It's not a version thing, it's what Dennis said... Need to change it
to Type 1 "if installed", not "always". This implies that you're
receiving a default route from your BGP providers at the edges. OSPF
doesn't know anything about what's happening with BGP. So you ask
your peers to send you a default route via BGP (even if they're
sending you a full table) and you configure OSPF to only originate a
default route when one exists in the table of THAT router (which it
would, via your BGP peers). When BGP goes down, your default will go
away from that router and it will stop advertising one to the rest
of the network forcing your other routers to all go the other way.
On 7/19/18 4:26 PM, Nate Burke wrote:
All different mix of ROS Versions. There must be something
special about the default route.
On 7/19/2018 5:20 PM, Bill Prince
wrote:
What version(s) of ROS are you using? We have a couple of
the routers that are back-rev, and we're wondering if this
is a version issue. It's kind of ironic that the only route
that does not propagate is the default route. That seems
counter to what you would expect.
part15sbs{at}gmail{dot}com
Exactly, only the
default route is affected. I see the same thing. All
other routes are fine except for the default route.
On
7/19/2018 5:02 PM, Bill Prince wrote:
It's a mix of Mikrotik routers. Many MIPSBE, CCR,
and X86 architectures.
It seems to be specific to the OSPF routing table
and the default route. If it's not a default route,
but a loopback interface, or other ordinary
interface, the route will propagate from one end to
the other almost immediately. But if it's the
default route, it will drop from ip/routes, but hang
out as a zombie in OSPF.
part15sbs{at}gmail{dot}com
What types of
routers? I've seen this happen on Mikrotik. Any
OSPF change anywhere in the network will fix it.
I have dedicated routers hanging at points on the
network running a script that pings an Internet
Address, and downs it's OSPF network if it's
unreachable. To address this specific senario.
On
7/19/2018 2:02 PM, Sterling Jacobson wrote:
Maybe not quite a similar
problem, but I had forgotten to advertise
BGP one time on one of the routers when I
thought things would magically fail over.
From: AF <[email protected]>
On Behalf Of Bill Prince
Sent: Thursday, July 19, 2018 11:44
AM
To: AFMUG <[email protected]>
Subject: [AFMUG] OSPF not propagating
default route...
Our backbone is 9
hops from one end to the other. We have
2 ASes and 2 exits at either end. "Most"
of the time the traffic is 50/50 right
at the midpoint (4 hops one way, and 4
hops the other). However, when we kill
BGP at one end or the other, the default
routes do not switch over. The two BGP
routers are set to "distribute-default"
"always as type 1". Clearly, this is not
working the way we expect it to. Any
clues about what we are missing here?
part15sbs{at}gmail{dot}com
AF mailing list
[email protected]
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
AF mailing list
[email protected]
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
|
