Add test for proper handling of connection failure to avoid 'cannot wait on socket event without a socket' error

[knizhnik]

Problem

See https://github.com/neondatabase/cloud/issues/14289
and PR #8210

Summary of changes

Add test for problems fixed in #8210

Checklist before requesting a review

• I have performed a self-review of my code.
• If it is a core feature, I have added thorough tests.
• Do we need to implement analytics? if so did you add the relevant metrics to the dashboard?
• If this PR requires public announcement, mark it with /release-notes label and add several sentences in this section.

Checklist before merging

• Do not forget to reformat commit message to not include the above checklist

[MMeent]

This does a full PS connect timeout up to an error; can we reduce that time so that we don't have a test consuming a runner's testslot for ~50 seconds?

[kelvich]

I'll keep mentioning that we don't need https://github.com/neondatabase/neon/pull/8210/files#diff-d6b0f875ded76c09210876308df775e27adf4b424af1824c65a5d5871effe074L430-L434 now as PQstatus does the same check. Am I wrong about it or there is some reason to keep it?

[github-actions]

3006 tests run: 2891 passed, 0 failed, 115 skipped (full report)

---

Code coverage* (full report)

• functions: 32.7% (6932 of 21213 functions)
• lines: 50.0% (54306 of 108576 lines)

* collected from Rust tests only

---

_{The comment gets automatically updated with the latest test results
c63bc01 at 2024-07-02T17:10:58.198Z :recycle:}

[knizhnik]

I'll keep mentioning that we don't need https://github.com/neondatabase/neon/pull/8210/files#diff-d6b0f875ded76c09210876308df775e27adf4b424af1824c65a5d5871effe074L430-L434 now as PQstatus does the same check. Am I wrong about it or there is some reason to keep it?

Yes, this check seems to be redundant.
PQerrorMessage also correctly handle NULL pointer.

[knizhnik]

This does a full PS connect timeout up to an error; can we reduce that time so that we don't have a test consuming a runner's testslot for ~50 seconds?

I would rather consider it as question to implementation rather then test - if there is is non-recoverable error, should we retry connection to PS? From one hand it seems to have no sense... But from the other - it is our own PS and why there is any invalids connection string? And even if somebody somehow incorrectly specify such connection string, will it be better that all backlands fail immediately or will be blocked for 50 seconds trying to reconnect to PS, making possible for "somebody" to fix the problem and revert incorrect connection string setting?

[knizhnik]

This does a full PS connect timeout up to an error; can we reduce that time so that we don't have a test consuming a runner's testslot for ~50 seconds?

Adding statement_timeout fixes this problem.