Re: Hanging udev process on nfs-mounted /dev

[Frank Steiner <fsteiner-mail@bio.ifi.lmu.de>]

[-- Attachment #1: Type: text/plain, Size: 3353 bytes --]

Frank Steiner wrote

> Kay Sievers wrote
> 
> 
>>All here known failure paths are now covered in the attached patch by
>>limiting the iteration count for loops over data read from disk. Let's
>>see what happens next :)
> 
> 
> Ok :-) I'm running this patch in parallel on hosts using tmpfs for /dev
> as well as "old" ones still using NFS. I will report success or failures!

Well, we are not done yet :-) Another udev hanging with 100% CPU, with
the latest deadlock patch applied. Here's the trace (I removed the "(gdb) s"
lines):

knuth /root# gdb -p 3480
GNU gdb 5.3.92
Copyright 2003 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i586-suse-linux".
Attaching to process 3480
Reading symbols from /sbin/udev...done.
Reading symbols from /lib/i686/libc.so.6...done.
Loaded symbols for /lib/i686/libc.so.6
Reading symbols from /lib/ld-linux.so.2...done.
Loaded symbols for /lib/ld-linux.so.2
0x4010db79 in chown@@GLIBC_2.1 () from /lib/i686/libc.so.6
(gdb) bt
#0  0x4010db79 in chown@@GLIBC_2.1 () from /lib/i686/libc.so.6
#1  0x0804be05 in secure_unlink (filename=0xbffff360 "/dev/sg2173337579")
     at udev-remove.c:78
#2  0x0804c018 in delete_node (dev=0xbffff690) at udev-remove.c:127
#3  0x0804c2c5 in udev_remove_device (path=0xbfffff70 "/class/scsi_generic/sg2",
     subsystem=0xbfffff3b "scsi_generic") at udev-remove.c:185
#4  0x080497f1 in main (argc=2, argv=0xbffffd60, envp=0xfffffe00) at udev.c:188
(gdb) s
Single stepping until exit from function chown@@GLIBC_2.1,
which has no line number information.
secure_unlink (filename=0xbffff360 "/dev/sg2173337579") at udev-remove.c:79
79      udev-remove.c: No such file or directory.
         in udev-remove.c
81      in udev-remove.c
log_message (level=7, format=0x805a060 "%s: chown(%s, 0, 0) failed with error '%s'")
     at udev.c:52
52      udev.c: No such file or directory.
         in udev.c
55      in udev.c
56      in udev.c
58      in udev.c
secure_unlink (filename=0xbffff360 "/dev/sg2173337579") at udev-remove.c:87
87      udev-remove.c: No such file or directory.
         in udev-remove.c
88      in udev-remove.c
90      in udev-remove.c
log_message (level=7, format=0x805a0a0 "%s: chmod(%s, 0000) failed with error '%s'")
     at udev.c:52
52      udev.c: No such file or directory.
         in udev.c
55      in udev.c
56      in udev.c
...
and so on. The device in question does not exist (at least not after I exited
tdb):
knuth /root# ls -la /dev/sg2173337579
ls: /dev/sg2173337579: No such file or directory

Just in case I also attached the .udev.tdb from this state if it helps. I'll
leave the host in this state until you say you don't need more info from it :-)

All the hosts running tmpfs have not reported any errors yet!

cu,
Frank






-- 
Dipl.-Inform. Frank Steiner   Web:  http://www.bio.ifi.lmu.de/~steiner/
Lehrstuhl f. Bioinformatik    Mail: http://www.bio.ifi.lmu.de/~steiner/m/
LMU, Amalienstr. 17           Phone: +49 89 2180-4049
80333 Muenchen, Germany       Fax:   +49 89 2180-99-4049
* Rekursion kann man erst verstehen, wenn man Rekursion verstanden hat. *

[-- Attachment #2: .udev.tdb --]
[-- Type: application/octet-stream, Size: 81920 bytes --]