[if this mail arrived twice, excuse me. i don't trust my mail server at the moment;-)] hello diald-list i have been hunting a rather ambiguous problem for weeks now that i cannot realy center in on; this is why i am posting this humunguous mail: S.u.S.E. Linux 6.1, Kernel 2.2.5 Diald 0.16.x (and 0.99.1) Fetchmail 4.7.5 (daemon mode) pppd 2.3.5 route 1.96 (from net_tools 1.50) the symptoms are: 1) when fetchmail tries to get mail from the provider, diald brings the link up, but fetchmail gets no data (the configuration file is ok). diald-top shows the TTL countdown, identifying the Destination Source 0.0.0.0/1044 [provider ip-addr]/pop3 (funny enough, if i kill fetchmail and bring it back up, while the link is up, it gets the mail ...) when the TTL is over, the packet gets renewed with the original TTL from the "accept tcp [TTL] any"-rule in diald.conf. At the same time netstat displays that somebody is still waiting for a SYN-reply (pop.vol.at is the provider mailserver): $ netstat tcp 0 1 *:1154 pop.vol.at:pop3 SYN_SENT tcp 0 0 localhost:1152 localhost:1153 ESTABLISHED [...] (the top entry remains there for ever, unless i shut down the fetchmail daemon.) the link then goes down occasionally after about 5 minutes instead of 100 sec, and bounces back up again after 30 sec (if i leave fetchmail running) 2) on other occasions diald just dies when the last TTL is up, leaving the link up for infinity 3) when shutting down a link, diald reports to syslog: [...] Jun 29 18:48:20 marble diald[1320]: Closing down idle link. Jun 29 18:48:20 marble diald[1320]: Nonzero exit status (7) on command '/sbin/route add 127.0.0.2 metric 1 dev sl0' Jun 29 18:48:20 marble diald[1320]: Nonzero exit status (7) on command '/sbin/route add default metric 1 dev sl0' Jun 29 18:48:20 marble pppd[1435]: Terminating on signal 2. Jun 29 18:48:20 marble pppd[1435]: sent [LCP TermReq id=0x2 "User request"] Jun 29 18:48:20 marble pppd[1435]: rcvd [LCP TermAck id=0x2] Jun 29 18:48:20 marble pppd[1435]: Connection terminated. Jun 29 18:48:21 marble pppd[1435]: Exit. Jun 29 18:48:21 marble diald[1320]: Nonzero exit status (7) on command '/sbin/route add 127.0.0.2 metric 1 dev sl0' Jun 29 18:48:21 marble diald[1320]: Nonzero exit status (7) on command '/sbin/route add default metric 1 dev sl0' Jun 29 18:48:22 marble diald[1320]: Delaying 30 seconds before clear to dial. [...] sometimes dying after the "Nonzero exit status ..."-message I have recompiled and installed diald (updating from 0.16 to 0.99.1), fetchmail and the net_tools package - without success i do not know anymore where to look for the problem - even though i guess it has to do with the network config. Does anybody have an idea? I would VERY MUCH appreciate any hint. Thnx & cheers markus - To unsubscribe from this list: send the line "unsubscribe linux-diald" in the body of a message to [EMAIL PROTECTED]
