Hi, DM et. al,

I have a question about host-routed netmail to points. I've been banging my head on this for some time and am running into issues.

I'm an RC in fidonet and route netmail for my region. netmail from downlinks passes through OK to destination nodes, but if a downlink sends mail for a point, it appears to sit on my system and is not routed to the bossnode. I found a couple errors in my log where SBBS was trying to send to the point and failing.

How can I set sbbsecho to crashmail the boss node when point mail arrives?

--k

---
þ Synchronet þ .: realitycheckbbs.org :: scientia potentia est :.

Re: Host-routed netmail to points
By: poindexter FORTRAN to All on Wed May 25 2022 09:35 am

Subject: Host-routed netmail to points
@MSGID: <628E5AD1.64191.dove.sync@realitycheckbbs.org>
@TZ: c1e0

Hi, DM et. al,

I have a question about host-routed netmail to points. I've been banging my head on this for some time and am running into issues.

I'm an RC in fidonet and route netmail for my region. netmail from downlinks passes through OK to destination nodes, but if a downlink sends mail for a point, it appears to sit on my system and is not routed to the bossnode. I found a couple errors in my log where SBBS was trying to send to the point and failing.

How can I set sbbsecho to crashmail the boss node when point mail arrives?

Here's that routing logic in sbbsecho.c:

if(nodecfg == NULL && dest.point != 0) {
fidoaddr_t boss = dest;
boss.point = 0;
if((nodecfg = findnodecfg(&cfg, boss, /* exact: */true)) != NULL) {
dest = boss;
lprintf(LOG_INFO, "Routing packet (%s) to boss-node %s", tmp_pkt, smb_faddrtoa(&dest, NULL));
}
}


So it looks like the destination point node must not be configured in sbbsecho.ini while the boss node must be configured in sbbsecho.ini for this routing to occur.
--
digital man (rob)

Synchronet/BBS Terminology Definition #70:
SIGHUP = Hangup signal sent to a process when its controlling terminal is closed
Norco, CA WX: 73.1øF, 62.0% humidity, 5 mph ESE wind, 0.00 inches rain/24hrs ---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

Re: Host-routed netmail to points
By: Digital Man to poindexter FORTRAN on Wed May 25 2022 10:12 am

So it looks like the destination point node must not be configured in sbbsecho.ini while the boss node must be configured in sbbsecho.ini for thi

Would setting a rule for 1:ALL setting the status to CRASH meet that criteria? What I'd like to do is go ahead and crash anyone directly, and as expected, crashmail point netmail to the bossnode.
--- SBBSecho 3.15-Win32
* Origin: http://realitycheckbbs.org | tomorrow's retro tech (1:218/700)
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

Re: Host-routed netmail to points
By: Kurt Weiske to Digital Man on Wed May 25 2022 01:39 pm

Re: Host-routed netmail to points
By: Digital Man to poindexter FORTRAN on Wed May 25 2022 10:12 am

So it looks like the destination point node must not be configured in sbbsecho.ini while the boss node must be configured in sbbsecho.ini for thi

Would setting a rule for 1:ALL setting the status to CRASH meet that criteria?

I don't think so: "the boss node must be configured in sbbsecho.ini".

What I'd like to do is go ahead and crash anyone directly, and as
expected, crashmail point netmail to the bossnode.

The logic in SBBSecho is currently that the bossnode must be configured. Perhaps that could be changed to just strip the point number always and call that a "route", but nobody has ever asked for that behavior. Is that what you're asking for?
--
digital man (rob)

This Is Spinal Tap quote #6:
David St. Hubbins: He was the patron saint of quality footwear.
Norco, CA WX: 79.0øF, 52.0% humidity, 10 mph S wind, 0.00 inches rain/24hrs
---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

Re: Host-routed netmail to points
By: Digital Man to Kurt Weiske on Wed May 25 2022 02:16 pm

The logic in SBBSecho is currently that the bossnode must be configured. Perhaps that could be changed to just strip the point number always and

calI

I have 1:229/426 configured in sbbsecho and tried sending mail to a point off of Nick's system. I saw this in my log:

5/25 10:45:07p BINKOUT Connecting to 1:229/426.3@fidonet at p3.f426.n229.z1.binkd.net:24554
5/25 10:45:07p BINKOUT Connection to p3.f426.n229.z1.binkd.net:24554 failed (Error: No such host is known. ).

So, it looks like sbbsecho is treating a 4d address literally and trying to send directly to a point.

What I think it should be doing is evaluating the zone:net/node portion of pointmail and sending the mail according to the boss node's configuration or a wildcard configuration that applies.

---
þ Synchronet þ .: realitycheckbbs.org :: scientia potentia est :.

Hi poindexter,

On 2022-05-25 23:01:17, you wrote to Digital Man:

The logic in SBBSecho is currently that the bossnode must be
configured. Perhaps that could be changed to just strip the point
number always and

calI

I have 1:229/426 configured in sbbsecho and tried sending mail to a point off of Nick's system. I saw this in my log:

5/25 10:45:07p BINKOUT Connecting to 1:229/426.3@fidonet at p3.f426.n229.z1.binkd.net:24554 5/25 10:45:07p BINKOUT Connection to p3.f426.n229.z1.binkd.net:24554 failed (Error: No such host is known. ).

So, it looks like sbbsecho is treating a 4d address literally and trying to
send directly to a point.

Sometimes that works! For instance I have 2 points which are directly connectable because they have a binkp service running 24/7. And at what address they are connectable is also public information because it's published in the Z2PNT list.

So if DM is going to change anything in the routing logic, I think that possible situation should also be taken into account...


Bye, Wilfred.

--- FMail-lnx64 2.1.0.18-B20170815
* Origin: FMail development HQ (2:280/464)
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

Re: Host-routed netmail to points
By: poindexter FORTRAN to Digital Man on Wed May 25 2022 11:01 pm

Subject: Host-routed netmail to points
@MSGID: <628F17AD.64200.dove.sync@realitycheckbbs.org>
@REPLY: <628E9CC1.47335.sync@vert.synchro.net>
@TZ: c1e0
Re: Host-routed netmail to points
By: Digital Man to Kurt Weiske on Wed May 25 2022 02:16 pm

The logic in SBBSecho is currently that the bossnode must be configured. Perhaps that could be changed to just strip the point number always and

calI

I have 1:229/426 configured in sbbsecho and tried sending mail to a point off of Nick's system. I saw this in my log:

5/25 10:45:07p BINKOUT Connecting to 1:229/426.3@fidonet at p3.f426.n229.z1.binkd.net:24554
5/25 10:45:07p BINKOUT Connection to p3.f426.n229.z1.binkd.net:24554 failed (Error: No such host is known. ).

So, it looks like sbbsecho is treating a 4d address literally and trying to send directly to a point.

What I think it should be doing is evaluating the zone:net/node portion of pointmail and sending the mail according to the boss node's configuration or a wildcard configuration that applies.

According to the logic I pasted here and explained:

If 1:229/426 is configured in sbbsecho.ini and 1:229/426.3 is *not* configured in sbbsecho.ini, then you should see in your sbbsecho.log file a line similar to this one, when processing a packet for 1:229/426.3:

Routing packet (x) to boss-node 1:229/426
--
digital man (rob)

Sling Blade quote #16:
Karl Childers (to Doyle, re: lawn mower blade): I aim to kill you with it. Mmm. Norco, CA WX: 60.7øF, 92.0% humidity, 2 mph SE wind, 0.00 inches rain/24hrs
---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

Wilfred van Velzen wrote to poindexter FORTRAN <=-

Sometimes that works! For instance I have 2 points which are directly connectable because they have a binkp service running 24/7. And at what address they are connectable is also public information because it's published in the Z2PNT list.

I've never been sure of the reasoning behind a public point list; my
understanding is that the boss node just needs to know about its
points, and all routing can be done to the boss - and the idea behind
a point was to support a system that was inaccessible at other times
and couldn't keep up with the minimum of being CM or available during
ZMH?

Otherwise, why not just put the node in the nodelist?

So, the expected behavior should be, if a point configuration is
defined in sbbsecho and the point address is known, route directly.

If the boss has a configuration and the point doesn't, route to the
boss.

If there's no boss or point configuration, route to the boss according to
default rules.

What appears to be happening is that sbbsecho is trying to route
directly to the point by default.






... Discover your formulas and abandon them
--- MultiMail/DOS v0.52
* Origin: http://realitycheckbbs.org | tomorrow's retro tech (1:218/700)
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

Re: Host-routed netmail to points
By: Digital Man to poindexter FORTRAN on Thu May 26 2022 08:45 am

If 1:229/426 is configured in sbbsecho.ini and 1:229/426.3 is *not* configured in sbbsecho.ini, then you should see in your sbbsecho.log file a line similar to this one, when processing a packet for 1:229/426.3:

Routing packet (x) to boss-node 1:229/426

Therein lies the problem. 229/426 is defined -- Nick is my uplink. .3 is not defined. What I'm seeing is below.

5/25 10:45:07p BINKOUT Connecting to 1:229/426.3@fidonet at
p3.f426.n229.z1.binkd.net:24554
5/25 10:45:07p BINKOUT Connection to p3.f426.n229.z1.binkd.net:24554
failed (Error: No such host is known. ).

---
þ Synchronet þ .: realitycheckbbs.org :: scientia potentia est :.

Re: Host-routed netmail to points
By: poindexter FORTRAN to Digital Man on Thu May 26 2022 01:50 pm

Subject: Host-routed netmail to points
@MSGID: <628FE7FE.64204.dove.sync@realitycheckbbs.org>
@REPLY: <628FA08D.47339.sync@vert.synchro.net>
@TZ: c1e0
Re: Host-routed netmail to points
By: Digital Man to poindexter FORTRAN on Thu May 26 2022 08:45 am

If 1:229/426 is configured in sbbsecho.ini and 1:229/426.3 is *not* configured in sbbsecho.ini, then you should see in your sbbsecho.log file a line similar to this one, when processing a packet for 1:229/426.3:

Routing packet (x) to boss-node 1:229/426

Therein lies the problem. 229/426 is defined -- Nick is my uplink. .3 is not defined. What I'm seeing is below.

5/25 10:45:07p BINKOUT Connecting to 1:229/426.3@fidonet at
p3.f426.n229.z1.binkd.net:24554
5/25 10:45:07p BINKOUT Connection to p3.f426.n229.z1.binkd.net:24554
failed (Error: No such host is known. ).

The sbbsecho.log file is the relevant log to be looking at in this case.
--
digital man (rob)

Breaking Bad quote #25:
Now if I could only learn how to lick myself. - Hank Schrader
Norco, CA WX: 74.9øF, 57.0% humidity, 11 mph SW wind, 0.00 inches rain/24hrs ---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

Digital Man wrote to poindexter FORTRAN <=-

The sbbsecho.log file is the relevant log to be looking at in this
case. --

I looked in the log and saw some weird entries showing 5 digit zone numbers. On a lark, I tried changing "Fuzzy Zone Operation" in echocfg to "no", and
now I'm seeing the routing messages in my sbbsecho.log.

If I can reproduce the error again, I'll post the log files. For now, it
seems like it's behaving as expected.

Thanks for your help!


... Use an unacceptable colour
--- MultiMail/DOS v0.52
* Origin: http://realitycheckbbs.org | tomorrow's retro tech (1:218/700)
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net