Hey all. I am having a problem with my clients logging out at random
intervals. I will be using a client, everything will be going well, then the
client will log itself out to LDM. When I log back in, I cannot execute any
localapps programs, but everything on the server will run fine. I can't even
enter a client terminal. If I hit ctrl+alt+F2, it will show the following,
and just sit there forever:

tty=/dev/tty2


The following is everything I was able to retrieve from the client and
server the last time this happened. It appears that nbd stops running.

*DMESG results (on server):*

[415148.424879] frontend[32225]: segfault at 46 ip 00007f3dba9b5a6b sp
00007fff63f85530 error 4 in libperl.so.5.10.1[7f3dba977000+162000]
[415150.408096] adduser[32248]: segfault at 46 ip 00007f299a844a6b sp
00007fff5ba73800 error 4 in libperl.so.5.10.1[7f299a806000+162000]
[415150.410459] adduser[32249]: segfault at 46 ip 00007fa7f45aba6b sp
00007fff89128b90 error 4 in libperl.so.5.10.1[7fa7f456d000+162000]
[415156.079418] type=1505 audit(1282047964.687:24):
operation="profile_replace" pid=32282 name="/usr/lib/libvirt/virt-aa-helper"
[415157.109060] type=1505 audit(1282047965.714:25):
operation="profile_load" pid=32338
name="libvirt-1f09b150-637c-917e-6a16-edd7b2091a87"
[415157.126844] device vnet1 entered promiscuous mode
[415157.128030] kvmbr0: port 3(vnet1) entering learning state
[415157.558814] lo: Disabled Privacy Extensions
[415166.120010] kvmbr0: port 3(vnet1) entering forwarding state
[415167.800016] vnet1: no IPv6 routers present
[427412.623161] qemu-system-x86[32345]: segfault at 10000675702 ip
000000004a22d6a3 sp 00007ffff5a7c1d0 error 4
[427412.774341] kvmbr0: port 3(vnet1) entering disabled state
[427412.837468] type=1505 audit(1282060221.445:26):
operation="profile_remove" pid=6183
name="libvirt-1f09b150-637c-917e-6a16-edd7b2091a87" namespace="root"
[427412.853399] device vnet1 left promiscuous mode
[427412.853403] kvmbr0: port 3(vnet1) entering disabled state

*dmesg results (on client)*

[7954.585906]  nbd9: unknown partition table
[7954.592946] nbd9: NBD_DISCONNECT
[7954.593618] nbd9: Receive control failed (result -32)
[7954.593736] nbd9: queue cleared


*/var/log/syslog (server)*

Aug 17 12:14:21 dlws ldminfod[6473]: connect from 192.168.1.156
(192.168.1.156)
Aug 17 12:14:23 dlws nbdrootd[6476]: connect from 192.168.1.156
(192.168.1.156)
Aug 17 12:14:23 dlws nbd_server[6477]: connect from 192.168.1.156, assigned
file is /opt/ltsp/images/i386.img
Aug 17 12:14:23 dlws nbd_server[6477]: Size of exported file/device is
1060065280
Aug 17 12:14:23 dlws nbd_server[6477]: Disconnect request received.
Aug 17 12:14:24 dlws bonobo-activation-server (jsass-6479): could not
associate with desktop session: Failed to connect to socket
/tmp/dbus-zB85vc9Wjy: Connection refused


*/var/log/syslog (on client)*

Aug 17 12:15:24 kernel: [7954.585906]  nbd9: unknown partition table
Aug 17 12:15:24 kernel: [7954.592946] nbd9: NBD_DISCONNECT
Aug 17 12:15:24 kernel: [7954.593618] nbd9: Receive control failed (result
-32)
Aug 17 12:15:24 kernel: [7954.593736] nbd9: queue cleared


*/var/log/daemon (server)*

Aug 17 12:14:21 dlws ldminfod[6473]: connect from 192.168.1.156
(192.168.1.156)
Aug 17 12:14:23 dlws nbdrootd[6476]: connect from 192.168.1.156
(192.168.1.156)
Aug 17 12:14:23 dlws nbd_server[6477]: connect from 192.168.1.156, assigned
file is /opt/ltsp/images/i386.img
Aug 17 12:14:23 dlws nbd_server[6477]: Size of exported file/device is
1060065280
Aug 17 12:14:23 dlws nbd_server[6477]: Disconnect request received.


*/var/log/auth.log (server)*

Aug 17 12:14:19 dlws sshd[3901]: Received disconnect from 192.168.1.156: 2:
Packet corrupt
Aug 17 12:14:19 dlws sshd[3831]: pam_unix(sshd:session): session closed for
user jsass
Aug 17 12:14:19 dlws gnome-keyring-daemon[3973]: dbus failure unregistering
from session: Connection is closed
Aug 17 12:14:19 dlws gnome-keyring-daemon[3973]: dbus failure unregistering
from session: Connection is closed
Aug 17 12:14:19 dlws polkitd(authority=local): Unregistered Authentication
Agent for session /org/freedesktop/ConsoleKit/Session28 (system bus name
:1.374, object path /org/gnome/PolicyKit1/AuthenticationAgent, locale
en_US.utf8) (disconnected from bus)


*/var/log/kern.log (client)*

Aug 17 12:15:24 ltsp156 kernel: [7954.585906]  nbd9: unknown partition table
Aug 17 12:15:24 ltsp156 kernel: [7954.592946] nbd9: NBD_DISCONNECT
Aug 17 12:15:24 ltsp156 kernel: [7954.593618] nbd9: Receive control failed
(result -32)
Aug 17 12:15:24 ltsp156 kernel: [7954.593736] nbd9: queue cleared


*"df -h" (client)*

Filesystem        Size    Used    Avail    Use%    Mounted on
aufs            497M    804K    496M    1%    /
none            492M    232K    492M    1%    /dev
/dev/nbd0           1011M   1011M       0    100%    /rofs
tmpfs            497M    804K    496M    1%    /cow
none            497M       0    497M    0%    /dev/shm
none            497M     60K    497M    1%    /var/run
none            497M       0    497M    0%    /var/lock
none            497M       0    497M    0%    /lib/init/rw
none            497M    4.0K    497M    1%    /tmp


*"df -i" (client)*

Filesystem        Inodes    IUsed     IFree    IUse%    Mounted on
aufs            127112       83    127029    1%    /
none            125930      681    127029    1%    /dev
/dev/nbd0             44306  44306         0    100%    /rofs
tmpfs            127112       83    127029    1%    /cow
none            127112        1    127111    1%    /dev/shm
none            127112       30    127082    1%    /var/run
none            127112        1    127111    1%    /var/lock
none            127112        1    127111    1%    /lib/init/rw
none            127112        7    127105    1%    /tmp



Stranger still, this will only happen to one client at a time. One client
will be having problems, while another one in another room will be operating
properly. I have experimented with a lot with various clients, so the client
hardware does not appear to be an issue. Rebooting the client sometimes
fixes the issue, sometimes it does not. Rebooting the server appears to
"fix" the problem for about a week before it gets bad again. There are no
filled drives or caches on the server, and the RAM on the client and server
appear to have a large amount unused or free. I am not sure how to continue
the diagnosis process. Any suggestions would be appreciated.

Joel
------------------------------------------------------------------------------
This SF.net Dev2Dev email is sponsored by:

Show off your parallel programming skills.
Enter the Intel(R) Threading Challenge 2010.
http://p.sf.net/sfu/intel-thread-sfd
_____________________________________________________________________
Ltsp-discuss mailing list.   To un-subscribe, or change prefs, goto:
      https://lists.sourceforge.net/lists/listinfo/ltsp-discuss
For additional LTSP help,   try #ltsp channel on irc.freenode.net

Reply via email to