From: Michael DeSimone (michael@desimone.net)
Date: Fri Mar 12 2004 - 11:03:51 EST
Hello Managers,
I have a 420R that has suddenly began exhibiting bizarre issues with
it's date. We first noticed this in our iPlanet (4 SP11). It would be
logging along just swimmingly and then the date would jump exactly 4
hours ahead for a few entries and then come back in line. This box was
running xntpd and logged drift to /var/adm/messages - nothing this
drastic was noted there. Some developers had sworn up and down that they
were on the box and ran a `date` and saw the same time difference. They
blamed NTP so, even though no other box on our network exhibited this
issue NTP was turned off. Now this is the point at which I was pulled
into the issue. I set up a stupid little script that runs a date dumps
it to a file then sleeps for 60 seconds and repeats. I was hoping to see
somewhat normal operations. Of course I was not that lucky. Doing this
my clock will not display the *4 and *5 minutes i.e. it would not show
7:44 and 7:45 instead it would show 7:40 and 7:41 (see pasted script
output below). I ran the script on 3 other machines in the same
environment and they all report time normally, in sequence.
vmstat -i reports normal for the clock (100)
Doing a sum on date reports the same as other systems
Does anyone have any idea what could cause this? Any suggestions on what
else I could do to figure out what is broken?
Thanks - will Summarize,
Michael
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:28:15 EDT