RE: bri flapping with demand cirquit/igrp redistribution
Your correct Z filter the bri subnet from redistribution into IGRP and your LSA's should not continue to bring the link up. Alan -Original Message- From: Mask Of Zorro [mailto:[EMAIL PROTECTED]] Sent: Tuesday, March 27, 2001 2:32 PM To: [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Re: bri flapping with demand cirquit/igrp redistribution Redistribution brings it up. What happens is, the link is brought up and OSPF forms an adjacency. Then, since it is a demand circuit, periodic LSA's are squelched and OSPF routes associated with those LSA's do not age out of the routing table. Then, the layer 2 portion of the link drops, since there is no interesting traffic. Once the link drops, whatever protocol you are redistributing into OSPF sees it's link drop, and changes its tables in accordance with the topology change. This change gets redistributed into OSPF, and OSPF floods LSA's out announcing the change. These LSA's bring up the link while OSPF converges. Then, after a while, things are stable again, and the link drops - and guess what? That's right! The whole thing starts again... There are ways to stop it. Z From: "George Zhang" [EMAIL PROTECTED] Reply-To: "George Zhang" [EMAIL PROTECTED] To: [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED] Subject: Re: bri flapping with demand cirquit/igrp redistribution Date: Tue, 27 Mar 2001 13:50:52 -0600 I also had the same problem a few days ago. Here is how I fixed it: 1. Isolate the problem by shutting down all other interfaces besides the interfaces between the two related routers; 2. Disable all other routing protocols (non-OSPF ones); 3. Turn off IGRP redistribution to OSPF; 3. Now, verify the ISDN demand circuit. It should be quiet now. 4. Now, start unshut the interfaces you have shut down one by one and verify that the ISDN demand circuit. It should might come up briefly but it should go down and keep quiet after some interfaces are unshut. If the ISDN line keep dialing, you should know which interface is causing the problem. 5. If the ISDN line is still quiet after you unshut all interfaces, turn on your other router protocols such as IGRN, RIP, etc one by one. Again, verify the IDN line after each change as above. 5. If the ISDN line is still quiet after you enable all other routing protocols, turn on your redistribution one by one. Again verify ISDN line along the way. If you follow these steps, you should be able to pin down what is causing your ISDN line to stay up. Hope it helps. George Zhang "Donald B Johnson Jr" [EMAIL PROTECTED] 03/27/01 03:33PM It may keep the connection open though even after there is no intresting traffic. Don - Original Message - From: Alan Basinger [EMAIL PROTECTED] To: Chris Larson [EMAIL PROTECTED]; perez claude-vincent [EMAIL PROTECTED]; Ya Wen [EMAIL PROTECTED]; Patrick Murphy [EMAIL PROTECTED]; Leah Lynch [EMAIL PROTECTED]; 'Jay Chandradas' [EMAIL PROTECTED]; 'Bob Boone' [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Saturday, March 24, 2001 8:37 AM Subject: RE: bri flapping with demand cirquit/igrp redistribution CDP may work at layer 2 but if the line is brought up because of web traffic CDP packets would then traverse the line consume a small amount of bandwidth. I have installed a few ISDN DDR VPN's without turning off CDP and never had and issue but also didn't think about the bandwidth consumption at the time. Alan -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Chris Larson Sent: Saturday, March 24, 2001 9:00 AM To: perez claude-vincent; Ya Wen; Patrick Murphy; Leah Lynch; 'Jay Chandradas'; 'Bob Boone'; [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: RE: bri flapping with demand cirquit/igrp redistribution It may not, I have just always disabled it on dialer links as a habit. It makes sense that it shouldn't since the dialer-list defines layer 3 traffic only. You could always put an access-list on the dialer interface permitting all traffic with the log statement to see exactly what is trying to get accross the line. It will output to the console if you are consoled in. That may help you to see what is bringing the line up. -Original Message- From: perez claude-vincent [mailto:[EMAIL PROTECTED]] Sent: Saturday, March 24, 2001 9:22 AM To: Ya Wen; Patrick Murphy; Leah Lynch; 'Jay Chandradas'; 'Chris Larson'; 'Bob Boone'; [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: RE: bri flapping with demand cirquit/igrp redistribution Can somebody
RE: bri flapping with demand cirquit/igrp redistribution
CDP may work at layer 2 but if the line is brought up because of web traffic CDP packets would then traverse the line consume a small amount of bandwidth. I have installed a few ISDN DDR VPN's without turning off CDP and never had and issue but also didn't think about the bandwidth consumption at the time. Alan -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Chris Larson Sent: Saturday, March 24, 2001 9:00 AM To: perez claude-vincent; Ya Wen; Patrick Murphy; Leah Lynch; 'Jay Chandradas'; 'Bob Boone'; [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: RE: bri flapping with demand cirquit/igrp redistribution It may not, I have just always disabled it on dialer links as a habit. It makes sense that it shouldn't since the dialer-list defines layer 3 traffic only. You could always put an access-list on the dialer interface permitting all traffic with the log statement to see exactly what is trying to get accross the line. It will output to the console if you are consoled in. That may help you to see what is bringing the line up. -Original Message- From: perez claude-vincent [mailto:[EMAIL PROTECTED]] Sent: Saturday, March 24, 2001 9:22 AM To: Ya Wen; Patrick Murphy; Leah Lynch; 'Jay Chandradas'; 'Chris Larson'; 'Bob Boone'; [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: RE: bri flapping with demand cirquit/igrp redistribution Can somebody tell me how come CDP may bring the line up? As you know, it works only in layer 2 as said before. Your dialer-list works at layer 3 4 only. Did I miss something? :-( --- Ya Wen [EMAIL PROTECTED] wrote: Try remove the "log" from the access-list 15 associated with the route-map stuff. Also, you do not need the summary-address under OSPF. -Ya -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Patrick Murphy Sent: Friday, March 23, 2001 5:04 PM To: Leah Lynch; 'Jay Chandradas'; 'Chris Larson'; 'Bob Boone'; [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Re: bri flapping with demand cirquit/igrp redistribution Also check the BRI interface and see if you see IPCDP, it should disapper when you no cdp en! Patrick - Original Message - From: "Leah Lynch" [EMAIL PROTECTED] To: "'Jay Chandradas'" [EMAIL PROTECTED]; "'Chris Larson'" [EMAIL PROTECTED]; "'Bob Boone'" [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Friday, March 23, 2001 7:45 PM Subject: RE: bri flapping with demand cirquit/igrp redistribution I think you normally disable CDP in dialup lines for efficiency. Leah -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Jay Chandradas Sent: Friday, March 23, 2001 2:57 PM To: Chris Larson; Bob Boone; [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Re: bri flapping with demand cirquit/igrp redistribution I am not sure CDP will keep the line up ? And ur interesting traffic is permit ip any any . I dont think CDP will keep the line up. When u do a debug ip pack.. u can nvr see CDP.. CDP is layer 2. my 0.02 - Original Message - From: "Chris Larson" [EMAIL PROTECTED] To: "Bob Boone" [EMAIL PROTECTED]; "Jay Chandradas" [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Friday, March 23, 2001 2:40 PM Subject: RE: bri flapping with demand cirquit/igrp redistribution Will CDP keep the line up? Turn off CDP. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Bob Boone Sent: Friday, March 16, 2001 5:30 PM To: Jay Chandradas; [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Re: bri flapping with demand cirquit/igrp redistribution Yes i do have passive BRI on IGRP, and also, the way it is done now, it restricts ALL networks, if you look at the access-list 15 it has one statement and then explisit deny all. still not working. - Original Message - From: "Jay Chandradas" [EMAIL PROTECTED] To: "Netguy" [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Friday, March 16, 2001 12:22 PM Subject: Re: bri flapping with demand cirquit/igrp redistribution 1. DO u have a passive interface on bri0 under router IGRP 2. I wud do this way !! when u r redisributing into OSPG .. allow only the IGRP networks ( including the network conneted with is running IGRP ) Jay when u r redistributing into - Original Message - From: "Netguy" [EMAIL PROTECTED] To: [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Friday, March 16, 2001 12:01 PM Subject: bri flapping with demand cirquit/igrp redistribution Hello all you happy people. Router A has ospf/igrp mutual redistribution and bri int dialing elsewhere with demand circuit. it keeps flapping. i followed someone's advice and created a route/map filter to filter out bri
RE: OSPF and frame relay issues
Yes I have noticed that also. Though I just assumed it was a new feature much like setting tunnel MTU larger than the physical interface and sending packets with the df bit across :) I was however able to set the priority to 0 on the line by setting it on the interface and not with the neighbor command. This was confirmed to me when I did a "show ip ospf int" and the priority was correct. I did notice when trying to use the neighbor statement that in some routers it never even showed the priority had been changed nor did the command appear in my configs. I then decided to resort to the interface level command which apparently worked. One of those little things that bite... Alan -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Chuck Larrieu Sent: Thursday, March 08, 2001 10:55 PM To: Kevin Mahler; 'Ccielab' (E-mail); Cisco@Groupstudy. Com (E-mail) Subject: RE: OSPF and frame relay issues This brings to mind a question I have about the neighbor priority statement. According to the Doc CD, the default value for neighbor priority is 0 I have on a couple of recent labs entered neighbor a.b.c.d and when doing a show run seen the priority 1 added to the end of the line I have also specifically entered neighbor a.b.c.d priority 0 and done a show run only to see the priority 1 tagged on the end. Has anyone else seen this phenomenon? Chuck -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of Kevin Mahler Sent: Thursday, March 08, 2001 5:03 AM To: 'Ccielab' (E-mail); Cisco@Groupstudy. Com (E-mail) Subject:RE: OSPF and frame relay issues You are on the right track. Router 3 does need to be the DR but router's 4 and 5 need to NOT be a BDR. Also you will need neighbor statements. You can put the priority on the neighbor statement. So router 1 has a priority of say 90 and routers 4 and 5 have a priority of 0 to keep them from ever becomming a BDR or DR. router ospf 9 neighbor 133.9.4.4 priority 0 neighbor 133.9.5.5 priority 0 do the same on routers 4 and 5 with neighbor statements back to router 3. Kevin - Original Message - From: McCallum, Robert [EMAIL PROTECTED] To: 'Ccielab' (E-mail) [EMAIL PROTECTED]; Cisco@Groupstudy. Com (E-mail) [EMAIL PROTECTED] Sent: Thursday, March 08, 2001 6:22 AM Subject: OSPF and frame relay issues I am at the moment ploughing through a lab which for the life of me I can't get it to work. I'm going mad!!!. Scenario is this Router4---Router3Router5 ! ! ! ! Router6 There is of course a frame relay cloud making these connections. Connections are as above i.e. Router4 has a pvc to Router3, Router5 has a pvc to Router3 and Router6 has a pvc to Router3. Conditions are --- Router3 uses 1 sub interface. No more sub interfaces allowed on any router. All routers share the 10.10.X.X domain. All routers are in area 0. Only one pvc can be used on Routers 4,56 i.e. only to router3. You cannot use the command IP OSPF NETWORK nor can you use multiple frame relay map statements. SO the first thing that sprung to my mind is make router3 the DR and form manual neighbor relationships. WRONG. this doesn't work. Router3 can ping everything, All other routers can only ping router3. Every router has full routing tables of the full network. PING end to end is a complete no no. Any help? configs below. ROUTER3 version 12.0 service timestamps debug uptime service timestamps log uptime no service password-encryption ! hostname router3 ! ! ip subnet-zero no ip domain-lookup ! ! ! ! interface Serial0/0 no ip address no ip directed-broadcast encapsulation frame-relay no ip mroute-cache no fair-queue ! interface Serial0/0.1 multipoint ip address 10.10.1.3 255.255.0.0 no ip directed-broadcast frame-relay interface-dlci 304 frame-relay interface-dlci 305 frame-relay interface-dlci 306 ! interface TokenRing0/0 no ip address no ip directed-broadcast shutdown ring-speed 16 ! interface Virtual-TokenRing3 ip address 10.1.3.1 255.255.0.0 no ip directed-broadcast ring-speed 16 ! router ospf 64000 network 10.1.0.0 0.0.255.255 area 1 network 10.10.0.0 0.0.255.255 area 0 neighbor 10.10.1.5 neighbor 10.10.1.4 neighbor 10.10.1.6 ! ip classless ! ! line con 0 transport input none line aux 0 line vty 0 4 login ! end ROUTER4 version 12.1 service timestamps debug uptime service timestamps log uptime no service password-encryption ! hostname router4 ! ! ! ! ! ! ip subnet-zero no ip domain-lookup ! cns event-service server ! ! interface Ethernet0/0 ip address 137.20.20.1 255.255.255.0 no keepalive ! interface Serial1/0 ip address 10.10.1.4 255.255.0.0 encapsulation frame-relay ip ospf priority 0 clockrate 128000 frame-relay interface-dlci 403 ! router ospf 64000 network 10.10.0.0 0.0.255.255 area 0
