Hello,
I have an issue with ibgp peer flapping with the below error.
Log Message; {l_addr: x.x.x.90@*4, r_addr: x.x.x.x@*4} Received Notification [ code: (3,4), msg: (UPDATE error: bad attribute flags), data: D011000E02030000FDFC0000FFF000021BC1]
To me, the extended length flag is being set incorrectly from the MikroTik. Is this a bug?
Please see packet capture.
model: CCR2004-16G-2S+
revision: r3
firmware-type: al64
current-firmware: 7.20.8
Peer device; MS Server 2022 Version 21H2
BGP Configuration
name="0034 internal server test" instance=instance-1
remote.address=x.x.x.10 .as=65039
local.address=x.x.x.90 .role=ibgp
routing-table=0034 vrf=0034 as=65039 nexthop-choice=force-self multihop=yes
output.redistribute=bgp .filter-chain=plc-BGP-0034-internal-export
input.filter=plc-BGP-0034-internal-server-import
chain=plc-BGP-0034-internal-export rule="if (protocol bgp && dst == 10.0.0.0/8 || dst == 172.16.0.0/12) { set bgp-local-pref 200; accept}"
chain=plc-BGP-0034-internal-export rule="if (protocol bgp && dst in 10.0.0.0/8 || dst in 172.16.0.0/12 || dst in 192.168.0.0/16) {reject} else {accept}"
chain=plc-BGP-0034-internal-export rule="if (protocol bgp && dst == 0.0.0.0/0) {reject} else {set bgp-local-pref 200; accept}"
chain=plc-BGP-0034-internal-server-import rule="if (dst in 0.0.0.0/0) {reject}
Please be aware, the same setup works fine on a Juniper ACX which the MikroTik is replacing.