Alan King

Linking to https://github.com/irods/irods/issues/7521 as well. Something funny about this test suite.

I have not been able to reproduce this failure in the particular way described here (SYS_SOCK_READ_TIMEDOUT). I've seen other failures as described in #7389 and #7521, but not like this...

Possibly. I would need to try these things out myself to be sure. My default answer is that whatever wins the vote in hierarchy resolution for open-for-write is going to...

This is a caught exception that is logged for archaeological purposes, but does demonstrate a possible TOCTOU situation in replication as it relates to the catalog. I would expect that...

Oh, forgot to note that the test passes consistently when run at the bench in a clean environment. Not sure if there's much to do here, but wanted to capture...

I've run this test a few times in the testing environment alongside others and haven't seen it fail lately. I'm thinking that I may close this alongside #7521 as it...

I haven't observed this in some time, so I'm going to close this as not planned. We can open a new issue if it reappears.

The test suite passes when this file and its references are removed. We are going to wait on merging for now until we can confirm that this does not produce...

I understand the desire to instrument the dynamic PEPs from within rules. This is actually a concern I've had: configured policy for API PEPs can be circumvented by calling equivalent...

The weird part is that `test_control_plane` itself passes.