nbd-server "connection reset by peer" message is normal
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
logcheck (Ubuntu) |
Invalid
|
Undecided
|
Unassigned | ||
nbd (Ubuntu) |
Invalid
|
Undecided
|
Unassigned |
Bug Description
Binary package hint: logcheck-database
Following this thread on edubuntu-devel, a small configuration change will soon be made to /etc/hosts.allow so that SO_KEEPALIVE is enabled on nbd_server connections.
https:/
As I'm testing it now, the result seems to be that, when an nbd-client goes away without properly closing the connection, some time (>2hours by default) the connection gets closed by tcpd and nbd-server, perhaps taken a little by surprise emits:
May 9 13:34:43 medlycott nbd_server[28314]: Read failed: Connection reset by peer
It would appear this is now to become normal behaviour. As such, it seems likely logcheck should ignore it.
Is this symptom still reproducible in 8.10 beta?