[c-nsp] 7500 having problems after upgrade from 12.0.19S2 to12.0.31S1

[Corneliu Tanasa]

Rodney, I think that it depends on the feature set you're running.  At least
for ISV, I'm running 2 full feeds, plus one half feed on a RSP4 and it works
very well for months with 12.3.x (currently 12.3.16)
                Head    Total(b)     Used(b)     Free(b)   Lowest(b)
Largest(b)
Processor   4349E4E0   213261088   180534804    32726284     6923904
6028088
     Fast   4347E4E0      131072      102872       28200       28200
28156

Current uptime: uptime is 7 weeks, 4 days, 16 hours, 52 minutes

Corneliu

-----Original Message-----
From: cisco-nsp-bounces at puck.nether.net
[mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Rodney Dunn
Sent: Tuesday, November 22, 2005 3:48 PM
To: Robert E.Seastrom
Cc: CLAEREBOUDT Elke; cisco-nsp at puck.nether.net; Jon Lewis
Subject: Re: [c-nsp] 7500 having problems after upgrade from 12.0.19S2
to12.0.31S1

On Tue, Nov 22, 2005 at 08:14:42AM -0500, Robert E.Seastrom wrote:
> Jon Lewis <jlewis at lewis.org> writes:
> 
> > On Tue, 22 Nov 2005, CLAEREBOUDT Elke wrote:
> >
> >> We've upgraded a 7500 from 12.0.19S2 to 12.0.31S1 , RSP has 256K of
> >> memory. After 10 days router started to misbehave and we had to isolate
> >> it from the network. Cisco case opened gave us the answer, you need
512M
> >> on a rsp when it has more then 100K routes (has 170K). Cpu is very
> >> unstable and points to snmp engine.
> >
> > What was your free mem under 12.0.19S2?  Assuming it was considerably 
> > more, I'd say there's bugs/memory leaks in the new IOS.  I've got
several 
> > RSP4's with full BGP routes and 40-50mb free, all running 12.2(18)S.
> 
> ST merge was after 12.0(21)S.  The new BGP data structure is said to
> be more computationally efficient at the expense of memory efficiency.

Absolutely correct. However, there is a memory gain too in that
for a full reconvergence it takes much much less transient memory
to reconverge.

Not pointed at Robert but a general comment:

We can't code new features without making the image larger.
We do (even though some may not think so) try very very hard to
be extremely memory conscious when we code new features. The
mistake we've made is having parts where they memory can't be
upgraded and we are trying to not make that mistake again.

256M is simply not enough anymore for dual feeds and recent code.

Rodney

> 
>                                         ---Rob
> 
> 
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
_______________________________________________
cisco-nsp mailing list  cisco-nsp at puck.nether.net
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/