Alan King
Alan King
Apparently not. This is strange not only because the test is not failing, but because the change in 702d062703e234255f0619d761f23928a852f29c ALSO does not appear to clean up the /dev/shm files for...
I think that improving the error message is the only actionable request in this issue. Logical locking was introduced to prevent precisely the situation presented by point 2 in OP...
While no data corruption occurs now, the "database race" / TOCTOU problem is still present for this situation. ```bash $ iput -f irodsctl foo & iput -f irodsctl foo [1]...
We federated 2 zones running 4.2.11 with SSL enabled in the testing environment and found that I had issues when `irods_ssl_verify_server` is set to `cert`. I get an `SSL_HANDSHAKE_ERROR` on...
Similar and possibly related to #4984 and https://github.com/irods/irods/issues/6537.
Yes, I think we will need to modify `iadmin` as well as the GeneralAdmin API such that they are separated. It is possible that the database plugin will need to...
@FifthPotato - Please close if complete. Thanks!
opening a dataobject for write will not write to an existing copy if it is on a non-default resource
This seems at least tangentially related to #6600, although they are definitely distinct issues. Just leaving that here for later... This does seem strange mainly in that it is not...
opening a dataobject for write will not write to an existing copy if it is on a non-default resource
Okay, thanks. We'll take a look at this to see what's going on
opening a dataobject for write will not write to an existing copy if it is on a non-default resource
Is this possibly the same issue as #7475?