[OpenIndiana-discuss] /hipster: rsyslog question

[Predrag Zečević - Unix Systems Administrator]

Hi all,

I have enabled rsyslog service (instead of default system-log):

$ svcs system-log
STATE          STIME    FMRI
disabled        8:48:36 svc:/system/system-log:default
online          8:52:13 svc:/system/system-log:rsyslog

and I am faced with strange error during reboot! From 
/var/svc/log/system-system-log:rsyslog.log:

[ Jul 14 08:48:37 Executing start method ("/lib/svc/method/rsyslog"). ]
savecore: open("/dev/dump"): No such file or directory
[ Jul 14 08:48:42 Method "start" exited with status 0. ]
[ Jul 14 08:48:43 Stopping because process dumped core. ]
[ Jul 14 08:48:43 Executing stop method (:kill). ]
[ Jul 14 08:48:43 Executing start method ("/lib/svc/method/rsyslog"). ]
savecore: open("/dev/dump"): No such file or directory
[ Jul 14 08:48:43 Method "start" exited with status 0. ]
[ Jul 14 08:48:43 Stopping because process dumped core. ]
[ Jul 14 08:48:43 Executing stop method (:kill). ]
[ Jul 14 08:48:43 Executing start method ("/lib/svc/method/rsyslog"). ]
savecore: open("/dev/dump"): No such file or directory
[ Jul 14 08:48:43 Method "start" exited with status 0. ]
[ Jul 14 08:48:43 Stopping because process dumped core. ]
[ Jul 14 08:48:43 Executing stop method (:kill). ]
[ Jul 14 08:48:43 Executing start method ("/lib/svc/method/rsyslog"). ]
savecore: open("/dev/dump"): No such file or directory
[ Jul 14 08:48:43 Method "start" exited with status 0. ]
[ Jul 14 08:48:43 Stopping because process dumped core. ]
[ Jul 14 08:48:43 Executing stop method (:kill). ]
[ Jul 14 08:48:43 Restarting too quickly, changing state to maintenance. ]

So, it went to maintenance, because /dev/dump link wasn't created BEFORE 
rsyslog is started (it is managed by svc:/system/dumpadm:default 
service, right?). From /var/svc/log/system-dumpadm:default.log we see:

[ Jul 14 08:48:39 Enabled. ]
[ Jul 14 08:49:41 Executing start method ("/lib/svc/method/svc-dumpadm"). ]
[ Jul 14 08:49:42 Method "start" exited with status 0. ]
savecore: bad magic number 6db6db6d


That dumpadm was started AFTER 1st attempt of rsyslog... Also, sym-link 
confirmes that:

$ ls -alL /dev/dump
crw-rw---- 1 root sys 281, 0 Jul 14 08:52 /dev/dump
$ ls -al /dev/dump
lrwxrwxrwx 1 root root 29 Nov 21  2013 /dev/dump -> 
../devices/pseudo/dump at 0:dump
$ ls -al /devices/pseudo/dump at 0:dump
crw-rw---- 1 root sys 281, 0 Jul 14 08:52 /devices/pseudo/dump at 0:dump

So, how I can make either that dumpadm runs BEFORE rsyslog or rsyslog 
AFTER dumpadm?

Any ideas?

With best regards.

-- 
Predrag Zečević
Technical Support Analyst
2e Systems GmbH

Telephone: +49 6196 9505 815, Facsimile: +49 6196 9505 894
Mobile:    +49  174 3109 288,     Skype: predrag.zecevic
E-mail:    predrag.zecevic at 2e-systems.com

Headquarter:          2e Systems GmbH, Königsteiner Str. 87,
                       65812 Bad Soden am Taunus, Germany
Company registration: Amtsgericht Königstein (Germany), HRB 7303
Managing director:    Phil Douglas

http://www.2e-systems.com/ - Making your business fly!