[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Ospfd looses 'cost' setting when interface goes down and then up again.
From: |
Łukasz Grochal |
Subject: |
Ospfd looses 'cost' setting when interface goes down and then up again. |
Date: |
Wed, 13 Feb 2002 22:16:53 +0100 |
User-agent: |
Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.1 (i386-debian-linux-gnu) |
Hello,
I'll try to keep this report as formal as possible, so here it goes:
1. I'm running zebra from CVS, checked out on 31.01.2002.
2. The problem affects ospfd, but I also use zebra itself and ripd.
I use Debian/GNU Linux 2.2 (stable) with kernel 2.4.17. Ospfd
was compiled without any special options like --enable-nssa or
such.
3. I use OSPF on number of point-to-point VPN links. When I set
specific cost for an interface (with 'ip ospf cost 3' for example),
and then the interface goes down and up for some reason (like
temporary network outage causing encrypting daemon to exit), zebra
is running all right when the interface comes back, except that the
cost is reset to default. This requires shutting ospfd down and
starting it again anytime such outage occurs.
4. Here is an example setup before the interface goes down:
ospfd# show running-config
Current configuration:
!
[...]
!
interface cipcb3
ip ospf network point-to-point
ip ospf cost 3
!
... when cipcb3 goes down and I start the link again, runtime configuration
reads:
!
interface cipcb3
ip ospf network point-to-point
!
Saved configuration retains - of course - the cost that was set.
5. n/a
6. Here is a snip from syslog:
Feb 13 22:07:20 walker ospfd[10861]: Zebra: Interface[cipcb3] state change to
down.
Feb 13 22:07:20 walker ospfd[10861]: NSM[cipcb3:10.2.3.1:10.3.0.1]: Status
change Full -> Down
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): Start
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): looked through
areas
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): bb_configured: 0
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): bb_act_attached: 0
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): areas_configured:1
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status():
areas_act_attached: 1
Feb 13 22:07:20 walker ospfd[10861]: nsm_change_status(): scheduling new
router-LSA origination
Feb 13 22:07:20 walker ospfd[10861]: LSA[Type1]: Cancel previous router-LSA
timer
Feb 13 22:07:20 walker ospfd[10861]: LSA[Type1]: Scheduling router-LSA
origination right away
Feb 13 22:07:20 walker ospfd[10861]: ISM[cipcb3:10.2.3.1]: ism_ignore called
Feb 13 22:07:20 walker ospfd[10861]: ISM[cipcb3:10.2.3.1]: Point-To-Point
(NeighborChange)
Feb 13 22:07:20 walker ospfd[10861]: NSM[cipcb3:10.2.3.1:10.3.0.1]: neighbor
deleted
Feb 13 22:07:20 walker ospfd[10861]: ISM[cipcb3:10.2.3.1]: Point-To-Point
(InterfaceDown)
Feb 13 22:07:20 walker ospfd[10861]: ISM[cipcb3:10.2.3.1]: Status change
Point-To-Point -> Down
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): Start
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): looked through
areas
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): bb_configured: 0
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): bb_act_attached: 0
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): areas_configured:
1
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status():
areas_act_attached: 1
Feb 13 22:07:20 walker ospfd[10861]: LSA[Type1]: Cancel previous router-LSA
timer
Feb 13 22:07:20 walker ospfd[10861]: LSA[Type1]: Scheduling router-LSA
origination right away
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): Start
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): looked through
areas
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): bb_configured: 0
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): bb_act_attached: 0
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status(): areas_configured:
1
Feb 13 22:07:20 walker ospfd[10861]: ospf_check_abr_status():
areas_act_attached: 1
Feb 13 22:07:20 walker ospfd[10861]: interface 10.2.3.1 leave AllSPFRouters
Multicast group.
Feb 13 22:07:20 walker ospfd[10861]: Timer[router-LSA]: (router-LSA Refresh
expire)
Feb 13 22:07:20 walker ospfd[10861]: LSA[Type1]: Create router-LSA instance
Feb 13 22:07:20 walker ospfd[10861]: counting fully adjacent virtual neighbors
in area 10.0.0.0
Feb 13 22:07:20 walker ospfd[10861]: there are 0 of them
Feb 13 22:07:20 walker ospfd[10861]: LSA[Type1]: Set link Point-to-Point
Feb 13 22:07:20 walker ospfd[10861]: SPF: calculation timer scheduled
Feb 13 22:07:20 walker ospfd[10861]: SPF: calculation timer delay = 5
Feb 13 22:07:20 walker ospfd[10861]: LSA: freed 0x810c550
Feb 13 22:07:20 walker ospfd[10861]: LSA[Type1:10.2.0.1]: data freed 0x8120718
Feb 13 22:07:20 walker ospfd[10861]: LSA[Type1]: ID 10.2.0.1 is self-originated
Feb 13 22:07:20 walker ospfd[10861]: LSA[Type1,id(10.2.0.1),ar(10.2.0.1)]:
Install router-LSA to Area 10.0.0.0
[...] and then:
Feb 13 22:07:21 walker ospfd[10861]: Zebra: interface add cipcb3 index 22 flags
144 metric 1 mtu 1442
Feb 13 22:07:21 walker ospfd[10861]: ISM[cipcb3:10.2.3.1]: Down (InterfaceDown)
Feb 13 22:07:21 walker ospfd[10861]: can't setsockopt IP_DROP_MEMBERSHIP
(AllSPFRouters): Cannot assign requested address
Feb 13 22:07:21 walker ospfd[10861]: Zebra: Interface[cipcb3] state change to
up.
Feb 13 22:07:21 walker ospfd[10861]: interface 10.2.3.1 join AllSPFRouters
Multicast group.
Feb 13 22:07:21 walker ospfd[10861]: ISM[cipcb3:10.2.3.1]: Down (InterfaceUp)
Feb 13 22:07:21 walker ospfd[10861]: ISM[cipcb3:10.2.3.1]: Status change Down
-> Point-To-Point
Feb 13 22:07:21 walker ospfd[10861]: ospf_check_abr_status(): Start
Feb 13 22:07:21 walker ospfd[10861]: ospf_check_abr_status(): looked through
areas
Regards,
--
(-) Łukasz Grochal address@hidden
(for PGP key visit:)
_____________________________________________ http://www.rotfl.eu.org/ __
... all in all it's just another rule in the firewall. /Ping Flood/
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- Ospfd looses 'cost' setting when interface goes down and then up again.,
Łukasz Grochal <=