Some insights about spt-threshold behavior

[stormshield-rlibaert]

Hello Joachim,

We had some more issues when testing some configurations with spt-threshold enabled. We tried to understand the mechanisms used to do the switch to SPT and realised a few things. I have a branch with a few modifications:

stormshield-rlibaert@b8c5442

To sum up:

• There is a bugfix when parsing configuration
• I removed a call to try_switch_to_spt which was done regardless of the provided configuration
• I moved around the init_route initialization function

The last one is the most risky to me as I don't have a wide understanding of pimd, so maybe it brings breaking changes. But the reason for this is that the spt_threshold timer was set to the default value (defined by SPT_THRESHOLD_DEFAULT_INTERVAL), as the configuration of this is done through init_vifs.

[troglobit]

Cool, I say we try it out and see what breaks! 😎

Please submit a pull request and I'll try to find some time to get back to my backlog (of yet unpushed fixes) in pimd. It's easier for me to test as well.

Update: nice catch there with the bugfix! 👍 Much facepalming over here right now .... 🤦‍♂️

(I still haven't gotten around to debugging the BSR + flood-to-RP-when-RP-changes issue, but this might trigger me to get back to it)

[troglobit]

There, tested and merged! Sorry it took me a while to get around to, been busy getting up to speed at a new job.