Earlier today I changed the IPv6 address used for the LUV server. I didn't keep the old address working because almost no-one uses IPv6 and IPv6 clients can generally fall back to IPv4 if necessary, so the time that the old address remains in DNS caches shouldn't be a problem. Hetzner had decided to remove the address range they had assigned to that server and I had to apply for a new /64. On Friday afternoon the hardware hosting the LUV VM crashed for unknown reasons. A couple of hours later it booted again of it's own accord and there was nothing in the logs. So that I can fix such things faster in future I have setup 2 monitoring systems with Jabber that monitor each other. Now I just need to monitor the Jabber client on my phone (see my previous message). Some time recently an update to Dovecot had caused it to stop working correctly. I didn't notice as I don't have an IMAP account on the LUV server. To prevent that sort of thing happening again I have written a monitoring script for IMAP that will alert if there is no new mail for more than 10 minutes in a test account. I believe that some mail to the LUV president was delayed because of this. But it won't happen again. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/