-
Notifications
You must be signed in to change notification settings - Fork 91
Open
Description
Hello,
We have noticed bad behavior for several months because of this commit:
xcp-ng@d28dcc1
In practice the commit that we revert is useful in the case where ISCSI connections are no longer valid, but sometimes even without any config change or bad connections, a disconnect/connect is launched preventing the correct use of the SR.
This means that it can interrupt backup tasks, for example.
Wouldn't this piece of code be a little too violent to restart the ISCSI service while the SR is still functional?
Wouldn't the code be a little too sensitive to a particular trigger, when there is no reason to restart the service?
Thanks!
olivierlambertbenjamreis
Metadata
Metadata
Assignees
Labels
No labels