rsc: Implement several bug fixes #1543
Conversation
| // TODO: The logging is incredibly spammy right now and causes significant slow down. | ||
| // for now, logging is disabled but this should be turned back on once logging is pruned. |
There was a problem hiding this comment.
We should probably just make this an option right?
There was a problem hiding this comment.
Yeah, though I want to spend some time to get it right. i.e. a matrix of where to log X what level to log at. I feel like it'll be totally fine for "log errors to file" as the default. I just wanted to focus on making it work first before circling back around to that.
| @@ -162,66 +162,79 @@ export def mkRemoteCacheRunner (rscApi: RemoteCacheApi) (hashFn: Result RunnerIn | |||
| else failWithError "rsc: Failed to make output directory" | |||
|
|
|||
| def doDownload (CacheSearchOutputFile path mode blob) = | |||
| rscApiGetFileBlob blob "{input.getRunnerInputDirectory}/{path}" mode | |||
| rscApiGetFileBlob blob path mode | |||
There was a problem hiding this comment.
We might want to write down in the RunnerInput/Plan tuple documentation that only Command is relative to Directory both Visible/Input and Output are relative to workspace. It seems reasonable to assume either and I don't see it written down.
There was a problem hiding this comment.
I made a comment/note on this line but it would probably be good for us to put together a high level document covering all these details/designs/limitations.
We have a bunch of smaller docs, but I don't think we have the top level doc covering all the content. Mainly because most of this was just iterating on the local cache learnings/design
The following bugs are fixed in this PR
With these changes larger internal builds will complete successfully