Recently I had to login to OS 11.3 via tty, but was unable to. Tty screen was flooded with syslog output. Instead of outputting info only on tty10 it was throwing it on every tty (1-6), I switched to.
How to fix this behaviour and restrict syslog output only to tty10?
did this suddenly begin happening?
was the system performing as expected for some time prior to this
discovery?
what is on tty7?
–
DenverD
CAVEAT: http://is.gd/bpoMD [posted via NNTP w/openSUSE 10.3]
When it comes to chocolate, resistance is futile.
Honestly I don’t use tty login on everyday basis, so I didn’t spot the exact moment at which it begin to happen. I was able to successfully login via tty some time ago, there was no interference from syslog with keyboard input and syslog messages were only on tty10. There is KDE on tty7.
D E wrote:
> I was able to successfully
> login via tty some time ago, there was no interference from syslog with
> keyboard input and syslog messages were only on tty10. There is KDE on
> tty7.
hmmm…something has (obviously) changed…and, without knowing what
that is i have no idea where to begin, except:
might try shutting down KDE (just log out of the session) and see if
anything changes…
and/or, shutdown KDE and then drop to runlevel 3 and check again for
change…i don’t know, but i guess there might be a clue in there
somewhere…
if 11.3 works great for a while and then stopped it could very well be
due to one of the (several) kernel changes, or any of the multitude of
package updates…now, which one went sour is really impossible to
tell from where i’m sitting…
i could only guess a best course of action might be to restore your
system for a backup made at a point in time before you “some time ago”
and test for expected service, and then slowly and methodically update
testing all along until the rogue package is found…
and, then be sure to submit a bug report on it, see:
http://en.opensuse.org/Submitting_Bug_Reports
of course, there is always the chance to google and see if anyone else
has seen this symptom, and solved the problem…
–
DenverD
CAVEAT: http://is.gd/bpoMD [posted via NNTP w/openSUSE 10.3]
When it comes to chocolate, resistance is futile.
On 2010-09-29 15:06, D E wrote:
>
> Recently I had to login to OS 11.3 via tty, but was unable to. Tty
> screen was flooded with syslog output. Instead of outputting info only
> on tty10 it was throwing it on every tty (1-6), I switched to.
>
> How to fix this behaviour and restrict syslog output only to tty10?
Run, blindly, as root:
klogconsole -r 10
That should stop it. Then, investigate.
Did you change or upgrade something?
Are you using vmware, perchance?
–
Cheers / Saludos,
Carlos E. R.
(from 11.2 x86_64 “Emerald” at Telcontar)
Thank you! This helped at least until reboot.
Related installs are updates from update repo, KDE 4.5.1 from KDE:Release:4.5 repo,
rsyslog 5.4.0-2.7.
No vmware.
On 2010-09-30 15:36, D E wrote:
> Thank you! This helped at least until reboot.
> Related installs are updates from update repo, KDE 4.5.1 from
> KDE:Release:4.5 repo,
> rsyslog 5.4.0-2.7.
> No vmware.
Search the Bugzilla database to see if someone reported something.
Check the date of the file “/etc/rsyslog.conf”, and see if it contains something like this:
kern.warning;.err;authpriv.none /dev/tty10;RSYSLOG_TraditionalFileFormat
kern.warning;.err;authpriv.none |/dev/xconsole;RSYSLOG_TraditionalFileFormat
*.emerg *
–
Cheers / Saludos,
Carlos E. R.
(from 11.2 x86_64 “Emerald” at Telcontar)
Already checked bugzilla. Nothing similar to this issue.
I’d not edited rsyslog.conf, so it is default one. Full rsyslog.conf available here
The related part is somewhat different than that in your snippet:
###
# print most important on tty10 and on the xconsole pipe
#
if ( \
/* kernel up to warning except of firewall */ \
($syslogfacility-text == 'kern') and \
($syslogseverity <= 4 /* warning */ ) and not \
($msg contains 'IN=' and $msg contains 'OUT=') \
) or ( \
/* up to errors except of facility authpriv */ \
($syslogseverity <= 3 /* errors */ ) and not \
($syslogfacility-text == 'authpriv') \
) \
then /dev/tty10
& |/dev/xconsole
# Emergency messages to everyone logged on (wall)
*.emerg *
# enable this, if you want that root is informed
# immediately, e.g. of logins
#*.alert root
#
# firewall messages into separate file and stop their further processing
#
if ($syslogfacility-text == 'kern') and \
($msg contains 'IN=' and $msg contains 'OUT=') \
then -/var/log/firewall
& ~
I suspect, that the line
*.emerg *
causing all the mess
On 2010-10-01 12:36, D E wrote:
>
> Carlos E. R.;2231034 Wrote:
> Already checked bugzilla. Nothing similar to this issue.
I wrote a bugzilla similar to this issue, time ago - that’s it why I knew the “hack”. The culprit
was vmware server 1, found no solution - except the hack I told you.
> I’d not edited rsyslog.conf, so it is default one. Full rsyslog.conf
> available ‘here’ (http://dpaste.org/hJgj/)
>
> The related part is somewhat different than that in your snippet:
Ah, yes, that’s 11.3. I checked on my 11.3 test partition and it has the same config - and no problems.
> I suspect, that the line
>
> Code:
> --------------------
> *.emerg *
> --------------------
>
> causing all the mess
The line is correct. It would display on all terminals (included those opened in X) only emergency
messages.
Something is messing with this, I have no idea what. You have the hack I told you, at least.
You could start in runlevel 3 first, and check. Then switch to runlevel 5, and check again. Then log
into X, without starting any app, and check. Finally start apps one by one, checking if the messages
appear on each app.
Do you have any proprietary app, something not coming from the suse repos?
–
Cheers / Saludos,
Carlos E. R.
(from 11.2 x86_64 “Emerald” at Telcontar)
After some research I’ve found that it happens when computer had been waked up from hibernation. With normal boot\power up all syslog output is working as expected.
Can it be some hibernation\resume script fault?
On 2010-10-14 12:06, D E wrote:
>
> After some research I’ve found that it happens when computer had been
> waked up from hibernation. With normal boot\power up all syslog output
> is working as expected.
>
> Can it be some hibernation\resume script fault?
I hibernate daily and have no problems there.
Probably some program is not restarting properly in your system.
–
Cheers / Saludos,
Carlos E. R.
(from 11.2 x86_64 “Emerald” at Telcontar)
On 2010-10-14 12:06, D E wrote:
> Can it be some hibernation\resume script fault?
Local scripts are in /etc/pm/sleep.d/
System scripts are in /usr/lib/pm-utils
–
Cheers / Saludos,
Carlos E. R.
(from 11.2 x86_64 “Emerald” at Telcontar)