frr icon indicating copy to clipboard operation
frr copied to clipboard

isisd: Three-Way Handshake State Change from Expiring

Open beith12 opened this issue 1 year ago • 7 comments
trafficstars

Under certain conditions the ISIS three-way handshake becomes stuck in an 'Expiring' mode and does not clear the neighbor entry. This fix will clear the neighbor entry if this condition becomes true.

beith12 avatar Jun 12 '24 14:06 beith12

@aceelindem Valid point - I can do another commit for the Broadcast circuit type.

beith12 avatar Jun 24 '24 18:06 beith12

I'm okay with a separate commit for broadcast type ... should we open an issue to track it?

riw777 avatar Jun 24 '24 21:06 riw777

I'm okay with a separate commit for broadcast type ... should we open an issue to track it?

I'm okay if you are. Do we need issues for every fix? I haven't been opening them.

aceelindem avatar Jun 24 '24 21:06 aceelindem

Looks good. Normally, you'd squash the two commits though.

I'm okay with a separate commit for broadcast type ... should we open an issue to track it?

I'm okay if you are. Do we need issues for every fix? I haven't been opening them.

nah ... just don't want to forget this additional work needs to be done ...

riw777 avatar Jun 25 '24 11:06 riw777

Looks good. Normally, you'd squash the two commits though.

I'm okay with a separate commit for broadcast type ... should we open an issue to track it?

I'm okay if you are. Do we need issues for every fix? I haven't been opening them.

nah ... just don't want to forget this additional work needs to be done ...

Looks good. Normally, you'd squash the two commits though.

I'm okay with a separate commit for broadcast type ... should we open an issue to track it?

I'm okay if you are. Do we need issues for every fix? I haven't been opening them.

nah ... just don't want to forget this additional work needs to be done ...

The second commit commit handles the broadcast circuit. It is just not squashed - which I agree is ok.

aceelindem avatar Jun 25 '24 15:06 aceelindem

please fix the lints, and then we can try to figure the failures out ... thanks!

riw777 avatar Jun 25 '24 15:06 riw777

Thanks @riw777 - style now corrected

beith12 avatar Jun 26 '24 13:06 beith12

This pull request has conflicts, please resolve those before we can evaluate the pull request.

github-actions[bot] avatar Jul 02 '24 12:07 github-actions[bot]

failing in ospf single switch test ... reunning just the failure to see if we can clear this

riw777 avatar Jul 09 '24 03:07 riw777

failing in ospf single switch test ... reunning just the failure to see if we can clear this

I've seen this failure on PRs as well - I think it is related to the topotest using 8 routers.

aceelindem avatar Jul 09 '24 13:07 aceelindem

please rebase to and let's see if that fixes this ci failure we also need to clear the conflicts now ... :-(

riw777 avatar Jul 09 '24 15:07 riw777

When i did the rebase i got a conflict and found that a previous commit has actually dealt with this same issue.

https://github.com/FRRouting/frr/commit/5009f7539ad6bf496158499917dcf3f7ab760753

beith12 avatar Jul 12 '24 17:07 beith12

it seems like we should close this one then?

@frrbot autoclose in 1 month

riw777 avatar Jul 16 '24 13:07 riw777