Date: Wed, 19 Aug 1998 14:14:27 -0400 (EDT) From: "John W. DeBoskey" <jwd@unx.sas.com> To: freebsd-current@FreeBSD.ORG Subject: 13 months of user time? Message-ID: <199808191814.AA28841@mozart>
next in thread | raw e-mail | index | archive | help
Hi folks,
I found this on a 3 month old snap, and have now replicated it
on 3.0-980818-SNAP.
for i in /bin/*; do
time rsh nodename uptime
done
Yields the following output:
.....
2:00PM up 1 day, 1:39, 1 user, load averages: 0.00, 0.00, 0.00
0.19s real 0.00s user 0.00s system
2:00PM up 1 day, 1:39, 1 user, load averages: 0.00, 0.00, 0.00
0.20s real 33554431.00s user 0.00s system
2:00PM up 1 day, 1:39, 1 user, load averages: 0.00, 0.00, 0.00
0.20s real 0.02s user 0.00s system
2:00PM up 1 day, 1:39, 1 user, load averages: 0.00, 0.00, 0.00
0.19s real 0.00s user 0.01s system
.....
The values are very consistent until that 13 month value of
33554431.00s user time sneaks in. fyi: The value 33554431
repeats in all occurances of this bug...
It seems to be coming from /usr/src/usr.bin/time/time.c:
while (wait3(&status, 0, &ru) != pid); /* XXX use waitpid */
It would appear that there is a critter in our resource accounting?
Has anyone else seen this problem?
Comments? Critiques? Stupid user that can't count? :-)
thanks,
John
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199808191814.AA28841>