Hi guys,
We have a problem on some LDP targeted sessions which hang up every 180 seconds.
The problem occures on a R1 CCR1016 (6.24) and only toward some peers, for example R13 CCR 1009 6.42.12 and R15 RB3011 4.45.9 (R1 is connected to many routers, most of them Mikrotik and no problems, some of them from other vendors, toward none of them there are problems)
sniffing traffic (on R13) we can see:
[*] new tcp connection established ( R13 port 40901 →R1 port 646)
[*] Initialization message, (from R13 Session KeepAlive Time: 180)
[*] some ldp exchange (Label Mapping Messages)
[*] regular LDP UDP Hello Messages from R1 to R13
[*] NO LDP UDP Hello Messages from R1 to R13
[*] regular KeepAlive message from R1 to R13 (every around 60 sec)
[*] NO regular KeepAlive message from R13 to R1
[*] after 180 sec, R1 sends Notification Message and then RESET (Fatal Error, KeepAlive Timer Expired)
[*] new tcp connection starts
The corresponding vpls drops and then returns functioning.
R1 CCR1016 (6.24) is pppoe concentrar, it has many vlps, 98% not affected by this issue (we would not like to update it)
As we have some dozen of routers and it looks like only some combination of RouterOS have this behaviuor, do You have any suggestion on similar issue and or compatibility matrix for LDP between different versions of RouterOS?
Waiting for hearing from You.
Regards
Alberto