ENH:Attempt to add timeout to get, put, post

[Deleted user (ghost)]

No description provided.

[Deleted user (ghost)]

Dan Allan (@danielballan) give this a shot

[Dan Allan (danielballan)]

That looks good. I took a quick look at the docs for request and read, in this section, "it is good practice to set your timeout to slightly larger than a multiple of 3 [seconds], which is the default TCP transmission window." Would 3.5 be a better default value than 2?

[Deleted user (ghost)]

Thomas A Caswell (@tacaswell) and I picked 2, we were off by a second and a half :). I think even 4.5 wouldn't hurt given olog also authenticates. I will test it and make the corresponding changes to the PR

[Deleted user (ghost)]

attn Dan Allan (@danielballan)

[Dan Allan (danielballan)]

Attn cmazzoli This should solve the issue with the Olog subscription hanging scans. If the Olog does not respond after ~4 seconds, an error will be raised in Python. The RunEngine will catch that error, print a warning, and continue with the scan.

[Dan Allan (danielballan)]

attn arkilic By coincidence, when I visited CSX to install the updated pyolog, the Olog happened to be down (what are the odds?!) and so I had the opportunity to test that, with this upgrade, bluesky is able to run the scan after a ~5 second delay, showing a warning about a ReadTImeoutError.

[Deleted user (ghost)]

Mission accomplished with a 27 character commit 👍