Hello all,
we noticed a lot of LockError exceptions in Sentry and are suspicious, whether it's troublesome or not. We have many incidents with a pause of 1-2 seconds and then it suddenly stops. Per day it may happen a hundred times, but then few days/weeks nothing. To illustrate this, I have added the diagram of events over time:
The site itself runs seemingly smooth as far we can tell.
The traceback:
LockError
Couldn't lock '/var/zeoclient2-2.zec.lock'zc/lockfile/init.py in _lock_file at line 59
zc/lockfile/init.py in init at line 84
ZEO/cache.py in init at line 199
ZEO/ClientStorage.py in init at line 404
ZODB/config.py in open at line 220
ZODB/config.py in open at line 101
Zope2/Startup/datatypes.py in createDB at line 182
Zope2/Startup/datatypes.py in open at line 185
Zope2/Startup/datatypes.py in getDatabase at line 287
eggs/Zope2-2.13.28-py2.7.egg/Products/ZODBMountPoint/MountedObject.py in _getDB at line 152
eggs/Zope2-2.13.28-py2.7.egg/Products/ZODBMountPoint/MountedObject.py in _getMountedConnection at line 142
eggs/Zope2-2.13.28-py2.7.egg/Products/ZODBMountPoint/MountedObject.py in _getOrOpenObject at line 247
It seems that there is a problem with locking the persistent cache. The files' owner is the very same as the one running the instance, so I don't think it's a permission problem.
Do you have any other ideas where to investigate?
best, Paul