Hey guys

I have a CCR2116-12G-4S, this is a core router working mostly as WAN aggregator implementing BGP IPv4 and IPv6 full routes. Since day zero I noticed a estrange behavior with this router, once I have at leas 2-3 sessions (about 2.0+ million routes) any command under ip/route or routing/bgp takes a lot of time and sometime the ip/route service crashes.

When I say crashes I mean if I look for some route using ip/route/print or try to ad a route nothing happens, when this happens all of my BGP announcements stops working, also as example if I open the route/bgp/sessions in Winbox the information shown there is different (uptime, session state, etc) from the same route when searching through SSH. All these symptoms are fixed after manually rebooting the router.

I have noticed what triggers this bug is by doing bulk operations very often, for example when a BGP peer goes down, up and down in less than 30 minutes, or for example when updating BGP filters and manually running routing/bgp/session/refresh commands.

I used to have these same peer and configuration in a CCR 2004 running V6 and this issue never happened. This has become a serious issue because in a few words my core router becomes glitched at any moment.

All my BGP connections are using the affinity value Alone for input and output filters but I see my CPU load is very low so I don't know if this could be causing this issue. I'm attaching a couple screenshots

Any advice or recommendation will be highly appreciated