ctld restarted, not reloaded when creating extent or associating extent to target
obviously service ctld restart is issued, what causes connection error.
as i read sending SIGHUP, what service ctld reload does, would be sufficient.
tested with latest 9.x and proxmox on an iscsi-target.
all vm's running on the target stop when manually adding an extent or associating an extent to the target in the gui,
can be reproduced quite reliable.
tested the same with
service ctld reload -> good
service ctld restart -> same behaviour, vm's stop, connection error
The same happens when creating extents / assigning to targets via REST API (where i initially noticed the vm's stopping)
#3 Updated by Alexander Motin almost 3 years ago
- Status changed from Screened to Closed: Cannot reproduce
- Target version changed from 11.1 to N/A
It seems this problem was already fixed at some point. I looked into the code and tested recent nightly build and as I see ctld is always reloaed, not restarted. May be it is not good either, since not all properties can be changed that way now, but that is what it is right now.