| Summary: | sys-fs/reiserfsprogs - reiserfsck crashes system when using option --adjust-size | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Navid Zamani <navid.zamani> |
| Component: | [OLD] Core system | Assignee: | Gentoo Kernel Bug Wranglers and Kernel Maintainers <kernel> |
| Status: | RESOLVED NEEDINFO | ||
| Severity: | critical | ||
| Priority: | High | ||
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: |
.config for kernel 2.6.20-hardened-r6
Hardware-Information |
||
|
Description
Navid Zamani
2007-09-04 16:41:56 UTC
the first thing to do is to actually upgrade to the latest kernel and see if it works there This actually is the latest kernel... for the hardened profile. I even tried the latest "unstable (~x86)" kernel (2.6.22-r3 if i remember it right), and could not boot with it. Additionally i have a fear of destroying even more data by producing another crash while rebuilding the fs. :( It's a production server. A test-system would be better suited for this kind of tests. I had a hope that someone other could run the exact commandlines with my kernel and see if it crashes too. Then try the newest kernel and see if it runs. If yes he could test the lines on that kernel too. This would help locate the source of the problem. Eg. Is it a hardware problem, a kernel problem, a reiserfsprogs problem or some other software/strange config in my system. When you say it crashes the kernel, do you mean you see an oops/call trace? If so, can you post that here? Can you post your dmesg output from the time of the crash? Any info you can give us will help. Please also post your kernel .config. Created attachment 131230 [details]
.config for kernel 2.6.20-hardened-r6
(In reply to comment #3) > When you say it crashes the kernel, do you mean you see an oops/call trace? > If so, can you post that here? Can you post your dmesg output from the time > of the crash? Nope. At the moment where i switch on the monitor, i only get a black screen. That's it. And as i said there is no network-connection possible. So the only choice left is to hard-reset the machine. Then i look at the log files and there's no error message. Just the usual entries like "temperature is ok" "smart is fine" "a mail came in". Then the first bootup message follows. So there is no unusual dmesg-ouput too. I can post it the next time, but i don't think there will be a next time, because - as you may understand - i will not "repair" (read destroy) the FS even more by running reiserfsck another time. > Any info you can give us will help. Please also post your kernel .config. Ok. It's attached now. Oh, i found something interesting: While writing to a DVD-RAM on teh same SATA-controller via pktdvd i get the following errors several times:
Sep 14 01:54:52 [kernel] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
Sep 14 01:54:52 [kernel] ata1.00: cmd a0/01:00:00:00:00/00:00:00:00:00/a0 tag 0 cdb 0x5a data 12 in
Sep 14 01:54:52 [kernel] res 40/00:03:00:00:00/00:00:00:00:00/a0 Emask 0x4 (timeout)
Sep 14 01:54:53 [kernel] ata1: soft resetting port
Sep 14 01:55:00 [kernel] ata1: port is slow to respond, please be patient (Status 0xf8)
Sep 14 01:55:23 [kernel] ata1: port failed to respond (30 secs, Status 0xf8)
Sep 14 01:55:23 [kernel] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
Sep 14 01:55:23 [kernel] ATA: abnormal status 0xF8 on port 0xC88DC087
- Last output repeated 5 times -
Sep 14 01:55:53 [kernel] ata1.00: qc timeout (cmd 0xa1)
Sep 14 01:55:53 [kernel] ata1.00: failed to IDENTIFY (I/O error, err_mask=0x4)
Sep 14 01:55:53 [kernel] ata1.00: revalidation failed (errno=-5)
Sep 14 01:55:53 [kernel] ata1: failed to recover some devices, retrying in 5 secs
Sep 14 01:55:58 [kernel] ata1: hard resetting port
Sep 14 01:55:58 [kernel] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
Sep 14 01:55:59 [kernel] ata1.00: configured for UDMA/33
Sep 14 01:55:59 [kernel] ata1: EH complete
It happens sporadically, but is completely reproducible and the messages always are exactly the same. I just do a
growisofs -Z /dev/sr0=/somepath/someimage.iso
Additionally the formatting of a DVD+RW is not possible. It causes the same errors and then the program exits in them mittle of the formatting.
So there is a possibility that it's a problem with the sata-controller's driver . (I heard they still are experimental so i guess such a bug report would be appreciated there.)
Created attachment 131231 [details]
Hardware-Information
By the way... some infos on the hardware, created with lshw.
Can you confirm that the error you get when writing to a DVD-RAM is reproducible on the latest development kernel (2.6.23-rc9 as of this writing)? If so, please post the kernel .config and the complete dmesg output, including the error(s). Please reopen if/when you can provide the info requested in comment #8. |
