Skip to contentSkip to conversation

XFS _apparent_ corruption: "DATA POINT" (worked around); 2.6.15.4-biglowmem

DISCUSSION Linda Walsh 1 participant 1 message Mar 5, 2006

lore
Share & tools
lore
#0 lore raw

Running 2.6.15.4 with the "biglowmem" patch (to allow using last 128M of 1G address space w/o calling it HIGHMEM, and using a 3+1G memory split.

System has been _stable_: uptime was 20days+11:04.

I tried doing an 'ls' of a directory and my system hung -- no panic, no message. Had been doing compiles/tests on same disk w/no problems (~26G used, 94G total, 68G free).

* Rebooted, went back to same dir -- hung again. * Rebooted, unmounted partition

Quoted 1 line

xfs_check claimed a journal needed to play.

 *  Remounted partition -- no problem; unmount;
 >  xfs_check -- claimed journal present
 >  xfs_repair -- claimed journal present
*>  remount & unmount; xfs_repair still sees journal;

* xfs_logprint gave:
----
ls ->hang
fs_logprint:
xfs_logprint: /dev/hde1 contains a mounted and writable filesystem
data device: 0x2101
log device: 0x2101 daddr: 100663328 length: 95392

Header 0x3ef wanted 0xfeedbabe
**********************************************************************
* ERROR: header cycle=1007 block=38747 *
**********************************************************************
Bad log record header
--------
* Decide to delete bad log: run xfs_repair -L /dev/hde1 :
runs completely through: NO ERRORS;
* run xfs_check again (partition is unmounted(!)):
output: sb_ifree 47759, counted 47758
* mount partition, try to access bad directory:

Quoted 1 line

immediate system hang

* reboot under earlier kernel (2.6.14.4 - vanilla)
* go to same directory, ls:

Quoted 1 line

NO HANG;

* unmount and respect with xfs_{check,repair}: expect to see no errors

Quoted 1 line

WRONG (sorta): both believe there is a log again:

* xfs_logprint:
------ (slightly different output)
Bad log record header
xfs_logprint:
data device: 0x2101
log device: 0x2101 daddr: 156288352 length: 152624

Header 0x84 wanted 0xfeedbabe
**********************************************************************
* ERROR: header cycle=132 block=52801 *
**********************************************************************
---------
* This time, I run xfs_repair with -L; remembering "no errors, and not
wanting to wait through another full "xfs_fsck", I abort after it starts (and after -L has removed problem log).
* I run xfs_check:

Quoted 1 line

no output (no errors found).

* To try to avoid problem, I copy the troublesome directory to another
directory name and delete the old directory. * run xfs_check:

Quoted 1 line

no output (no error indicated)

=> Problem seems to have been dealt with; this report is meant as a datapoint in case other problems come in...

-linda

_
_