PDA

View Full Version : boot-error sles11 sp2: sysfs: cannot create duplicate file



vinnie65
06-Aug-2012, 15:19
After installing sles11 sp2 on an HP DL380 the following error is displayed during boot on several servers:
[ 15.130156] scsi 0:0:2:1: Direct-Access EMC SYMMETRIX 5874 PQ: 0 ANSI: 4^M
[ 15.539137] scsi 0:0:2:2: Direct-Access EMC SYMMETRIX 5874 PQ: 0 ANSI: 4^M
[ 15.901360] lpfc 0000:0b:00.1: 1:1303 Link Up Event x1 received Data: x1 x1 x20 x0 x0 x0 0^M
[ 16.361938] ------------[ cut here ]------------^M
[ 16.593961] WARNING: at /usr/src/packages/BUILD/kernel-default-3.0.13/linux-3.0/fs/sysfs/dir.c:455 sysfs_add_one+0xb7/0xe0()^M
[ 17.157151] Hardware name: ProLiant DL380 G7^M
[ 17.371612] sysfs: cannot create duplicate filename '/devices/pci0000:00/0000:00:08.0/0000:0b:00.0/host0/rport-0:0-4/target0:0:2/0:0:2:2/dh_state'^M
[ 18.032335] Modules linked in: scsi_dh lpfc scsi_transport_fc scsi_tgt hpsa cciss scsi_mod^M
[ 18.444105] Supported: Yes^M
[ 18.579900] Pid: 5, comm: kworker/u:0 Not tainted 3.0.13-0.27-default #1^M
[ 18.915992] Call Trace:^M
[ 19.038553] [<ffffffff810048b5>] dump_trace+0x75/0x300^M
[ 19.300724] [<ffffffff8143ea0f>] dump_stack+0x69/0x6f^M
....
facts:
# the server was running sles11 sp1 for over a year before, no problems
# the same boot problem on a handfull of similar servers
# but 30+ servers also installed with sp2 (and sp1 before) do not show this problem
# all our servers boot from SAN (EMC DMX/VMAX)
# the problematic sp2-servers all have at least 4 SAN-volumes, where as the working sp2-servers usually have one boot and one data volume

Has any one ever had the same or a similar problem or behaviour? and what could be the cause or the cure?

malcolmlewis
06-Aug-2012, 15:49
After installing sles11 sp2 on an HP DL380 the following error is
displayed during boot on several servers:
[ 15.130156] scsi 0:0:2:1: Direct-Access EMC SYMMETRIX
5874 PQ: 0 ANSI: 4^M
[ 15.539137] scsi 0:0:2:2: Direct-Access EMC SYMMETRIX
5874 PQ: 0 ANSI: 4^M
[ 15.901360] lpfc 0000:0b:00.1: 1:1303 Link Up Event x1 received
Data: x1 x1 x20 x0 x0 x0 0^M
[ 16.361938] ------------[ cut here ]------------^M
[ 16.593961] WARNING: at
/usr/src/packages/BUILD/kernel-default-3.0.13/linux-3.0/fs/sysfs/dir.c:455
sysfs_add_one+0xb7/0xe0()^M
[ 17.157151] Hardware name: ProLiant DL380 G7^M
[ 17.371612] sysfs: cannot create duplicate filename
'/devices/pci0000:00/0000:00:08.0/0000:0b:00.0/host0/rport-0:0-4/target0:0:2/0:0:2:2/dh_state'^M
[ 18.032335] Modules linked in: scsi_dh lpfc scsi_transport_fc
scsi_tgt hpsa cciss scsi_mod^M
[ 18.444105] Supported: Yes^M
[ 18.579900] Pid: 5, comm: kworker/u:0 Not tainted
3.0.13-0.27-default #1^M
[ 18.915992] Call Trace:^M
[ 19.038553] [<ffffffff810048b5>] dump_trace+0x75/0x300^M
[ 19.300724] [<ffffffff8143ea0f>] dump_stack+0x69/0x6f^M
....
facts:
# the server was running sles11 sp1 for over a year before, no
problems
# the same boot problem on a handfull of similar servers
# but 30+ servers also installed with sp2 (and sp1 before) do not show
this problem
# all our servers boot from SAN (EMC DMX/VMAX)
# the problematic sp2-servers all have at least 4 SAN-volumes, where as
the working sp2-servers usually have one boot and one data volume

Has any one ever had the same or a similar problem or behaviour? and
what could be the cause or the cure?



Hi
On the updated servers affected, have all had the latest updates esp the
kernel?

On a side note, when pasting console text, can you use code tags ;)

--
Cheers Malcolm °¿° (Linux Counter #276890)
SUSE Linux Enterprise Desktop 11 (x86_64) Kernel 3.0.34-0.7-default
up 8 days 11:21, 2 users, load average: 0.21, 0.42, 0.43
CPU Intel i5 CPU M520@2.40GHz | Intel Arrandale GPU

vinnie65
10-Aug-2012, 09:20
Hi Malcom,

we did some more investigation:
1) if we set the kernel option maxcpus=1, the system installs and boots correctly
2) after that, we install the latest updates, ie. uname -a shows
Linux lx00142 3.0.34-0.7-default #1 SMP Tue Jun 19 09:56:30 UTC 2012 (fbfc70c) x86_64 x86_64 x86_64 GNU/Linux
3) next we reboot the system without the maxcpu-option (=all cores/threads on)
and the system hangs again with same problem

thanks,
Vincent

vinnie65
10-Aug-2012, 09:31
oops, "uname -a" from wrong system :-), nevertheless:
lx00241:~ # uname -a
Linux lx00241 3.0.34-0.7-default #1 SMP Tue Jun 19 09:56:30 UTC 2012 (fbfc70c) x86_64 x86_64 x86_64 GNU/Linux

malcolmlewis
10-Aug-2012, 14:20
oops, "uname -a" from wrong system :-), nevertheless:
lx00241:~ # uname -a
Linux lx00241 3.0.34-0.7-default #1 SMP Tue Jun 19 09:56:30 UTC 2012
(fbfc70c) x86_64 x86_64 x86_64 GNU/Linux



Hi
Sounds like you might need to try some other grub boot options (there
are many), is there a later BIOS for the system available?

Can you try the following grub boot option;


acpi_osi=Linux


--
Cheers Malcolm °¿° (Linux Counter #276890)
SUSE Linux Enterprise Desktop 11 (x86_64) Kernel 3.0.34-0.7-default
up 12 days 9:50, 2 users, load average: 0.43, 0.54, 0.53
CPU Intel i5 CPU M520@2.40GHz | Intel Arrandale GPU