sudden errors in /var/log/messages

sathyguy · 01-17-2008, 02:27 AM

Friends,

OS: RHEL AS 3.0

Yesterday evening, i entered into the GUI using "startx" to take oracle backup in the tapedrive, i saw the exclamatory on the tapedrive icon. then i checked the /var/log/messages

Quote:

Jan 17 10:04:28 proddb login(pam_unix)[17903]: session opened for user root by LOGIN(uid=0)
Jan 17 10:04:28 proddb -- root[17903]: ROOT LOGIN ON tty1
Jan 17 10:04:39 proddb modprobe: modprobe: Can't locate module char-major-10-134
Jan 17 10:04:43 proddb kernel: mtrr: type mismatch for f6000000,800000 old: uncachable new: write-combining
Jan 17 10:04:43 proddb kernel: mtrr: type mismatch for f6000000,800000 old: uncachable new: write-combining
Jan 17 10:04:47 proddb gconfd (root-5787): starting (version 2.2.1), pid 5787 user 'root'
Jan 17 10:04:47 proddb gconfd (root-5787): Resolved address "xml:readonly:/etc/gconf/gconf.xml.mandatory" to a read-only config source at position 0
Jan 17 10:04:47 proddb gconfd (root-5787): Resolved address "xml:readwrite:/root/.gconf" to a writable config source at position 1
Jan 17 10:04:47 proddb gconfd (root-5787): Resolved address "xml:readonly:/etc/gconf/gconf.xml.defaults" to a read-only config source at position 2
Jan 17 10:04:48 proddb xinetd[5797]: warning: can't get client address: Transport endpoint is not connected
Jan 17 10:05:52 proddb modprobe: modprobe: Can't locate module sound-slot-0
Jan 17 10:05:52 proddb modprobe: modprobe: Can't locate module sound-service-0-3
Jan 17 10:05:54 proddb gconfd (root-5787): Received signal 1, shutting down cleanly
Jan 17 10:05:54 proddb gconfd (root-5787): Exiting
Jan 17 10:06:01 proddb login(pam_unix)[17903]: session closed for user root

what im suppose to do?
can i restart the server?

thanks

unSpawn · 01-17-2008, 03:38 AM

Quote:

Originally Posted by sathyguy

modprobe: modprobe: Can't locate module char-major-10-134

That's APM. If not used it's not fatal.

Quote:

Originally Posted by sathyguy

modprobe: Can't locate module sound

If not used it's not fatal.

The rest are also messages of the informational level (level 6, KERN_INFO). The module messages you can avoid by putting in "alias $modulename off" in /etc/modprobe.conf. There's no messages related to your tapedrive there. The only time you need to reboot the server is when things can't be fixed in other ways, like loading a newly installed kernel, recovering from a kernel crash or some hardware device changes.

sathyguy · 01-17-2008, 04:08 AM

Thanks for your info....still some doubts

what is APM?

"That's APM. If not used it's not fatal."

so, no need to worry about these errors....

? if yes, then i will be happy

then how can i know why the tape drive is not working.
I am able to login to tape drive in the graphical mode.
but, when i right click on the drive to select the folder for backup nothing is coming only the properties is coming....
this is happening only from yesterday onwards.
to solve this tapedrive issue can i restart the server or without restarting is it possible to find out the cause?

unSpawn · 01-17-2008, 04:48 AM

Quote:

Originally Posted by sathyguy

what is APM?

The TLA stands for "Advanced Power Management". APM is the precursor of ACPI. Now you've got 70+ posts here. You *must* have come across the LQ search function. Please make use of that and your local documents.

Quote:

Originally Posted by sathyguy

this is happening only from yesterday onwards.

Review your system logs from the last time you where able to backup succesfully.

Quote:

Originally Posted by sathyguy

is it possible to find out the cause?

Sure. Posting the brand and type of the tapedrive could help. Is the tape at /dev/st0? What does 'mt -f /dev/st0 status' return? Are all necessary modules loaded?

sathyguy · 01-17-2008, 05:09 AM

Sorry for not using search function......

The brand is "HP Storage Works dat 40"

here is the output of mt....

Quote:

[oracle@proddb oracle]$ mt -f /dev/st0 status
/dev/st0: Permission denied
[oracle@proddb oracle]$ su
Password:
[root@proddb oracle]# mt -f /dev/st0 status
SCSI 2 tape drive:
File number=-1, block number=-1, partition=0.
Tape block size 0 bytes. Density code 0x0 (default).
Soft error count since last status=0
General status bits on (50000):
DR_OPEN IM_REP_EN

Thanks

unSpawn · 01-17-2008, 06:04 AM

Quote:

Originally Posted by sathyguy

DR_OPEN

That means "door open". If it isn't could try another tape.

sathyguy · 01-19-2008, 02:07 AM

hi,

i restarted the server now the tapedrive is working.
but when i work in "Command interface" no problem i didnt see any messages in my /var/log/messages.
when i enter into gui im getting the following messages in "/var/log/messages"

Quote:

Jan 19 08:22:06 proddb -- root[2591]: ROOT LOGIN ON tty1
Jan 19 08:22:09 proddb modprobe: modprobe: Can't locate module char-major-10-134
Jan 19 08:22:13 proddb kernel: mtrr: type mismatch for
f6000000,800000 old: uncachable new: write-combining
Jan 19 08:22:13 proddb kernel: mtrr: type mismatch for
f6000000,800000 old: uncachable new: write-combining
Jan 19 08:22:16 proddb gconfd (root-15878): starting (version 2.2.1), pid 15878 user 'root'
Jan 19 08:22:16 proddb gconfd (root-15878):
Resolved address "xml:readonly:/etc/gconf/gconf.xml.mandatory" to a read-only config source at position 0
Jan 19 08:22:16 proddb gconfd (root-15878):
Resolved address "xml:readwrite:/root/.gconf" to a writable config source at position 1
Jan 19 08:22:16 proddb gconfd (root-15878):
Resolved address "xml:readonly:/etc/gconf/gconf.xml.defaults" to a read-only config source at position 2
Jan 19 08:22:17 proddb xinetd[15888]: warning: can't get client address:
Transport endpoint is not connected
Jan 19 08:23:35 proddb modprobe: modprobe: Can't locate module sound-slot-0
Jan 19 08:23:35 proddb modprobe: modprobe: Can't locate module sound-service-0-3
Jan 19 08:23:37 proddb gconfd (root-15878): Received signal 1, shutting down cleanly
Jan 19 08:23:37 proddb gconfd (root-15878): Exiting
[root@proddb oracle]#

what i should do for the following message?

Quote:

Jan 19 08:22:09 proddb modprobe: modprobe: Can't locate module char-major-10-134

can i.....edit the "grub.conf" and add apm=off
or
add "alias char-major-10-134 off" in modules.conf
by adding the above will do any harm?

like the above can i edit the modules.conf for sound card for the below messages

Quote:

Jan 19 08:23:35 proddb modprobe: modprobe: Can't locate module sound-slot-0
Jan 19 08:23:35 proddb modprobe: modprobe: Can't locate module sound-service-0-3

but what about the below messages? what should i do?
i didnt find a file "/etc/modprobe.conf"
instead of that there is a file "/etc/modules.conf"
i added "alias $modulename off" (without quotes) in modules.conf
still the error appears.

Quote:

Jan 19 08:22:13 proddb kernel: mtrr: type mismatch for
f6000000,800000 old: uncachable new: write-combining
Jan 19 08:22:13 proddb kernel: mtrr: type mismatch for
f6000000,800000 old: uncachable new: write-combining
Jan 19 08:22:16 proddb gconfd (root-15878): starting (version 2.2.1), pid 15878 user 'root'
Jan 19 08:22:16 proddb gconfd (root-15878):
Resolved address "xml:readonly:/etc/gconf/gconf.xml.mandatory" to a read-only config source at position 0
Jan 19 08:22:16 proddb gconfd (root-15878):
Resolved address "xml:readwrite:/root/.gconf" to a writable config source at position 1
Jan 19 08:22:16 proddb gconfd (root-15878):
Resolved address "xml:readonly:/etc/gconf/gconf.xml.defaults" to a read-only config source at position 2
Jan 19 08:22:17 proddb xinetd[15888]: warning: can't get client address: Transport endpoint is not connected
Jan 19 08:23:37 proddb gconfd (root-15878): Received signal 1, shutting down cleanly
Jan 19 08:23:37 proddb gconfd (root-15878): Exiting

am i have to reboot the server if i edit the modules.conf or grub.conf?

thanks

unSpawn · 01-19-2008, 10:22 AM

Quote:

Originally Posted by sathyguy

i added "alias $modulename off" (without quotes) in modules.conf, still the error appears. am i have to reboot the server if i edit the modules.conf or grub.conf?

That's because the module is still loaded. Again, no need to reboot.