Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[🐛] {PROD4POD-1779} Try to fix lockfile-hashing timeout #1077

Merged
merged 4 commits into from
Jul 21, 2022

Conversation

JJ
Copy link
Contributor

@JJ JJ commented Jul 20, 2022

Initial idea

The timeout has happened in #1076, so it's likely it's not triggered here. Anway, it's a bit over the brink, so maybe an upgrade of the cache code will fix it; among other things, it changes node from version 12 to version 16 (although, for some reason, that part is written in Csharp?).
Anyway, if this does not work, there are no easy alternatives. This PR is waiting to be vetted, so I'm afraid we will have to disconnect the cache (it's very likely it's not making any good in this specific runner, too)

Resolution

Eliminate cache. Best case scenario, we were obtaining marginal gains; installing modules took anywhere from 2 minutes to 6 minutes; but downloading and uncompressing the cache never took less than 4 minutes. So even in the best case scenario we were losing around two minutes; I have seen uncompression times of up to 11 minutes, so it's really a very bad bet and needs to be eliminated. The fact that the hashing was also timing out was only the last nail in the coffin.

I have also upgraded the action that sets up node, and used a version for node, 18, which is different from the "main" one (still 16). Can give us a taste of some possible failures, before we upgrade; besides, that way we have at least two versions of node tested.

Juan Julián (JJ) Merelo added 4 commits July 20, 2022 09:31
And add self path to trigger workflow
Since performance of the MacOS runner is dismal, it takes an inordinate amount of time to unpack the cache and to compute the hashes. So I think overall we're better off eliminating it. Installing the modules is I/O bound, and it's not that bad in that area, so we might actually see our workflows run *faster* without the cache
We're no longer testing Android here
Well, we can use it to see if there's any hiccup with the latest version
Copy link
Contributor

@RichardSto RichardSto left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Makes sense to me!

@JJ JJ merged commit 66062fb into main Jul 21, 2022
@JJ JJ deleted the 🐛-PROD4POD-1779-fix-timeouts branch July 21, 2022 07:26
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Development

Successfully merging this pull request may close these issues.

2 participants