Hi list Today my dear server had some problems.... I´ve arrived at work, and the server had hanged, after a reboot (and a loooong wait while system was checking MD0) I was able to "destroy" the RAID (hda3 + hdc3) change fstab to mount / just with hda3.... after the system was running, i´ve tried fsck.ext3 an hdc3, and.... server crashed after another reboot, system was unable to mount hdd too.... Well, everything points to hardware failure... but... 2 discs at once???? I´m afraid of IDE controller error, and after a google harvest, have found some people with the same error messages, but apparently with config problems... Or, I hope... cable. Any clues??? Nov 30 18:35:59 samba kernel: hdc: status timeout: status=0x80 { Busy } Nov 30 18:35:59 samba kernel: hdc: DMA disabled Nov 30 18:35:59 samba kernel: hdd: DMA disabled Nov 30 18:35:59 samba kernel: hdc: drive not ready for command Nov 30 18:36:01 samba kernel: ide1: reset: success ..... Dec 1 07:11:16 samba kernel: Cannot find map file. Dec 1 07:11:16 samba kernel: hdd1: bad access: block=2, count=2 Dec 1 07:11:16 samba kernel: end_request: I/O error, dev 16:41 (hdd), sector 2 Dec 1 07:11:16 samba kernel: EXT3-fs: unable to read superblock Dec 1 07:11:16 samba kernel: EXT3-fs warning: checktime reached, running e2fsck is recommended Dec 1 07:11:16 samba kernel: hdd: status error: status=0x00 { } Dec 1 07:11:16 samba kernel: hdd: drive not ready for command Dec 1 07:11:16 samba kernel: hdd: status error: status=0x00 { } Dec 1 07:11:16 samba kernel: hdd: drive not ready for command Joao Marka Calçados Jacob S/A
On Thu, 1 Dec 2005, joao marka wrote:
Hi list
Today my dear server had some problems.... I�ve arrived at work, and the server had hanged, after a reboot (and a loooong wait while system was checking MD0) I was able to "destroy" the RAID (hda3 + hdc3) change fstab to mount / just with hda3....
after the system was running, i�ve tried fsck.ext3 an hdc3, and.... server crashed after another reboot, system was unable to mount hdd too....
Well, everything points to hardware failure... but... 2 discs at once???? I�m afraid of IDE controller error, and after a google harvest, have found some people with the same error messages, but apparently with config problems... Or, I hope... cable.
Run a test with smartmon tools: smartctl -t short /dev/hda (there are other tests). For peace of mind you might prefer to take the disks out and do the test in another machine. I recommend booting off CD, especially if you do it in the same machine. I believe that Knoppox (4.0 is current last I looked) has the nexessary tools. At least one new drive is indicated regardless: it's safer recovering if you do it on a duplicate, and there's no telling what corruption you have now. I have had success copying a drive with DD, even when the target was larger, but I use ext2/ext3i* filesystems, and that may make a difference. * I saw your messages re EXT3, but I'd hate someone to assume that it will necessarily work with other fileystems: it might, but I don't have the experience.
Any clues???
Nov 30 18:35:59 samba kernel: hdc: status timeout: status=0x80 { Busy } Nov 30 18:35:59 samba kernel: hdc: DMA disabled Nov 30 18:35:59 samba kernel: hdd: DMA disabled Nov 30 18:35:59 samba kernel: hdc: drive not ready for command Nov 30 18:36:01 samba kernel: ide1: reset: success ..... Dec 1 07:11:16 samba kernel: Cannot find map file. Dec 1 07:11:16 samba kernel: hdd1: bad access: block=2, count=2 Dec 1 07:11:16 samba kernel: end_request: I/O error, dev 16:41 (hdd), sector 2 Dec 1 07:11:16 samba kernel: EXT3-fs: unable to read superblock Dec 1 07:11:16 samba kernel: EXT3-fs warning: checktime reached, running e2fsck is recommended Dec 1 07:11:16 samba kernel: hdd: status error: status=0x00 { } Dec 1 07:11:16 samba kernel: hdd: drive not ready for command Dec 1 07:11:16 samba kernel: hdd: status error: status=0x00 { } Dec 1 07:11:16 samba kernel: hdd: drive not ready for command
Joao Marka Cal�ados Jacob S/A
On Thu, 2005-12-01 at 15:21 -0200, joao marka wrote: [snip]
after the system was running, i´ve tried fsck.ext3 an hdc3, and.... server crashed after another reboot, system was unable to mount hdd too....
Well, everything points to hardware failure... but... 2 discs at once????
Depends on a lot of factors, you can have two discs fail simultaneously if your server suffered a strong enough power surge (a UPS doesn't always protect you, especially if it's a line-interactive). If it is a Maxtor disc, the firmware might have gotten damaged (their firmware isn't particularly stable and/or robust, in my experience). Controller/motherboard might have gotten damaged too - this seems to happen more often with power irregularities. Test the discs in another box, you'll be able to see if the problem is with the discs or not. Hans
On Mon, 05 Dec 2005 10:49:05 +0200, you wrote:
On Thu, 2005-12-01 at 15:21 -0200, joao marka wrote: [snip]
after the system was running, i´ve tried fsck.ext3 an hdc3, and.... server crashed after another reboot, system was unable to mount hdd too....
Well, everything points to hardware failure... but... 2 discs at once????
Depends on a lot of factors, you can have two discs fail simultaneously if your server suffered a strong enough power surge (a UPS doesn't always protect you, especially if it's a line-interactive).
If it is a Maxtor disc, the firmware might have gotten damaged (their firmware isn't particularly stable and/or robust, in my experience).
Controller/motherboard might have gotten damaged too - this seems to happen more often with power irregularities.
Test the discs in another box, you'll be able to see if the problem is with the discs or not.
Hans
Just a P.S. Maxtor disks have been dying faster and faster here - I've gone thru a few dozen warranty replacements in the last 2 years. I don't use them anymore. If you've got Maxtors, I'd say it was quite possible for 2 of them to go at the same time. Mike- -- Mornings: Evolution in action. Only the grumpy will survive. -- Please note - Due to the intense volume of spam, we have installed site-wide spam filters at catherders.com. If email from you bounces, try non-HTML, non-encoded, non-attachments.
On Mon, 2005-12-05 at 07:56 -0500, Michael W Cocke wrote:
Just a P.S. Maxtor disks have been dying faster and faster here - I've gone thru a few dozen warranty replacements in the last 2 years. I don't use them anymore. If you've got Maxtors, I'd say it was quite possible for 2 of them to go at the same time.
I'll second that. I've had so many returns on Maxtor discs, that the brand as a whole became a loss. Same with Fujitsu, until their IDE discs kinda disappeared from the market. These days I stick to Seagate - in the last three years I've sold hundreds of discs, two of them returned within the first few days (was problematic from the beginning), four more were damaged in power surges (but still fully readable with no data corruption, wile maxtor discs in those same surges were not recoverable by mere mortals like me). I think six returns in two years is not too bad. Hans
participants (4)
-
H du Plooy -
joao marka -
John Summerfield -
Michael W Cocke