** Reply to note from Lowell Gilbert <[EMAIL PROTECTED]> 29 Jul 2003 16:16:01 -0400

> > I understand, but upgrading this machine has a cost and I must be 
> > absolutely sure this will not introduce new problems as previous 
> > upgrades did. 
> I understand, but my time has a cost as well.   

No doubt about that. As I said in my first message, I *AM* willing to take an upgrade, 
I'd only like to have a bit more
insight. In case this depends on other things, maybe it's a well-known problem, an 
error of mine, a
misconfiguration, maybe about mbufs as you suggested, I don't think the upgrade would 
matter and/or help.

> > >I don't have enough time to go into this in depth, but see if the 
> > >problem is affected by increasing the number of mbufs. 
> >  
> > Nice suggestion, how do I do that? 
> Nice sarcasm; the obvious response is that you could read the manual. 

I had read the handbook, but found this not so clear.
I also tried "man 9 mbuf", as suggested in netstat's man page, but this doesn't seem 
to be available on my system

I was also a bit scared, since it is said to possibly lead to boot time crashes and I 
don't easily have physical access
to the machine (which would then keep crashing and crashing, since the problem would 
be in the startup configuration

I did instead produce another crash and measured mbufs usage: several seconds before 
the crash (I don't think it was
more than a minute and a half) "netstat -m" gave:

207/496/4096 mbufs in use (current/peak/max):
        203 mbufs allocated to data
        2 mbufs allocated to packet headers
        2 mbufs allocated to fragment reassembly queue headers
198/248/1024 mbuf clusters in use (current/peak/max)
620 Kbytes allocated to network (20% of mb_map in use)
0 requests for memory denied
0 requests for memory delayed
0 calls to protocol drain routines

So I would guess I'm not short on these. Does this answer your suggestion or should I 
try anyway?
If so, I'll try and do this next time I happen to be physically there (and possibly 
upgrade too).

 bye & Thanks

[EMAIL PROTECTED] mailing list
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Reply via email to