Please include errno when raising LockError

Bug #1071766 reported by nutjob
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
zc.lockfile
New
Undecided
Unassigned

Bug Description

I've been getting intermittent startup errors on a Zope instance from zc.lockfile 1.0.0, zc/lockfile/__init__.py line 57, in the call to fcntl. The message is displayed as "LockError: Couldn't lock FILENAME".

Unfortunately, there's no further information, as the try block catches the IOError and raises a fresh LockError.

That IOError does have a useful bit of data, though: the errno value left by the failed call to fcntl. With that, I'd have a clue as to what the underlying issue is.

Perhaps the errno could be included in the raised LockError's message, as suggested by the patch.

Thanks,
Sean "nutjob" Kelly.

Revision history for this message
nutjob (kelly-seankelly) wrote :
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.