search

mit-pdos / noria

Fast web applications through dynamic, partially-stateful dataflow
Apache License 2.0
4.99k stars 244 forks source link

Fix test failures #55

Closed ekmartin closed 6 years ago

jonhoo commented 6 years ago

I think you also need to bump the version of arccstr to 1.0.4 for the current nightly

ekmartin commented 6 years ago

Done!

ms705 commented 6 years ago

Looks good to me! :+1:

@fintelia, can you double-check that all the new unwraps added in the consensus code are in places where we actually want to crash on failure? I think so, but another pair of eyes would be good.

jonhoo commented 6 years ago

Also, this still doesn't seem to pass?

ekmartin commented 6 years ago

@jonhoo: it fails because of build errors in benchmarks. Didn't think it was worth fixing since it seemed like you were doing a pretty significant refactor in the orchestrator branch. Could add each subcrate manually with --manifest-path in the meantime though, at least then the tests would run.

jonhoo commented 6 years ago

Ah, yes, fair. I haven't done anything to the hotsoup binary, but you're right, that seems fine. Once @fintelia signs off, I'd be happy for this to be merged, and then I'll merge it into orchestrator (which tbh should probably become master soon).

ms705 commented 6 years ago

@fintelia Can you have a quick look over this?

Let's merge it before master diverges again.

fintelia commented 6 years ago

These changes look good to me. Lets merge and then just double check that all the tests do in fact work.

jonhoo commented 6 years ago

@ekmartin can you rebase on top of master? For Cargo.lock, just remove the file, run cargo update, and commit the resulting one.

ekmartin commented 6 years ago

@jonhoo: rebased 👍

jonhoo commented 6 years ago

:tada: @ms705 can you take a look at the hotsoup failures? Maybe we should just delete that binary from master?