Rerender the feedstock to fix downstream issues

[xhochy]

As discussed in conda-forge/openmpi-feedstock#142

[conda-forge-webservices]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe) and found it was in an excellent condition.

[xhochy]

@conda-forge-admin please restart ci

[pentschev]

Thanks @xhochy , if I understand correctly conda-forge/openmpi-feedstock#142 (comment) is the specific reason for this. Should we mark the previous build as broken too?

[xhochy]

Yes, but only once we have confirmed that this build also works.

[dalcinl]

Merge Forrest, merge!

[pentschev]

LGTM.

[jakirkham]

Should we mark the previous build as broken too?

Was it just one build that was broken?

@mbargull is there a good way for us to check which packages were affected?

[mbargull]

Was it just one build that was broken?

See conda-forge/openmpi-feedstock#142 (comment)

@mbargull is there a good way for us to check which packages were affected?

I don't know/don't think so.

[jakirkham]

Ok let's start with marking those broken

If we missed something, the errors are pretty loud. So hopefully we catch them quickly

Submitted PR: conda-forge/admin-requests#956

[h-vetinari]

I don't think it's a good idea to mark these builds as broken. They're over a month old, and they only break if someone calls ldconfig (AFAIU). On top of that, while there is a functional replacement for rdma_core 50.0*=_0, there's no replacement for rdma_core 49.1, and so we're very likely to break otherwise working environments with this.

[jakirkham]

We could rebuild rdma-core 49.1 too

[h-vetinari]

Of course. I just didn't see the urgent need to mark them broken in the first place. That should be a measure of last resort, and aside from the build issues in openmpi, there didn't seem to be a problem AFAICT?

[jakirkham]

Submitting a rebuild with PR: #18

[pentschev]

Both #18 and conda-forge/admin-requests#956 are now merged. With that I believe the issue on rdma-core front is resolved. Thanks all!