Extending snapshots in dmeventd will be soon implemented. We must handle the situation when the snapshot overflows before dmeventd does it's job more gracefully; losing all data (what it does now) is really the worst possibility for some workloads. Possible ideas: * block writes to the origin and snapshot until dmeventd does its job (disadvantage --- dmeventd may wait for some memory allocation that results in write to the blocked origin) * drop the snapshot (current behavior; prevents system crash and origin problems but destroys all data on the snapshot) * return -EIO on all writes to the origin and snapshot --- doesn't cause system crash, causes problems on the origin (if the origin is root fs, it will cause a crash). On the other hand, it won't cause data corruption neither on the origin nor on the snapshot, after reboot data is intact. * throttle writes to the snapshot --- hardest to implement but I think it is the best workaround for the deadlock.
This bug appears to have been reported against 'rawhide' during the Fedora 19 development cycle. Changing version to '19'. (As we did not run this process for some time, it could affect also pre-Fedora 19 development cycle bugs. We are very sorry. It will help us with cleanup during Fedora 19 End Of Life. Thank you.) More information and reason for this action is here: https://fedoraproject.org/wiki/BugZappers/HouseKeeping/Fedora19