Hello,
A while ago I made a post related to an issue with OSPF and BGP where we saw some peers flapping and then all the OSPF adjacency going down as well (basically, as soon the flapping ended, the OSPF was stable again), you can check it here: http://forum.mikrotik.com/t/serious-issue-with-bgp-and-ospf/106342/5
I’ve tried to recreate this issue and I managed to do so, basically “inducing” bgp peering flappings. The device seems to keep forwarding traffic using the static routes, but all the OSPF neighbors are lost. One of the weird things here is that I also notice the routes are deleted really slow from FIB and that the CPU keeps really low on usage all the time (I’m using CCRs), so it doesn’t really look like it’s a hardware issue (lack of capacity) but a software one.
So here comes my question for the community and specially MikroTik staff, does RouterOS use the same process for all routing related, as in, the process called “routing” manages everything related to OSPF, BGP, RIP, etc. And if such, if this process for any reason hangs, freezes or just resets itself, every protocol is affected? I’d really like an answer on this, since (to me at least) this is a serious matter.