Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Match BGP connection source address to router ID #422

Closed
briantopping opened this issue Apr 6, 2019 · 7 comments
Closed

Match BGP connection source address to router ID #422

briantopping opened this issue Apr 6, 2019 · 7 comments

Comments

@briantopping
Copy link
Contributor

briantopping commented Apr 6, 2019

Bug Report

What happened:

Peering BIRD and MLB on the same host node was possible by fiddling with the addresses that BIRD and MLB expect to see each other on. The processes were peering, but the origin for the propagated routes from MLB was somehow showing as the upstream peer to the BIRD instance, not the BIRD host itself. This caused routing loops as soon as the unaggregated MLB routes were propagated.

What you expected to happen:

The goal was that the upstream would know what host to direct traffic at per address. But with the routing loop, there was no way to get that properly working.

Resolution

Forthcoming small patch will set the source address of the TCP connection to the router ID specified in configuration, rather than relying on the Go resolver to provide one. With the patch, MLB connects to BIRD from the matching router ID. This apparently avoids some issue that might more properly be rejected by it for a session opening.

@russellb
Copy link
Collaborator

The associated PR was closed over a year ago, so I assume this issue is no longer needed.

@briantopping
Copy link
Contributor Author

@russellb Problem still exists, unfortunately. I just keep manually patching on my end when I need to upgrade.

@russellb
Copy link
Collaborator

@russellb Problem still exists, unfortunately. I just keep manually patching on my end when I need to upgrade.

Sorry about that. I'm closing stuff a bit quickly in a cleanup round. Should the PR be reopened then? Is that the patch you're carrying to make your environment work at each upgrade?

@russellb russellb reopened this Jun 18, 2020
@briantopping
Copy link
Contributor Author

Should the PR be reopened then? Is that the patch you're carrying to make your environment work at each upgrade?

Probably. I don't remember what the objections were, but I've been using it cleanly for a long time.

@russellb
Copy link
Collaborator

Should the PR be reopened then? Is that the patch you're carrying to make your environment work at each upgrade?

Probably. I don't remember what the objections were, but I've been using it cleanly for a long time.

It looks like more of a misunderstanding about whether it was still needed?

@briantopping
Copy link
Contributor Author

Not really sure. I could update the PR if that's desirable.

@briantopping
Copy link
Contributor Author

Closed in deference to #670

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants