test: Fix intermittent timeout in p2p_permissions.py #26854

The sync is based on bytesrecv_per_msg["verack"]. However, the bytes are counted before processing the message, so they are not sufficient to ensure the connection is fully up.

<!--e57a25ab6845829454e8d69fc972939a-->

The following sections might be updated with supplementary metadata relevant to reviewers and maintainers.

<!--021abf342d371248e50ceaed478a90ca-->

Reviews

See the guideline for information on the review process.

If your review is incorrectly listed, please react with 👎 to this comment and the bot will ignore it on the next update.

This runs into other timeouts (p2p_node_network_limited.py, see CI) although the test passes for me locally.

Thanks. Fixed by replacing == with >=. (There can only be one verack msg, but there can be multiple pong msgs per connection)

Is this also a fix for the mempool timeouts in #26440?

Seems likely, but not sure if this is possible to tell without having Cirrus CI upload the full datadir + test log of the failure

ACK fa1bf4e7052e617dd0e5c8c54969d84314af9577

This makes the test code dependent on their being a ping right after connection setup (which is the case today because m_ping_start=0us initially, but one could imagine other ways of initialization.

Another possibility might be to expose fSuccessfullyConnected via RPC (e.g. in getpeerinfo) and use that, but I'm not sure if that is better if that field is not really interesting to user and just used by tests.

(which is the case today because m_ping_start=0us initially, but one could imagine other ways of initialization.

In the unlikely case that this is changed, it should be possible to just call the ping RPC at the right time to achieve the same.

ACK fa1bf4e7052e617dd0e5c8c54969d84314af9577

I verified I could reproduce the failure on master with rr but not on this branch.