Don't consider unschedulable pods unrecoverable

[sseago]

Thank you for contributing to Velero!

Please add a summary of your change

Don't consider "unschedulable" pods unrecoverable (for fast fail purposes), since pods will be briefly unschedulable while waiting for PV provisioning.

Does your change fix a particular issue?

Fixes #7898

Please indicate you've done the following:

• [x ] Accepted the DCO. Commits without the DCO will delay acceptance.
• [x ] Created a changelog file or added /kind changelog-not-required as a comment on this pull request.
• [x ] Updated the corresponding documentation in site/content/docs/main.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 58.79%. Comparing base (89229d3) to head (9614ead).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #7899      +/-   ##
==========================================
- Coverage   58.80%   58.79%   -0.01%     
==========================================
  Files         345      345              
  Lines       28771    28766       -5     
==========================================
- Hits        16919    16914       -5     
  Misses      10423    10423              
  Partials     1429     1429              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[Lyndon-Li]

@sseago Please help to cherry-pick this PR to 1.14. Thanks.

[itayrin]

Hi, I just want to say that lately I was trying to integrate Velero 1.14.0 with our product and finding this specific commit made me downgrade to 13.2, since it's still not on any official version. This is an important fix, took me a few days to find the issue.

[sseago]

@itayrin Yes, unfortunately we didn't discover this bug until 1.14.0 was out. Some storage providers provision volumes quickly enough that we don't hit it, so our pre-release testing didn't show this -- but other providers take a bit longer for volumes to come up, and for those providers this bug was hit all the time. 1.14.1 includes the fix, and we're hoping to release that around the end of the month.