fix(ping): gracefully handle remote stream closure.

[privacyguard]

Gracefully handle remote stream closure

Description

This PR fixes the UnexpectedEOFError issue. This issue happens because the ping protocol does not handle the case where the remote peer closes the stream and with the introduction of connectionMonitor this became an issue. It opens a stream for the ping protocol, sends a ping, gets the response then closes the stream. The local peer throws an exception and triggers a yamux reset.

Notes & open questions

We only tested these changes in the context described above. Not sure if there are other cases to handle.

Change checklist

• [x] I have performed a self-review of my own code
• [ ] I have made corresponding changes to the documentation if necessary (this includes comments as well)
• [ ] I have added tests that prove my fix is effective or that my feature works

[achingbrain]

Thanks for opening this. I think we should handle the case where the remote has pinged us and closed the connection, it's in the spec so we shouldn't be logging error messages.

I can't push changes to this branch (possibly due to your org settings) so I've opened #2822 which incorporates the commit from here.

[achingbrain]

Closing in favour of #2822

[privacyguard]

@achingbrain no worries! Glad we could help.