falc410

[Log in to get rid of this advertisement]

I have a weird problem with two RedHat 5 machines.

When I type date I get the correct time returned, but when I check the syslog all the times are one hour off.
I have searched the forums but I didn't really find a solution. When I create a new file, it also has the correct time, so it really only appears wrong in all the log files under /var/log which is very very annoying!

Now I have a ntp server configured and I think the hardware clock is fine, maybe it is a timezone problem, but I'm not sure.

I'm located in Germany, so that should be GMT+1.

The only thing I found which has anything to do with the timezone is /etc/sysconfing/clock and it looks like this:

ZONE="UTC"
UTC=true
ARC=false

I don't know if these are the correct values or not. Any help is appreciated.

pinniped

Are you confusing the system timestamps (UTC) with your local time? UNIX has historically used UTC for file timestamps (and almost everything else). Individual users can set their time zone on a per-process basis, so you can run multiple instances of 'pingus' each with a different time zone. Using local times to mark files etc just doesn't make any sense on any *NIX system. Even before UNIX, many computer applications (bank transactions, airline tickets, etc) involved multiple users from all over the planet performing operations on the computer. If someone updated data, how do you know when the last change was made? Simple - use a single time (UTC) for all transactions. If a user wanted to see results displayed in their own time zone, all they had to do was set the time zone appropriately and any timezone-aware program would adjust what the user sees.

falc410

Well I thank you for the explanation but I did only understand half of it.

I will try to explain my problem once more. Right now it is 2 pm in Germany. I log in to the server and run the following commands:

[root@nagios2 nagios]# date
Fri Jan 30 14:01:06 CET 2009
[root@nagios2 nagios]# wall test

Broadcast message from root (pts/0) (Fri Jan 30 13:01:18 2009)

Do you see the difference?
Funny thing is, depending on the software if I check /var/log/messages the timestamp is either correct or one hour off:

Jan 30 14:01:18 nagios2 wall[1636]: wall: user root broadcasted 1 lines (6 chars)

This is the correct time, but other software is sometimes one hour off, this causes quite a few problems. Here is a copy and paste of my /var/log/messages:

Do you see what is happening? All those entries were in a timeframe of 7 minutes, but the logfiles recorded it with different hours, making the logfiles quite unreadable since you never can be sure what time it really was and you can't be sure if this happened in the right order or not.