03-22-2018 06:13 AM - edited 03-08-2019 02:21 PM
Good day;
I have a pair of layer 3 switches one is an Arista switch and the other is a Cisco switch. They are connected together over a PTP 10 gig circuit. What is going on is that about 2-3 times a day I get am getting am OSPF Adjacency teardown/establishment dude to DD mismatch. Now I have checked the MTU size on both ends and they are set to 1500 bytes, carries says that the circuit is clean. The only other difference I see on the config is that the Arista switch the max/min hold times for SPFs are 5000msec while on Cisco it is set to 10000msec. Is there anything else I am missing? What else can I check?
Thank you in advance!!!
Warren
Solved! Go to Solution.
03-22-2018 07:50 AM - edited 03-22-2018 08:03 AM
Both devices saw the link drop at the same time. That, in turn will of course drop your OSPF adjacency.
I believe, you'll need to investigate and remediate the cause of the link dropping.
03-22-2018 08:27 AM
Hi,
You are facing interface flapping issue.
Mar 22 01:19:09: %LINK-SW2_SP-3-UPDOWN: Interface TenGigabitEthernet1/4/4, changed state to down
Mar 22 01:28:00: %LINK-3-UPDOWN: Interface TenGigabitEthernet1/4/4, changed state to up
!
!
Mar 22 01:19:09 SW-AUCL-CORE1 Ebra: 12394: %LINEPROTO-5-UPDOWN: Line protocol on Interface Ethernet21, changed state to down
Mar 22 01:19:09 SW-AUCL-CORE1 Rib: %OSPF-4-OSPF_ADJACENCY_TEARDOWN: NGB 10.16.160.9, interface 10.0.2.45 adjacency dropped: interface went down, state was: FULL
Mar 22 01:19:09 SW-AUCL-CORE1 Rib: %PIM-5-NBRCHG: neighbor 10.0.2.44 DOWN on interface et21
Mar 22 01:28:00 SW-AUCL-CORE1 Ebra: 12395: %LINEPROTO-5-UPDOWN: Line protocol on Interface Ethernet21, changed state to up
Please check your media, connector or any bad hardware.
Regards,
Deepak Kumar
03-22-2018 06:26 AM
Hi,
Some parameters must match as:
1. Hello and Dead intervals
2. Area ID
3. Authentication type and password
4. Stub Area flag
5. Subnet ID and Subnet mask
Please check which are applied to your configuration.
Regards,
Deepak Kumar
03-22-2018 06:41 AM
Warren, Deepak,
Please allow me to join.
The DD packets (Database Description) would be used only when the OSPF routers detect themselves and go through ExStart and Exchange states. DD packets are not used when the OSPF routers are in the FULL mode anymore. Any problem reported with the DD packets is therefore somehow related to the OSPF neighbors falling back from the FULL adjacency state to a lesser state, and attempting to come back to the FULL state again.
Do you believe you could share the exact logging messages both from the Cisco and the Arista devices? Their exact wording and contents are very important. Thanks!
Best regards,
Peter
03-22-2018 07:03 AM
Hi Peter and Deepak
Deepak
all configs match thank you for the suggestion I only checked MTU size
Peter
agreed I thought perhaps since they where on a PTP circuit that the circuit bounce but carrier says it is all clear here is the exact log messages
Cisco:
Mar 22 01:19:09: %LINEPROTO-5-UPDOWN: Line protocol on Interface TenGigabitEthernet1/4/4, changed state to down
Mar 22 01:19:09: %LINEPROTO-SW2_SP-5-UPDOWN: Line protocol on Interface TenGigabitEthernet1/4/4, changed state to down
Mar 22 01:19:09: %LINK-3-UPDOWN: Interface TenGigabitEthernet1/4/4, changed state to down
Mar 22 01:19:09: %PIM-5-NBRCHG: neighbor 10.0.2.45 DOWN on interface TenGigabitEthernet1/4/4 DR
Mar 22 01:19:09: %PIM-5-DRCHG: DR change from neighbor 10.0.2.45 to 10.0.2.44 on interface TenGigabitEthernet1/4/4
Mar 22 01:19:09: %OSPF-5-ADJCHG: Process 100, Nbr 10.2.1.1 on TenGigabitEthernet1/4/4 from FULL to DOWN, Neighbor Down: Interface down or detached
Mar 22 01:19:09: %LINK-SW2_SP-3-UPDOWN: Interface TenGigabitEthernet1/4/4, changed state to down
Mar 22 01:28:00: %LINK-3-UPDOWN: Interface TenGigabitEthernet1/4/4, changed state to up
Mar 22 01:28:00: %LINK-SW2_SP-3-UPDOWN: Interface TenGigabitEthernet1/4/4, changed state to up
Mar 22 01:28:00: %LINEPROTO-5-UPDOWN: Line protocol on Interface TenGigabitEthernet1/4/4, changed state to up
Mar 22 01:28:00: %LINEPROTO-SW2_SP-5-UPDOWN: Line protocol on Interface TenGigabitEthernet1/4/4, changed state to up
Mar 22 01:28:01: %PIM-5-DRCHG: DR change from neighbor 0.0.0.0 to 10.0.2.44 on interface TenGigabitEthernet1/4/4
Mar 22 01:28:03: %PIM-5-NBRCHG: neighbor 10.0.2.45 UP on interface TenGigabitEthernet1/4/4
Mar 22 01:28:03: %PIM-5-DRCHG: DR change from neighbor 10.0.2.44 to 10.0.2.45 on interface TenGigabitEthernet1/4/4
Mar 22 01:28:07: %OSPF-5-ADJCHG: Process 100, Nbr 10.2.1.1 on TenGigabitEthernet1/4/4 from LOADING to FULL, Loading Done
Arista:
Mar 22 01:19:09 SW-AUCL-CORE1 Ebra: 12394: %LINEPROTO-5-UPDOWN: Line protocol on Interface Ethernet21, changed state to down
Mar 22 01:19:09 SW-AUCL-CORE1 Rib: %OSPF-4-OSPF_ADJACENCY_TEARDOWN: NGB 10.16.160.9, interface 10.0.2.45 adjacency dropped: interface went down, state was: FULL
Mar 22 01:19:09 SW-AUCL-CORE1 Rib: %PIM-5-NBRCHG: neighbor 10.0.2.44 DOWN on interface et21
Mar 22 01:28:00 SW-AUCL-CORE1 Ebra: 12395: %LINEPROTO-5-UPDOWN: Line protocol on Interface Ethernet21, changed state to up
Mar 22 01:28:03 SW-AUCL-CORE1 Rib: %PIM-5-NBRCHG: neighbor 10.0.2.44 UP on interface et21
Mar 22 01:28:07 SW-AUCL-CORE1 Rib: %OSPF-4-OSPF_ADJACENCY_ESTABLISHED: NGB 10.16.160.9, interface 10.0.2.45 adjacency established
Mar 22 06:08:49 SW-AUCL-CORE1 Rib: %OSPF-4-OSPF_ADJACENCY_TEARDOWN: NGB 10.16.160.9, interface 10.0.2.45 adjacency dropped: DD packet mismatch, state was: FULL
Mar 22 06:08:49 SW-AUCL-CORE1 Rib: OSPF: invalid DD packet from 10.0.2.44: Initialize flag mismatch (15)
Mar 22 06:08:54 SW-AUCL-CORE1 Rib: %OSPF-4-OSPF_ADJACENCY_ESTABLISHED: NGB 10.16.160.9, interface 10.0.2.45 adjacency established
03-22-2018 07:50 AM - edited 03-22-2018 08:03 AM
Both devices saw the link drop at the same time. That, in turn will of course drop your OSPF adjacency.
I believe, you'll need to investigate and remediate the cause of the link dropping.
03-22-2018 08:27 AM
Hi,
You are facing interface flapping issue.
Mar 22 01:19:09: %LINK-SW2_SP-3-UPDOWN: Interface TenGigabitEthernet1/4/4, changed state to down
Mar 22 01:28:00: %LINK-3-UPDOWN: Interface TenGigabitEthernet1/4/4, changed state to up
!
!
Mar 22 01:19:09 SW-AUCL-CORE1 Ebra: 12394: %LINEPROTO-5-UPDOWN: Line protocol on Interface Ethernet21, changed state to down
Mar 22 01:19:09 SW-AUCL-CORE1 Rib: %OSPF-4-OSPF_ADJACENCY_TEARDOWN: NGB 10.16.160.9, interface 10.0.2.45 adjacency dropped: interface went down, state was: FULL
Mar 22 01:19:09 SW-AUCL-CORE1 Rib: %PIM-5-NBRCHG: neighbor 10.0.2.44 DOWN on interface et21
Mar 22 01:28:00 SW-AUCL-CORE1 Ebra: 12395: %LINEPROTO-5-UPDOWN: Line protocol on Interface Ethernet21, changed state to up
Please check your media, connector or any bad hardware.
Regards,
Deepak Kumar
03-22-2018 09:05 AM
I believe its a Physical layer issue between anywhere both nodes.
Please update whenever you will be able to fix it.
Thanks,
Lovejit
03-22-2018 09:13 AM
ok issue solved thank you all for your help but here is what happened I called back to the
carrier to ask if there was any issues last night, originally when I cal they person said no issues.
When I called back again after talking to you all I was told their was maintenance going on last
night that cause a hit on my circuit so that is what caused the interface to go down and back up, which in turn cause the OSPF ADJ change as well.
Thank you again for your help much appreciated!!!!
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: