Support multiple evaluators on a single machine #673
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This has been deployed for the past week or so and nothing bad happened (...after I worked out all the initial kinks -- if you saw out-of-space errors followed by me prodding ofborg to re-eval... no you didn't).
We now run 3 evaluators on each of the 5 machines we have access to through EM. I saw about 160GiB peak when all 3 were running the outpaths evaluation at the same time, out of a total of 256GiB of RAM. This (hopefully) leaves room for the single builder that resides on the same machine as well as for the outpaths check to consume even more ( 🙄 ) memory.
This also seems to more than handle the (current) Nixpkgs PR cadence, as all the machines chewed through the ~160 queue in about 10 hours:
We also run
git gcevery time we clone / before we clone so as to help curb the inode usage of these perpetual Nixpkgs checkouts. Before this, I had seen checkouts of more than 70GiB on disk!!! It's not fast (I think it takes 10 minutes tops), but it certainly helps cut down on inode clutter.It's not beautiful, but it works.