CI: Troubleshoot

[jbrockmendel]

• closes #xxxx
• tests added / passed
• Ensure all linting tests pass, see here for how to run them
• whatsnew entry

[mroeschke]

My theory proved incorrect in #45500. Looks like you're having more luck here?

[jbrockmendel]

looks like azure is green. time to whittle these down...

[jreback]

great

[jreback]

any luck here?

[jreback]

can we force the windows builds to a single cpu? (and maybe disable some expensive tests / keep 2 builds)

[jbrockmendel]

any luck here?

slowly narrowing it down

[jbrockmendel]

OK! looks like its somewhere in usecols...

[jbrockmendel]

The build that just failed is almost identical to one that passed a few hours ago (see the "remove usecols" commit above 71bf824). That commit also removed the entire usecols directory, while this one left usecols/__init__.py

[mroeschke]

can we force the windows builds to a single cpu? (and maybe disable some expensive tests / keep 2 builds)

Trying this approach in #45547. The CI is really backed up right now so it might take a while to get to

[jbrockmendel]

The CI is really backed up right now so it might take a while to get to

Could/should we just cancel some of the non-fixing-this azure jobs to speed this up?

[mroeschke]

The CI is really backed up right now so it might take a while to get to

Could/should we just cancel some of the non-fixing-this azure jobs to speed this up?

Yeah if there are lot of queued up jobs it would allow to debug this quicker. But looks like the queue cleared up overnight

[jbrockmendel]

azure green!

[mroeschke]

azure green!

Confident it was this or do you think retriggering the CI will confirm this?

[jreback]

ok let's merge on green (i retriggered)

[jreback]

let's also an issue to troubleshoot this as well (this case you are skipping)

[jreback]

@meeseeksdev backport 1.4.x

[lumberbot-app]

Something went wrong ... Please have a look at my logs.