaired with no further problems. Booted perfectly after that. This was the result of a crash that had unusually occurred for me. I have just started using 9.2. Other than this minor problem, 9.2 is working fine for me. I like it so far. Gary
On Sunday 14 November 2004 11:31, bruce@scharlau.co.uk wrote:
Hi all,
discovered a bad hard disk, and now the system won't boot fully. Instead I get message :
fsck.reiserf /dev/hdb3 failed (status 0x8) run manually!
for each of the hard drive partitions (/dev/hdb1, /dev/hdb2 and /dev/hdb3) and a message at the end saying
fsck failed. Repair manually and reboot. The root filesystem is currently mounted read-only. ...
Anyways, I try mounting the partitions, but get /dev/hdb1 is not a valid block device.
I saw something about vgcfgrestore, which suggested that there should be some files about partitions under /etc/lvmconf, but don't see that directory in my Suse 9.0 system.
Any suggestions?
Bruce
Try:
fsck.reiserf /dev/hdb1
and so forth for each hdb 2 and hdb3. Here's some details:
I had a similar problem. I'm not sure this is exactly what you're describing but the instruction to run fsck manually is the same. Here's what I did to recover:
I went through everything carefully and repeatedly to try to repair the problem to see what was wrong. I went through all the options using the DVD install disk. Repair could not restore ext2 in my situation and said so advising a manual repair without the -p or -a options.
Root address was corrupted, and I could to get to it. I used the installation disk to boot initially then used its boot from the hard disk option. This put me into root ready to enter fsck manually which seems to be where you are above according to the message you got. I then entered:
fsck.ext2 /dev/hda4 (for my configuration)
I then just answered "yes" to each question, and it repaired with no further problems. Booted perfectly after that. This was the result of a crash that had unusually occurred for me. I have just started using 9.2. Other than this minor problem, 9.2 is working fine for me. I like it so far.
Gary
Gary, I tried that, but still got the 'not a valid block device' message. I also tried booting the original boot disk that I made for my ftp installation to see if it would help me out, but no luck there either. Anyone else have any ideas about how to sort this out, or I am going to have to just reinstall everything onto a new hard disk and kiss my old files goodbye? cheers, Bruce
bruce@scharlau.co.uk wrote:
On Sunday 14 November 2004 11:31, bruce@scharlau.co.uk wrote:
Hi all,
discovered a bad hard disk, and now the system won't boot fully. Instead I get message :
fsck.reiserf /dev/hdb3 failed (status 0x8) run manually!
for each of the hard drive partitions (/dev/hdb1, /dev/hdb2 and /dev/hdb3) and a message at the end saying
fsck failed. Repair manually and reboot. The root filesystem is currently mounted read-only. ...
Anyways, I try mounting the partitions, but get /dev/hdb1 is not a valid block device.
[snipped intermediate stuff]
Anyone else have any ideas about how to sort this out, or I am going to have to just reinstall everything onto a new hard disk and kiss my old files goodbye?
Just to clarify: * You can mount and run /dev/hda without problem. * Any partition on /dev/hdb that you try to access cannot be reached. * 'resierfsck /dev/hdb3' (and so on) all return a bad status. Well, I'd say your hard disk is stuffed. Just a couple of things you might want to try before chucking it: 1. Replace the IDE cable with a new one. 2. If /dev/hdb is on the second IDE controller, try moving it onto the first controller. I just had a /dev/hdb disk that wouldh't mount. I have /dev/hda on IDE1 and /dev/hdb on IDE2 and I shifted /dev/hdb onto IDE1 (as a slave) and it works again. Whether the problem was the cable or the mobo I'm not sure... I'll probably replace the mobo soon anyway. Hope that helps jalal -- GPG fingerprint = 3D45 5509 D380 26A4 523E A9D8 A66A 5F38 CA43 BB0E
jalal wrote:
bruce@scharlau.co.uk wrote:
On Sunday 14 November 2004 11:31, bruce@scharlau.co.uk wrote:
Hi all,
discovered a bad hard disk, and now the system won't boot fully. Instead I get message :
fsck.reiserf /dev/hdb3 failed (status 0x8) run manually!
for each of the hard drive partitions (/dev/hdb1, /dev/hdb2 and /dev/hdb3) and a message at the end saying
fsck failed. Repair manually and reboot. The root filesystem is currently mounted read-only. ...
Anyways, I try mounting the partitions, but get /dev/hdb1 is not a valid block device.
[snipped intermediate stuff]
Anyone else have any ideas about how to sort this out, or I am going to have to just reinstall everything onto a new hard disk and kiss my old files goodbye?
Just to clarify: * You can mount and run /dev/hda without problem. * Any partition on /dev/hdb that you try to access cannot be reached. * 'resierfsck /dev/hdb3' (and so on) all return a bad status.
Well, I'd say your hard disk is stuffed.
Just a couple of things you might want to try before chucking it: 1. Replace the IDE cable with a new one. 2. If /dev/hdb is on the second IDE controller, try moving it onto the first controller.
I just had a /dev/hdb disk that wouldh't mount. I have /dev/hda on IDE1 and /dev/hdb on IDE2 and I shifted /dev/hdb onto IDE1 (as a slave) and it works again. Whether the problem was the cable or the mobo I'm not sure... I'll probably replace the mobo soon anyway.
Hope that helps jalal
I installed a new A7N8X-E mobo in this box and the same problems with HDA. I discovered that IDE1 had gone bad and there was no way of recovering the disk, I couldn't mount it, reiserfsck, debugreiserfs all failed. I had to change the motherboard and rebuild SuSE. It wasn't immediately obvious, it was only when it failed completely and I could only boot from the CD if it was on IDE2 that I realised what the problem was. Regards Sid. -- Sid Boyce .... Hamradio G3VBV and keen Flyer =====LINUX ONLY USED HERE=====
On Monday 15 November 2004 03:37 am, jalal wrote:
I just had a /dev/hdb disk that wouldh't mount. I have /dev/hda on IDE1 and /dev/hdb on IDE2 and I shifted /dev/hdb onto IDE1 (as a slave) and it works again. Whether the problem was the cable or the mobo I'm not sure... I'll probably replace the mobo soon anyway.
I must be misunderstanding something. I thought /dev/hdb is *always* the second device (slave) on the first IDE controller. If a drive is on the second IDE controller (master), I thought it was /dev/hdc and /dev/hdd as a slave (even with no device on the first IDE controller slave- postition. If that is the case, you wouldn't find /dev/hdb on the second IDE controller. hd...a,b,c,d,e,f.etc shows where the drive is, not which drive 'number' it is. But maybe I'm misunderstanding something. Doug
Doug B wrote:
On Monday 15 November 2004 03:37 am, jalal wrote:
I just had a /dev/hdb disk that wouldh't mount. I have /dev/hda on IDE1 and /dev/hdb on IDE2 and I shifted /dev/hdb onto IDE1 (as a slave) and it works again. Whether the problem was the cable or the mobo I'm not sure... I'll probably replace the mobo soon anyway.
I must be misunderstanding something.
I thought /dev/hdb is *always* the second device (slave) on the first IDE controller.
If a drive is on the second IDE controller (master), I thought it was /dev/hdc and /dev/hdd as a slave (even with no device on the first IDE controller slave- postition.
If that is the case, you wouldn't find /dev/hdb on the second IDE controller.
hd...a,b,c,d,e,f.etc shows where the drive is, not which drive 'number' it is.
But maybe I'm misunderstanding something.
Doug
No you're right... originally the disk was /dev/hdc, then I moved it to /dev/hdb and it worked correctly. cheers jalal -- GPG fingerprint = 3D45 5509 D380 26A4 523E A9D8 A66A 5F38 CA43 BB0E
participants (5)
-
bruce@scharlau.co.uk
-
Doug B
-
jalal
-
Sid Boyce
-
Tiscali Gary D Chance