On Mon, Mar 10, 2014 at 01:50:46PM +0000, Justin Marshall wrote: > I will answer this as soon as I catch it again... It's so > intermittent, hard to catch it unless I stare at it...(most of the > times I don't have to wait too long) > > > What does "/ip route print where dst-address in x.x.213.0/24" look > > like on MikroTik A and B? > > Tower A: > [admin@Tower-VB_SBA] /ip route> print where dst-address in x.x.213.0/26 > Flags: X - disabled, A - active, D - dynamic, C - connect, S - static, r - > rip, b - bgp, o - ospf, m - mme, > B - blackhole, U - unreachable, P - prohibit > # DST-ADDRESS PREF-SRC GATEWAY DISTANCE > 28 ADo x.x.213.0/26 198.172.214.66 110
I wanted to see "in x.x.213.0/24" so that we could see if there was any covering route which would send traffic a different direction when the /26 went away. It may have a covering route in the /20 or /16 or whatever you were allocated. I probably should have asked for : "/routing ospf lsa print detail where id in x.x.213.0/24" That can be a lot of data. I am not sure I know how to interpret that data fully. I'm just trying to find a way to look for multiple possible paths. It's entirely possible that this is the symptom of a bug. Make a supout while it's happening the next time and send it to [email protected]. Maybe they'll see something they can fix. But they will probably just say to upgrade everything to 6.11 beta. > Tower B: > [admin@Tower-FP_Selvitz] /ip route> print where dst-address x.x.213.0/26 > Flags: X - disabled, A - active, D - dynamic, C - connect, S - static, r - > rip, b - bgp, o - ospf, m - mme, > B - blackhole, U - unreachable, P - prohibit > # DST-ADDRESS PREF-SRC GATEWAY DISTANCE Whoa, this is the MikroTik which is supposed to have x.x.213.1 on a local interface? Did you just paste one too few lines? We should have seen a connected route on that box. If that's not just a copy and paste-o, are we sure the IP is defined on the bridge interface rather than an underlying physical interface? I think it can work that way sometimes, when the moon is waning and pluto is at parahelion, or some such thing. I think I've seen that work on my network while I was shuffling things. That's probably a bug in itself. But it has been nice to not lose access to the box on those occasions when I forgot I was connected to the IP on the interface I was putting into the bridge from 2 hours away. > > Is x.x.213.1 the OSPF router ID of your MikroTik at Tower B? Have > > you made sure your all of your OSPF router IDs are all manually > > specified and unique within your network? > > No, the router ID is x.x.214.66. And Yes, all the OSPF router IDs are > all manually specified and unique. > > > Does the problem resolve itself after a bit? how long? or do you > > have to take some action to clear it? > > Yes, the problem resolves itself after about 8 to 10 seconds, haven't > seen it go much longer than that... > > Thanks again for any input you can and have provided :) > > Justin > > > > -----Original Message----- > From: [email protected] > [mailto:[email protected]] On Behalf Of Scott Lambert > Sent: Friday, March 07, 2014 5:15 PM > To: Mikrotik discussions > Subject: Re: [Mikrotik] OSPF Issue > > On Wed, Mar 05, 2014 at 08:05:16PM +0000, Justin Marshall wrote: > > Hi, > > > > We are having an OSPF issue. At least i'm assuming it's an OSPF > > issue because there are no static routes involved. > > > > We have our core router (imagestream) connected via Fiber to a > > Mikrotik at Tower A, Which in turn is connected via 2 RB411AH's with > > Atheros AR922X's to a Mikrotik at Tower B. Every 500-1000 pings we > > are loosing pings to 1 of the IP addresses on a bridge interface on > > the Mikrotik at Tower B. When this happens I see an IP address that > > is bound to one of the interfaces on our imagestream, show up on an > > MTR like so... > > > > How is the wireless link from Tower A to Tower B configured? Are the > RB411s bridged or do they participate in OSPF? How are the wireless > AP and station interfaces configured? > > What is the OSPF interface configuration across the wireless link? > > Is the ImageStream up to date? > > Are you losing OSPF neighbor associations on the MikroTiks? Should > show in the MikroTik log or you can just see how long the OSPF > neighbor association has been installed. > > > Host Loss% Snt > > Last Avg Best Wrst StDev > > 1. 10.192.172.1 0.0% 3642 > > 0.7 0.4 0.4 27.1 1.0 > > 2. x.x.99.1 0.3% 3642 > > 0.8 0.6 0.5 36.5 1.2 > > 3. x.x.214.74 0.0% 3642 > > 9.7 0.7 0.6 33.4 1.3 > > x.x.36.209 > > 4. x.x.214.66 0.0% 3642 > > 8.0 10.3 0.7 75.1 9.1 > > x.x.99.1 > > 5. x.x.213.1 0.2% 3642 > > 32.7 26.1 0.8 288.3 21.3 > > x.x.36.209 > > > > > > This is how it should look: > > Host Loss% Snt > > Last Avg Best Wrst StDev > > 1. x.x.172.1 0.0% 4 > > 0.8 0.9 0.5 1.4 0.4 > > 2. x.x.99.1 0.0% 4 > > 0.7 0.8 0.6 1.0 0.2 > > 3. x.x.214.74 0.0% 4 > > 1.2 1.1 0.7 1.7 0.4 > > 4. x.x.214.66 0.0% 3 > > 29.8 15.5 5.7 29.8 12.7 > > 5. x.x.213.1 0.0% 3 > > 15.3 14.4 13.8 15.3 0.8 > > > > > > Both Mikrotik's are on 6.7, running OSPF. What we can't figure out > > is why this x.x.36.209 keeps showing up in the route. While this is > > happening, pings come back as: TTL exceeded. Also we aren't loosing > > any pings to any other IP's on the Mikrotik in question at Tower B, > > nor are we loosing pings to any other IP addresses bound to that > > bridge interface. > > > > The IP that we are loosing pings to is a public IP vs the others > > being private and not being in the /routing ospf networks. This may > > or may not be relevant. There is also no NAT involved whatsoever. > > I may be interpreting your message incorrectly. Could you clarify > if x.x.36.209 is an IP on an interface on the ImageStream? Or is it > on a router at Tower C which may also be connected to Tower A via a > wireless link? I suspect it is on the ImageStream. > > So, only the one subnet is falling out? At the time that you get > looping traceroutes for 213.1, other subnets run clean? > > Are you only looping for the one IP in that subnet or for all hosts in > the subnet? > > What does "show ip ospf route" look like from the ImageStream for > anything in that /24? During the problem and while normal. > > What does "/ip route print where dst-address in x.x.213.0/24" look > like on MikroTik A and B? > > Is x.x.213.1 the OSPF router ID of your MikroTik at Tower B? Have you > made sure your all of your OSPF router IDs are all manually specified > and unique within your network? > > Does the problem resolve itself after a bit? how long? or do you have > to take some action to clear it? > > RouterOS 6.x has been better for us than 5.x for wierd OSPF issues. > But it still seems a bit delicate. Especially if you have a > misconfiguration somewhere or the wrong OSPF network type on a > wireless bridge. > -- Scott Lambert KC5MLE Unix SysAdmin [email protected] How to be a "computer expert," http://www.xkcd.com/627/ _______________________________________________ Mikrotik mailing list [email protected] http://mail.butchevans.com/mailman/listinfo/mikrotik Visit http://blog.butchevans.com/ for tutorials related to Mikrotik RouterOS
