Closed ipspace closed 3 months ago
I am able to run that test case and it validates eventually. I did notice a delay, not sure where that is coming from
I had to tweak numerous BGP timers on Cisco IOS to get it to originate prefixes in reasonable time. No such problems on most other platforms.
It is interesting, though, that the prefix specified in the bgp.originate attribute gets advertised pretty quickly while the one configured on the loopback interface is not.
Test passes for me now
Test passes for me now
Correct. The test results @ https://tests.netlab.tools/_html/srlinux-clab-bgp have been updated.
Based on https://netlab-cicd.pages.dev/srlinux/clab/bgp/04-originate.yml-validate.log, it seems that SR Linux (under some circumstances?) does not add its loopback address into the BGP table even when the bgp.advertise_loopback is set to True.
The device that does not originate its loopback prefix in BGP is a standalone router with a single EBGP session, which makes it even more intriguing.
Topology @ https://github.com/ipspace/netlab/blob/dev/tests/integration/bgp/04-originate.yml