03-11-2010 08:26 AM - edited 03-04-2019 07:47 AM
We have a real doozy of a problem here. 2 T1's one side on a 2800 router, the other end on different DS3's on the same 7200 router. We have done this with other sites with no problems, but this one has us stumped. One T1 is missing the PPP keepalives, and causes the multilink to flap and not work. The T1 is testing 100% clean, both directions CSU to CSU. No errors at all, but every minute, it misses 5 keepalives and drops protocol, renegotiates, and comes back up. Right now I have it as a single PtP T1 with IP's at both ends, and I can slam it with max bytes forever, and only when the LCP drops do I lose packets, but again, no errors. When brought into the multilink though, massive errors, both on the T1 and the multilink.
Anyone give me an idea what to look for next? The only thing I can think of is a misprovisioned circuit, but there are no errors.
Here is the debug output
*Mar 11 16:04:40.440: Se0/3/0 PPP: Missed 5 keepalives, taking LCP down
*Mar 11 16:04:40.440: Se0/3/0 PPP: Sending Acct Event[Down] id[5B7F]
*Mar 11 16:04:40.440: Se0/3/0 LCP: State is Closed
*Mar 11 16:04:40.440: Se0/3/0 PPP: Phase is DOWN
*Mar 11 16:04:40.440: Se0/3/0 CDPCP: State is Closed
*Mar 11 16:04:40.440: Se0/3/0 IPCP: State is Closed
*Mar 11 16:04:40.444: Se0/3/0 PPP: Phase is ESTABLISHING, Passive Open
*Mar 11 16:04:40.444: Se0/3/0 LCP: State is Listen
*Mar 11 16:04:40.444: Se0/3/0 IPCP: Remove route to 10.100.106.17
*Mar 11 16:04:41.440: %LINEPROTO-5-UPDOWN: Line protocol on Interface Serial0/3/0, changed state to down
*Mar 11 16:04:42.444: Se0/3/0 PPP: Outbound cdp packet dropped
*Mar 11 16:04:42.456: Se0/3/0 LCP: Timeout: State Listen
*Mar 11 16:04:42.456: Se0/3/0 LCP: O CONFREQ [Listen] id 113 len 10
*Mar 11 16:04:42.456: Se0/3/0 LCP: MagicNumber 0x26ED9FC2 (0x050626ED9FC2)
*Mar 11 16:04:42.920: Se0/3/0 LCP: I CONFREQ [REQsent] id 1 len 10
*Mar 11 16:04:42.920: Se0/3/0 LCP: MagicNumber 0x36E8A91B (0x050636E8A91B)
*Mar 11 16:04:42.920: Se0/3/0 LCP: O CONFACK [REQsent] id 1 len 10
*Mar 11 16:04:42.920: Se0/3/0 LCP: MagicNumber 0x36E8A91B (0x050636E8A91B)
*Mar 11 16:04:42.920: Se0/3/0 LCP: I CONFACK [ACKsent] id 113 len 10
*Mar 11 16:04:42.920: Se0/3/0 LCP: MagicNumber 0x26ED9FC2 (0x050626ED9FC2)
*Mar 11 16:04:42.920: Se0/3/0 LCP: State is Open
*Mar 11 16:04:42.920: Se0/3/0 PPP: Phase is FORWARDING, Attempting Forward
*Mar 11 16:04:42.920: Se0/3/0 PPP: Phase is ESTABLISHING, Finish LCP
*Mar 11 16:04:42.924: Se0/3/0 PPP: Phase is UP
*Mar 11 16:04:42.924: Se0/3/0 IPCP: O CONFREQ [Closed] id 1 len 10
*Mar 11 16:04:42.924: Se0/3/0 IPCP: Address 10.100.106.18 (0x03060A646A12)
*Mar 11 16:04:42.924: Se0/3/0 CDPCP: O CONFREQ [Closed] id 1 len 4
*Mar 11 16:04:42.924: Se0/3/0 PPP: Process pending ncp packets
*Mar 11 16:04:43.372: Se0/3/0 IPCP: I CONFREQ [REQsent] id 1 len 10
*Mar 11 16:04:43.372: Se0/3/0 IPCP: Address 10.100.106.17 (0x03060A646A11)
*Mar 11 16:04:43.376: Se0/3/0 IPCP: O CONFACK [REQsent] id 1 len 10
*Mar 11 16:04:43.376: Se0/3/0 IPCP: Address 10.100.106.17 (0x03060A646A11)
*Mar 11 16:04:43.376: Se0/3/0 IPCP: I CONFACK [ACKsent] id 1 len 10
*Mar 11 16:04:43.376: Se0/3/0 IPCP: Address 10.100.106.18 (0x03060A646A12)
*Mar 11 16:04:43.376: Se0/3/0 IPCP: State is Open
*Mar 11 16:04:43.376: Se0/3/0 CDPCP: I CONFREQ [REQsent] id 1 len 4
*Mar 11 16:04:43.376: Se0/3/0 CDPCP: O CONFACK [REQsent] id 1 len 4
*Mar 11 16:04:43.380: Se0/3/0 IPCP: Install route to 10.100.106.17
*Mar 11 16:04:43.648: Se0/3/0 PPP: Outbound cdp packet dropped
*Mar 11 16:04:43.928: %LINEPROTO-5-UPDOWN: Line protocol on Interface Serial0/3/0, changed state to up
*Mar 11 16:04:43.928: Se0/3/0 PPP: Outbound cdp packet dropped
*Mar 11 16:04:44.920: Se0/3/0 CDPCP: Timeout: State ACKsent
*Mar 11 16:04:44.920: Se0/3/0 CDPCP: O CONFREQ [ACKsent] id 2 len 4
*Mar 11 16:04:44.920: Se0/3/0 CDPCP: I CONFACK [ACKsent] id 2 len 4
*Mar 11 16:04:44.920: Se0/3/0 CDPCP: State is Open
03-11-2010 08:40 AM
Understand that you may checked 1000 times, but symptoms are of "no keepalives" at one side.
5 missed keepalives = 50 seconds
03-11-2010 08:45 AM
That part I get, but what would be blocking the keepalives? The interfaces are configred all default, only IP and encapsulation are configured on both ends.
03-11-2010 08:52 AM
Take debug at the other side, we want to see the keepalive being sent out to begin with.
03-11-2010 08:55 AM
I will need to do that in a maintenance window, it is our core router with hundreds of othe connections.
03-11-2010 09:03 AM
May be a long shot, but the core may be suffering some data structure corruption and not sending keepalive even if supposed to.
Try switching encap to hdlc and back.
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: