[RESOLVED] Time problem?

Issues related to hardware problems
Post Reply
fugtruck
Posts: 39
Joined: 2006/08/18 18:24:08

[RESOLVED] Time problem?

Post by fugtruck » 2010/07/19 15:05:49

I have a CentOS 5.5 32-bit server with some problems that I think are all related to each other. First, ftp clients connecting to the server (running pure-ftpd) time out after about 10 seconds of inactivity, dispite the MaxIdleTime being set to 15 minutes. I increased this to 1500 minutes and clients now stay connected for a reasonable amount of time (I have not actually measured how long they stay connected).

Second, output from the 'top' command refreshes so fast that it is almost unreadable. Running the top command with a delay of 100 causes the output to refresh at a rate that we would normally expect to see.

So, between that and the ftp timeout issue, we are suspecting that there is some sort of time keeping problem on the server. If we run the command 'hwclock', we get the following error:

[i]
# hwclock
select() to /dev/rtc to wait for clock tick timed out

# hwclock --debug
hwclock from util-linux-2.13-pre7
Using /dev/rtc interface to clock.
Last drift adjustment done at 1275547673 seconds after 1969
Last calibration done at 1275547673 seconds after 1969
Hardware clock is on UTC time
Assuming hardware clock is kept in UTC time.
Waiting for clock tick...
select() to /dev/rtc to wait for clock tick timed out
...got clock tick
[/i]

Searching on the error, the only solution I could find was to add [i]CLOCKFLAGS="--directisa"[/i] to /etc/sysconfig/clock and rebooted. But that didn't help. The hwclock command does not always produce the error, sometimes returns a normal output, but 9 times out of 10 it gives the error.
Anyone else agree that these problems are related? Any suggestions on what to do?

As an fyi, the server is a VMWare virtual machine (hardware version 7) running on an ESXi 4 update1 server. None of the other VMs on this physical server are having problems.

fugtruck
Posts: 39
Joined: 2006/08/18 18:24:08

RESOLVED

Post by fugtruck » 2010/07/19 16:19:51

Well, I solved the problem by adding noacpi to the boot arguments. Then I found that the BIOS clock was off so I manually updated it. Not sure if the noapic and noacpi options mattered or not, but the problem is resolved.

Post Reply