I have the following topology:
R1--------R2=======R3
R2 has two paths to reach R3 (lets call one link as "main link" and other link as "bypass link")
There is MPLS-TE tunnel configured from Headend(R1) to tailend(R3)
This tunnel only has one path (strict explicit hops) configured
If there is a failure on the main link between R2 and R3, FRR kicks in and the PLR (R2) starts sending traffic over the bypass link.
The problem I am seeing in the debugs is as follows:
i. As soon as the failure happens on R2 and it switches traffic over the bypass tunnel, the PLR (as expected) sends a PERR message (Code 25 subcode 3) to the headend which is fine.
ii. But then on the headend, I constantly see a PERR message with (Code 24 subcode 1)
iii. The above causes the headend to delete the PSB associated with the LSP which is being protected using the FRR bypass tunnel on PLR
My question is:
- Is it normal for the PLR to send a PERR message (Code 24 subcode 1) after initially sending the PERR with Code 25 subcode 3 - even when its actually providing local protection for the tunnel because IMO it should be sufficient for the PLR to send the PERR with Code 25 subcode 3 ?
- If its normal for PLR to send this PERR msg after initial 25/3 coded message, I'd like to know more about the behavior of PLR and headend during failure and especially after the link has recovered (note that I have read RFC 4090 but need more details so if anyone knows deeper details about the behavior please let me know).
