Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 25 Jul 1998 23:33:12 -0500 (CDT)
From:      Kevin Day <toasty@home.dragondata.com>
To:        current@FreeBSD.ORG
Subject:   Last batch of -current bug reports
Message-ID:  <199807260433.XAA29040@home.dragondata.com>

next in thread | raw e-mail | index | archive | help

I assumed that these problems were well known, but in talking to a few
people, they'd never heard of any of this. I apologize if I'm getting
annoying with these bug reports that are probably already known by those who
care, but since October 15th is getting closer, I thought I'd repeat...

Random occurances of:

swap_pager: suggest more swap space: 254 MB

When there obviously is plenty of ram/swap available. I happened to be in
'top' the last time this occurred. I saw a flash of 'k_getuid: kvm_getprocs'
fly across the screen, and the above message suggesting more swap. The last
printout from top before it exited was:

last pid: 15377;  load averages:  1.63,  1.44,  1.40    23:07:02
153 processes: 1 running, 152 sleeping

Mem: 75M Active, 10M Inact, 23M Wired, 9116K Cache, 8348K Buf, 6396K Free
Swap: 164M Total, 46M Used, 118M Free, 28% Inuse


pindent complains a lot about bad addresses, and kvm_getprocs not working..
(yet still works 90% of the time)


Processes get stuck randomly in 'vmwait'. (can be killed with -9 though)

Processes can get stuck in 'nfsrcv' when an nfs server dies and comes back. (are
unkillable)

'killall -HUP mountd' causes 'panic: vfs_unlock: not locked' if the server
is busy. (this occurs in -STABLE, too)

de0 driver is still unstable under heavy (full-duplex 100mb 60-70%
utilization) traffic. (causes random page fault while in kernel mode)


This will probably be one of my last bug reports before I go back to
-STABLE. We bit the bullet and just bought some fast PII/400 servers instead
of dealing with the problems of SMP pentium/pentium pro's. (as much as I
like -current, our customers are complaining heavily about the once-a-week
crashes)

If there's any information anyone wants before we newfs our drives, please
tell me know. :)



Kevin

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?199807260433.XAA29040>