sbbsecho routes echomail for non-existant nodes to whatever blah:ALL catches it

well, that was fun and interesting... just found out that sbbsecho can and does route echomail... had a system that was linked here and then dropped out... i removed their entry from echocfg and thought that was all i needed to do... well, it wasn't... the node's number was still in the areas.bbs file so sbbsecho was still packing echomail for it... the thing was, since there was no node entry for that system any more, sbbsecho dutifully (re)routed the pkts to the system behind the 1:ALL system...

perhaps echocfg could/should be checking the lines in the areas.bbs file and removing entries from the lines for nodes that no longer exist in sbbsecho.ini? or sbbsecho would see that there's no node entry in sbbsecho.ini (instead of blindly(?) following areas.bbs) and not pack echomail for that node as well as removing it from the areas.bbs file... netmail processing would be ok, i think... the main thing is to not pack echomail for a non-existent node and (re)route it to some other system...

2021-05-06 15:34:42 SBBSecho 3.11-Linux r3.173 May 23 2020 GCC 7.5.0 (PID 25866) invoked with options: -ce
2021-05-06 15:34:42 Configured: 7 archivers, 79 linked-nodes, 11 echolists
2021-05-06 15:34:42 NetMail directory: /sbbs/ftn/netmail/
2021-05-06 15:34:42 Secure Inbound directory: /sbbs/ftn/in/secure/
2021-05-06 15:34:42 Non-secure Inbound directory: /sbbs/ftn/in/unsecure/
2021-05-06 15:34:42 Outbound (BSO root) directory: /sbbs/ftn/out/fidonet
2021-05-06 15:34:42 Read 251 areas from /sbbs/data/areas.bbs
2021-05-06 15:34:42 Read 5 areas from /sbbs/data/badareas.lst
2021-05-06 15:34:42 Read 258 echo statistics from /sbbs/data/echostats.ini
[...]
2021-05-06 15:34:42 Creating outbound packet from 1:3634/12 to 1:14/6: /sbbs/temp/sbbsecho/609444e1.pkt
2021-05-06 15:34:42 Adding FIDOTEST message from Wilfred van Velzen (2:280/464) to packet for 1:14/6: /sbbs/temp/sbbsecho/609444e1.pkt
2021-05-06 15:34:42 Appending outbound packet from 1:3634/12 to 1:14/6: /sbbs/temp/sbbsecho/609444e1.pkt
2021-05-06 15:34:42 Adding ALLFIX_FILE message from Htick (2:230/150) to packet for 1:14/6: /sbbs/temp/sbbsecho/609444e1.pkt
2021-05-06 15:34:42 Appending outbound packet from 1:3634/12 to 1:14/6: /sbbs/temp/sbbsecho/609444e1.pkt
2021-05-06 15:34:42 Adding RBERRYPI message from druck (3:770/3) to packet for 1:14/6: /sbbs/temp/sbbsecho/609444e1.pkt
2021-05-06 15:34:43 Finalizing outbound packet from 1:3634/12 to 1:14/6: /sbbs/temp/sbbsecho/609444e1.pkt
2021-05-06 15:34:43 Node (1:14/6) successfully locked via: /sbbs/ftn/out/fidonet/000e0006.bsy
2021-05-06 15:34:43 Sending packet (/sbbs/temp/sbbsecho/609444e1.pkt, 10.3KB) from 1:3634/12 to 1:14/6
2021-05-06 15:34:43 Routing packet (/sbbs/temp/sbbsecho/609444e1.pkt) to 1:229/426
2021-05-06 15:34:43 Moving packet for 1:229/426: /sbbs/temp/sbbsecho/609444e1.pkt to /sbbs/ftn/out/fidonet/609444e1.pkt
2021-05-06 15:34:43 Node (1:229/426) successfully locked via: /sbbs/ftn/out/fidonet/00e501aa.bsy
2021-05-06 15:34:43 File (/sbbs/ftn/out/fidonet/609444e1.pkt, 10.3KB) for 1:229/426 added to BSO/FLO file: /sbbs/ftn/out/fidonet/00e501aa.dlo

changed the description

Before April 19 2016, any linked nodes (addresses linked to areas in the area file) that were not configured as nodes in sbbsecho.ini, were ignored. This was changed ("Don't require an exact match [node:x] section in sbbsecho.ini for every link in the area file (areas.bbs)"): http://cvs.synchro.net/cgi-bin/viewcvs.cgi/src/sbbs3/sbbsecho.c?revision=3.6&view=markup and I'm not sure why.

Presumably, someone complained or asked for change in behavior. So I'll make this behavior (the requirement that linked nodes are specifically configured as nodes in sbbsecho.ini) optional.

closed