I struck this too. I thought "Cannot read filterfile" meant the file didn't exist or something, and by the time I figured out it was the syntax (and put an empty file in its place) I was offline, so the postinst fetchnews then failed for the servers being unreachable.
It'd be nice if that filterfile message was clearer, and better still if there was some sort of hint that you have to go in and update it manually. I think it'd also be nice if that postinst fetchnews wasn't a fatal error. I guess you lose access to your spool until it runs successfully, but the package may as well install and then let you fixup the configs later (assuming that's a manual process). Also, somehow or somewhere in my attempts at texpire / fetchnews / rnews, and seeing if the nntp was working, I managed to fill /var/log/syslog with 50 Mbytes of messages about groupinfo file lines being invalid (old format I guess). Not sure if this was my own fault, but it could be worth thinking about removing or massaging the groupinfo file in the upgrade process. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
