Quick fix for a failing harvesting test #8937

[landreev]

What this PR does / why we need it:

Which issue(s) this PR closes:

Closes #8937

Special notes for your reviewer:

Suggestions on how to test this:

Does this PR introduce a user interface change? If mockups are available, please link/include them here:

Is there a release notes update needed for this change?:

Additional documentation:

[coveralls]

Coverage remained the same at 20.045% when pulling 781ca74 on 8937-harvest-test-failure into 1ef73b0 on develop.

[pdurbin]

Looks like a doubling, as discussed. I'm waiting for the Jenkins tests to pass before I merge this.

[landreev]

Before increasing that sleep, I ran the test a few times and it was failing about 50% of the time. After doubling it I ran it a few more times and wasn't able to get it to fail.
So, looks like a solid fix. 😄

[landreev]

I'll take a look to see why it bombed once I'm in. ☹️

[landreev]

This is kind of insane that the harvesting test is still bombing. So maybe it's not the duration after all. (not sure yet what's up with the other test that failed)

[landreev]

Oh. Was that the wrong sleep interval to increase? - I.e., it may be failing not on account of the set create taking too long, but because we try to create the set before the publish command has completed on the dataset there...

[landreev]

OK, the tests have passed.
The other test that failed during the first run - it died in a really random place, with a 500 from an attempt to destroy a dataset at the end of a test. I feel like dismissing this one as a fluke.

[pdurbin]

Before merging I confirmed that indeed all API tests passed. Again, all we're doing is bumping up the amount of sleep.

[landreev]

Yeah, I edited the comment in the issue, documenting that it wasn't the doubling of the 5 sec. sleep, but an extra sleep added before that step.