slowing down every 60 seconds due to BGP scaner

Mailman List Mirror (Read Only)
1

Hi,
I found all of my backbone routers slow down every 60
seconds , it will last 5 seconds .
While this happens , the CPU may get more than 70%
utilization and BGP scaner is the process that chewing
the CPU (50%) .
Customer with video stream noticed that .
Cisco said it's normal .Any idea ?
Thanks you !

Ben

2

We've brought this concern up to Cisco before and they assured us that
everything is performing normally. You will see this when performing
router to router pings as well however, we have been told that packet
forwarding does not suffer. ICMP replies from the router (not through
the router) are given a very low CPU priority.

-Dave

tang bing wrote:

3

tang bing wrote:

Hi,
I found all of my backbone routers slow down every 60
seconds , it will last 5 seconds .
While this happens , the CPU may get more than 70%
utilization and BGP scaner is the process that chewing
the CPU (50%) .
Customer with video stream noticed that .
Cisco said it's normal .Any idea ?
Thanks you !

Ben

We see this every 67.5 seconds on average; utilization goes to about 99.9%.
Oddly, it seems to have started this summer. Were are currently running Version 12.2(2)T1.

4

"David McGaugh" <david_mcgaugh@eli.net> writes:

We've brought this concern up to Cisco before and they assured us that
everything is performing normally. You will see this when performing
router to router pings as well however, we have been told that packet
forwarding does not suffer. ICMP replies from the router (not through
the router) are given a very low CPU priority.

Seconded. Ping response times or lack of ping response from routers
signifies *nothing*. Ditto for traceroutes, &c. Ping *through* the
router, not *to* the router.

                                        ---Rob

5

"Robert E. Seastrom" wrote:

"David McGaugh" <david_mcgaugh@eli.net> writes:

> We've brought this concern up to Cisco before and they assured us that
> everything is performing normally. You will see this when performing
> router to router pings as well however, we have been told that packet
> forwarding does not suffer. ICMP replies from the router (not through
> the router) are given a very low CPU priority.

Seconded. Ping response times or lack of ping response from routers
signifies *nothing*. Ditto for traceroutes, &c. Ping *through* the
router, not *to* the router.

                                        ---Rob

We do streaming, and this causes a freeze up both in and out bound.
In bound there seem to be losses. Outbound all is buffered and nothing is
lost.

6

Marshall Eubanks <tme@21rst-century.com> writes:

"Robert E. Seastrom" wrote:

> "David McGaugh" <david_mcgaugh@eli.net> writes:
>
> > We've brought this concern up to Cisco before and they assured us that
> > everything is performing normally. You will see this when performing
> > router to router pings as well however, we have been told that packet
> > forwarding does not suffer. ICMP replies from the router (not through
> > the router) are given a very low CPU priority.
>
> Seconded. Ping response times or lack of ping response from routers
> signifies *nothing*. Ditto for traceroutes, &c. Ping *through* the
> router, not *to* the router.

We do streaming, and this causes a freeze up both in and out bound.
In bound there seem to be losses. Outbound all is buffered and nothing is
lost.

pinging the router causes a freeze-up? or the router is freezing up
on bgp updates every n seconds? platform/release ?

                                        ---rob

7

"Robert E. Seastrom" wrote:

Marshall Eubanks <tme@21rst-century.com> writes:

> "Robert E. Seastrom" wrote:
>
> > "David McGaugh" <david_mcgaugh@eli.net> writes:
> >
> > > We've brought this concern up to Cisco before and they assured us that
> > > everything is performing normally. You will see this when performing
> > > router to router pings as well however, we have been told that packet
> > > forwarding does not suffer. ICMP replies from the router (not through
> > > the router) are given a very low CPU priority.
> >
> > Seconded. Ping response times or lack of ping response from routers
> > signifies *nothing*. Ditto for traceroutes, &c. Ping *through* the
> > router, not *to* the router.
>
> We do streaming, and this causes a freeze up both in and out bound.
> In bound there seem to be losses. Outbound all is buffered and nothing is
> lost.

pinging the router causes a freeze-up? or the router is freezing up
on bgp updates every n seconds? platform/release ?

                                        ---rob

Dear Rob;

I said nothing about pings. The router is freezing (or at least slowing a lot) every
67.5 +- 3 seconds. We this for both inbound and outbound - outbound things are
buffered but inbound they seem to be lost.

A Cisco 7204 running Version 12.2(2)T1

8

Marshall Eubanks <tme@21rst-century.com> writes:

I said nothing about pings. The router is freezing (or at least
slowing a lot) every 67.5 +- 3 seconds. We this for both inbound and
outbound - outbound things are buffered but inbound they seem to be
lost.

A Cisco 7204 running Version 12.2(2)T1

What kind of NPE, how much memory, how many BGP views, what
interfaces, and is there a particular reason that you're running 12.2?

                                        ---rob

9

Cisco said this is normal? Normal default operation maybe. The default
BGP scan time is 60 seconds. This can be modified with the "bgp scan time
N" command.

  That's not a production router, is it? (T) code is test code. That's the
first code with IPv6 and MPLS Label Distribution Protocol support. Not the
type of stuff you'd want to be running on your backbone (unless you needed
some functionality that it introduced).

Sincerely,

Dennis J. Hartmann
White Pine Consulting
Global Knowledge-MPLS Course Director
http://www.globalknowledge.com/training/course.asp?PageID=9&courseid=1571
dennisjhartmann@hotmail.com
AOL IM: dennisjhartmann

10

Marshall Eubanks <tme@21rst-century.com> writes:

> I said nothing about pings. The router is freezing (or at least
> slowing a lot) every 67.5 +- 3 seconds. We this for both inbound and
> outbound - outbound things are buffered but inbound they seem to be
> lost.
>
> A Cisco 7204 running Version 12.2(2)T1

What kind of NPE, how much memory, how many BGP views, what
interfaces,

That + what else running on the router vesides BGP, is it a 7204VXR?
maybe just copy the output of "show hardware"

    and is there a particular reason that you're running 12.2?

                                        ---rob

  It's actually 12.2 "T" series and 12.2(2)T was the first FCS - not
exactly a good idea for a production ..
( see <http://www.cisco.com/warp/public/770/index.shtml?60,8>
for "Cisco IOS 12.2 and 12.2T Release Numbering Changes"

Do you need IPv6 with "official" support ? why the "T" series ?

If you really _must_ use 12.2T are you having problems with BGP next hop?
then try checking bug _CSCdu74704_ - supposedly fixed in IOS 12.2(2)T02