Open egrigore opened 3 years ago
Excuse my late response, we have not seen this before. AFAICS the acquire check in remote/client.py
may be to strict, as it requires the resources of a place to not be acquired at all, not even permitting the current place. OTOH this should not happen at all and indicates a synchronization/data race between the coordinator and exporter infrastructure.
@egrigore Is this still an issue at your end with recent labgrid versions? We've never seen.
Hi @Bastian-Krause , yes , it still appears from time to time (we rebased to latest upstream in the summer) - only for the named-match resources (if this helps).
@egrigore Okay. Since you seem to be the only one able to reproduce this, investigation is probably up to you.
@Bastian-Krause , yes , I will try to investigate - probably I need to think of a redesign of the current configurations logic. Thank you
Hi,
From time to time I meet this issue: that I can't acquire a place because one of its resources is already acquired by the place itself. I don't know in what situation it happens. Has anyone met the same issue?
Below is the error message I get and how the place resources look like (Matching resource '2' is acquired by the place):