We've been getting a lot of these messages lately on one of
our older Alphas. Eventually, the system freeezes and we
can't log on, or do much of anything except reboot if we
do manage a logon.
We tried upping the number of maxusers from 64
to 128, and modifying /sys/conf/param.c as listed below,
but these haven't really helped. The SAs who originally did
the configuration are long gone. Any suggestions? Sorry
if this has been covered before, we're new to this list.
/var/adm/messages excerpt:
May 29 09:47:06 be1 vmunix: 3>vnode table is full
May 29 09:47:06 be1 vmunix: vnode table is full
May 29 09:47:06 be1 last message repeated 861 times
May 29 09:47:06 be1 vmunix: 3>vnode table is full
May 29 09:47:06 be1 vmunix: vnode table is full
May 29 09:47:06 be1 last message repeated 655 times
May 29 09:47:06 be1 vmunix: 3>vnode table is full
May 29 09:47:06 be1 vmunix: vnode table is full
May 29 09:47:09 be1 last message repeated 885 times
May 29 09:50:12 be1 vmunix: Alpha boot: available memory from 0x92e000 to 0x1000
0000
May 29 09:50:12 be1 vmunix: DEC OSF/1 V2.0 (Rev. 240); Wed May 20 15:37:26 GMT-0
500 1998
May 29 09:50:12 be1 vmunix: physical memory = 254.00 megabytes.
May 29 09:50:12 be1 vmunix: available memory = 234.90 megabytes.
May 29 09:50:12 be1 vmunix: using 975 buffers containing 7.61 megabytes of memor
y
One of the SAs changes:
root1[be1]/sys/conf> diff param.c param.c.052098
211c211
< #define NVNODE (NPROC + (2*MAXUSERS) + 256)
---
> #define NVNODE (NPROC + (2*MAXUSERS) + 128)
--
*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=**
Michael Watson, EMT-P N6WAV Virginia Beach, Virginia
N50347 C150H _at_ KCPK mjwatson_at_livenet.net
*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=**
Received on Fri May 29 1998 - 16:29:02 NZST