Use varied command timeout length for different vcs commands

[sdboyer]

We use a system of subprocess monitoring (basically, we watch stderr/stdout for updates) to determine if a command is making progress. Commands that aren't making progress are terminated if a configurable timeout is exceeded since the last visible progress was made.

However, we aren't taking advantage of that configurability. Currently, all command invocations of underlying vcs binaries use a 2 minute timeout. This is not great - there's no reason we should allow such long timeouts on some of our commands.

So, let's expand the runFromRepoDir() func to take a configurable argument, and make more granular decisions for the different commands. This won't really solve any of the hang issues people have been having on its own, but it at least may help us isolate which commands are giving us problems.

[erizocosmico]

May I take this one? 😄

[sdboyer]

absolutely! good to have you back on this 😀🎉

…

On July 21, 2017 6:19:42 PM EDT, Miguel Molina ***@***.***> wrote: May I take this one? 😄 -- You are receiving this because you authored the thread. Reply to this email directly or view it on GitHub: #872 (comment)