[c-nsp] CEF fun in SXF

Gert Doering gert at greenie.muc.de
Tue Dec 13 13:22:54 EST 2005


Hi,

On Tue, Dec 13, 2005 at 09:13:04AM -0500, Phil Rosenthal wrote:
> On Dec 13, 2005, at 4:09 AM, Gert Doering wrote:
> >3B, non-XL, but 512 Mb normally are more than enough - after reboot  
> >and
> >BGP convergence, we have 140 Mb of free memory.  Over time, this went
> >down to about 25 Mb, which was low enough that we had scheduled a  
> >reboot
> >"next maintenance window" - but the remaining memory was fragmented
> >badly enough so that it made CEF seriously unhappy.  Or so.
> Mind giving some details on this BGP memory leak?

It's a bit unclear under which circumstances it happens.  It *seems*
to have something to do with BGP update-groups getting out of sync
(CSCsb09852), but when re-syncing things ("clear ip b * soft out"), 
the memory isn't released - which should be the case - and thus I got
a new bug number, CSCsc78308.

The effect is that if you plot "show proc mem", BGP will grow by about
1-2 Mbyte per day, and free mem will go down by the same amount...

Quite obvious :)

gert

-- 
USENET is *not* the non-clickable part of WWW!
                                                           //www.muc.de/~gert/
Gert Doering - Munich, Germany                             gert at greenie.muc.de
fax: +49-89-35655025                        gert at net.informatik.tu-muenchen.de


More information about the cisco-nsp mailing list