hallo everybody

i have a problem with my server powering down unexpectedly.

i want to check the syslog but as you know, it is lost in every restart so this is a problem. i want to find out how i can append syslog results to a file (and in a way that when the server reboots, the syslog wont be erased by the creation of a new syslog).

i am sure i have seen this mentioned somewhere in the forum but the search did not help.

can you please help?

 

PS: btw, apparently my server was constructing the parity disk from some HDDs to which i had loaded my data and the ball next to the parity drive was orange. I dont know if the parity sync was finished. but when i rebooted, the ball next to the parity drive is now green (i guess it must have finished), yet it is performing a parity-check. why would that be?

Look here: http://lime-technology.com/forum/index.php?topic=5623.msg52444#msg52444

aha

so i...

 

1.put these 4 lines to my go script

mv -f /boot/syslog.current /boot/syslog.last 2>/dev/null

cat /var/log/syslog > /boot/syslog.current

echo "*.debug  /boot/syslog.current" >> /etc/syslog.conf

/etc/rc.d/rc.syslog restart

2.reboot

3. erase the lines since the syslog has been permanently moved?

Put them in the "go" script, but you do not need to reboot if you also invoke then on the command line.

 

If you remove them from the go script they will NOT be in effect the next time you reboot.  The syslog is not permanently moved.

 

On the other hand, since flash drives have a limited number of "write" cycles before they wear out, do NOT leave the lines in longer than you need to find the errors causing the reboots, otherwise you will eventually find the flash drive unusable.

 

An alternative was posted here by WeeboTech : http://lime-technology.com/forum/index.php?topic=5623.msg58761#msg58761

 

Joe L.

thanx starcat and Joe

i think i will go with the first solution since it seems easier for me

lwt's hope i find the root of the problem..