Something I havn't really thought about until recently....

What do you do if your feed goes down?

I noticed this morning that my FidoNet hub is no longer responding via my polls. Now I know this isn't FidoNet but it got me thinking about the best way to stop this from happenning and provide some redundancy.

Some people might argue that this is not even needed, but it seems crazy to me that this is not built into software that people use for "hubs".

How will I know if my single FidoNet hub run by an elderly gentleman is ever coming back?

All it would take is a syncronisation of the credentials for feeds/points and a log of their suscribed echos/areas. (This could even be done via a FTN echo if needed)

This could then be duplicated between however many systems for redundancy.

"Clients" would then poll the DNS CNAME which includes all the hubs. As long as one hub is still active (perhaps after a few retries) mail/files would still flow.

BinkP caters for this also I believe using optional multiple hosts on the "node" line.

I'm now wondering if I have to find another FidoNet feed from somewhere....

--- ENiGMA 1/2 v0.0.12-beta (linux; x64; 12.18.3)
* Origin: Sonic BBS (21:4/144)

On 15 Sep 2020, Satchmo said the following...

What do you do if your feed goes down?

How will I know if my single FidoNet hub run by an elderly gentleman is ever coming back?

You are able to gets feeds from more than one source. I get my "main" feed from Nick (atreyu) (Z1) and another feed from Paul (Avon) (Z3). You can ask them if they'd be willing to set you up, I'm sure there are others here who'd be willing to give you another feed as well.

Jay

... Why don't ants get covid-19? Because they have little anty bodies.

--- Mystic BBS v1.12 A46 2020/08/26 (Raspberry Pi/32)
* Origin: Northern Realms (21:3/110)

On 15 Sep 20 14:19:14, Satchmo said the following to All:

All it would take is a syncronisation of the credentials for feeds/points an a log of their suscribed echos/areas. (This could even be done via a FTN ech if needed)

This could then be duplicated between however many systems for redundancy.

I understand your thoughts and frustrations, but first of all please try to work it out with your Hub. If you can't, then you are more than happy to get a feed direct from my system. I administer Fidonet for Zone 1 and feed close to a hundred systems.

The second problem is that the DNS CNAME and shared-credentials and all those ideas just wouldn't work in Fidonet politically. It may work in an Othernet like this one or others where theres a bit more enthusiasm by some to give it a try but Fido is mostly dead on accepting new innovations.

Mail redundancy is already being done by systems "meshing" with eachother. What this means is that instead of echomail flowing in the traditional
path of A->B->C we now have mail exchanged directly amongst A<->B<->C.

This has worked well for the most part, for the last few years or so.

Atreyu

--- Renegade vY2Ka2
* Origin: Joey, do you like movies about gladiators? (21:1/176)

On 15/09/2020 3:30 pm Atreyu said...

I understand your thoughts and frustrations, but first of all please try to work it out with your Hub.

Thanks for the reply and the kind offer of a feed. I will try and work it out with my current hub. I have emailed him this morning. Hope it's just a change in addressing that I've missed.

Working in redundant systems and net works all day, I'm used to some redundancy. NOT relying on single points of failure. What If I had several nodes/points downstream?

Satchmo





--- ENiGMA 1/2 v0.0.12-beta (linux; x64; 12.18.3)
* Origin: Sonic BBS (21:4/144)

On 15 Sep 2020, Satchmo said the following...

What do you do if your feed goes down?

I noticed this morning that my FidoNet hub is no longer responding via my polls. Now I know this isn't FidoNet but it got me thinking about the
best way to stop this from happenning and provide some redundancy.

Provided your mail setup is smart enough not to backfeed and has decent dupe detection there's nothing stopping you from finding alternate or having multiple feeds.

Historically this wasn't a practical option because of calling circles, hardware resources and time spent polling, so if your hub went down someone else in your area would have to take up the reins and you'd get set back up with them.

Nowadays redundancy requires minimal bandwidth and if your hub goes down for good it's easy to find and connect to a new one.

--- Mystic BBS v1.12 A46 2020/08/26 (Linux/64)
* Origin: Subcarrier BBS (21:2/140)

On 15 Sep 2020, Satchmo said the following...

What do you do if your feed goes down?

I noticed this morning that my FidoNet hub is no longer responding via my polls. Now I know this isn't FidoNet but it got me thinking about the
best way to stop this from happenning and provide some redundancy.

I've actually got 5 different fido feeds here. I do end up with a lot of dupe messages getting caught, but at least I know the messages are coming through.

If you need a fido feed, send me a netmail and I can get you set up here.


---

Black Panther(RCS)
aka Dan Richter
Castle Rock BBS
telnet://bbs.castlerockbbs.com
http://www.castlerockbbs.com
http://github.com/DRPanther
The sparrows are flying again...

--- Mystic BBS v1.12 A46 2020/08/25 (Linux/64)
* Origin: Castle Rock BBS - bbs.castlerockbbs.com (21:1/186)

On 15/09/2020 6:45 pm Black Panther said...

I've actually got 5 different fido feeds here. I do end up with a lot of dupe messages getting caught, but at least I know the messages are coming through.

If you need a fido feed, send me a netmail and I can get you set up here.

Thanks, netmail incoming :)


--- ENiGMA 1/2 v0.0.12-beta (linux; x64; 12.18.3)
* Origin: Sonic BBS (21:4/144)

On 15 Sep 20 15:41:52, Satchmo said the following to Atreyu:

Working in redundant systems and net works all day, I'm used to some redundancy. NOT relying on single points of failure. What If I had several nodes/points downstream?

You're telling me. I have close to a hundred links on my system thats running on outdated hardware running an outdated OS using outdated software tools.

Oh sure, I have spare parts and the inclination to keep it going for the hell of it, but when there is a failure, then yes nobody gets mail for the moment. Most Sysops who are concerned with redundancy have meshed-echomail feeds. The duplicate messages are weeded out as long as the SEEN-BY lines are not screwed around with or stripped out.

Thats the best that can be done for now. To have a much cleaner level of redundancy in the way you mentioned would require retooling software and processes which have worked for many Sysops for years, decades. There just isn't a clear incentive to do so, and Fidonet Sysops can be a picky bunch.

Atreyu

--- Renegade vY2Ka2
* Origin: Joey, do you like movies about gladiators? (21:1/176)

On 15 Sep 2020 at 07:32p, Atreyu pondered and said...

meshed-echomail feeds. The duplicate messages are weeded out as long as the SEEN-BY lines are not screwed around with or stripped out.

Side node - I still do plan to fix 3:770/1 .. sorry it's taking so long. Building a new box soon and migrating it to HPT when done.

--- Mystic BBS v1.12 A46 2020/08/26 (Windows/32)
* Origin: Agency BBS | Dunedin, New Zealand | agency.bbs.nz (21:1/101)

I understand your thoughts and frustrations, but first of all please try to work it out with your Hub. If you can't, then you are more than happy to get a feed direct from my system. I administer Fidonet for Zone 1 and feed close to a hundred systems.

I've always wondered about how that works. If a Hub drops out unexpectedly for any reason, other than a short delay because of hardware issues of the like,. how does the word make it around and everyone down stream get the information to get their new feed info?

The second problem is that the DNS CNAME and shared-credentials and all those ideas just wouldn't work in Fidonet politically. It may work in an Othernet like this one or others where theres a bit more enthusiasm by some to give it a try but Fido is mostly dead on accepting new innovations.

I've never understood that, until I wound up working in a large corporation. Resistance to change simply because it's always been done this way just isn't part of my thinking.



--- AfterShock/Android 1.6.7
* Origin: HOUSTON, TX (21:4/111)

On 15 Sep 20 21:49:52, Charles Pierson said the following to Atreyu:

I've always wondered about how that works. If a Hub drops out unexpectedly for any reason, other than a short delay because of hardware issues of the like,. how does the word make it around and everyone down stream get the information to get their new feed info?

In that scenario word sortof travels "fast" now because of the Internet and system-meshing; and generally a hub that feeds many systems is ran by someone "known" to the rest of the people in the hobby.

On the other hand I've seen systems appear to be running fine when in fact the Sysop had died and the family was so consumed with grief and funeral arrangements that it ran for weeks on end before finally being shutdown.

Thats the thing... anything can happen to anyone, anytime, a BBS or
hubbing a silly message network is the last sort of arrangement made in a will or instructions to loved ones.

For the longest time I was strongly against having a Co-Sysop. I was bombarded in the 90's with many people wanting to be Co-Sysop. Nowadays with running a busy Fido system I see how its important to some people and important enough that it needs to be taken over by someone if I'm hit by a bus tomorrow.

Othernet like this one or others where theres a bit more enthusiasm by some to give it a try but Fido is mostly dead on accepting new innovations.

I've never understood that, until I wound up working in a large corporation Resistance to change simply because it's always been done this way just isn part of my thinking.

I kindof understand some of the mentality as I too run a system that has remained unchanged for years, decades even. I'd be very skeptical of
anything new; but stuff like Internet Rex and BinkD were loved by everyone.

Atreyu

--- Renegade vY2Ka2
* Origin: Joey, do you like movies about gladiators? (21:1/176)

Quoting Atreyu to Charles Pierson <=-

On the other hand I've seen systems appear to be running fine when in
fact the Sysop had died and the family was so consumed with grief and funeral arrangements that it ran for weeks on end before finally being shutdown.

Mine is not even run in my home anymore, I'd say it would run until
the BBS databases got corrupted. Since windows updates and reboots on
it's own... Likely would keep coming back online for years. LOL

Thats the thing... anything can happen to anyone, anytime, a BBS or hubbing a silly message network is the last sort of arrangement made
in a will or instructions to loved ones.

True enough, it's not in my will anywhere that's for sure.

Shawn

... Hot water heaters... Hot water needs heating?

--- EzyBlueWave V3.00 01FB001F
* Origin: Tiny's BBS - telnet://tinysbbs.com:3023 (21:1/130)

I kindof understand some of the mentality as I too run a system that has remained unchanged for years, decades even. I'd be very skeptical of anything new; but stuff like Internet Rex and BinkD were loved by everyone.

Seems like it's a, "this is a pain point for most of you, even now, and this
is a solution" sort of deal.

As opposed to doing something like switching a programming language or keyboarding style, where the benefits have to be pretty massive before it's moderately reasonable to spend the time to switch.

Which reminds me of the time I switched over to the Dvorak keyboard for a while. I only got up to 25wpm or something, and am unconvinced that it'd be measurably different from my typing speed with QWERTY, even under ideal conditions.

But all the same, if I still had my AlphaGrips, I'd kinda want to learn that keyboard a bit better, because having a keyboard in my hands while I lay back sounds pretty cool.

(The AlphaGrips is a two-handed controller that does look like a game controller, but a bit bigger, with a track ball, and a _ton_ of rocker
buttons and shift keys to be able to type the characters you need.)

--- Mystic BBS v1.12 A46 2020/08/26 (Linux/64)
* Origin: Storm BBS (21:2/108)

On 16 Sep 20 12:15:43, Adept said the following to Atreyu:

I kindof understand some of the mentality as I too run a system that ha remained unchanged for years, decades even. I'd be very skeptical of anything new; but stuff like Internet Rex and BinkD were loved by everyone.

Seems like it's a, "this is a pain point for most of you, even now, and this is a solution" sort of deal.

It was basically that way with Internet Rex and BinkD.... lots of Sysops had to reconfigure their dialup-modem systems to integrate those new things at the time.

Which reminds me of the time I switched over to the Dvorak keyboard for a while. I only got up to 25wpm or something, and am unconvinced that it'd be

I too had a Dvorak and could never get past 25wpm.

Atreyu

--- Renegade vY2Ka2
* Origin: Joey, do you like movies about gladiators? (21:1/176)

Satchmo wrote to All <=-

I noticed this morning that my FidoNet hub is no longer responding via
my polls. Now I know this isn't FidoNet but it got me thinking about
the best way to stop this from happenning and provide some redundancy.

People in Zone 2 refer to the "fidoweb" - set up redundant uplinks
and let your BBS's dupe detection software sort out the duplicate
messages. While censorship is the usual inspiration for duplicate
links, it also provides redundancy for situations like this.

Having echolists synchronized between multiple feeds would be nice.

Here's to hoping your uplink is OK.



... The robots can go off-script?
--- MultiMail/XT v0.52
* Origin: realitycheckBBS.org -- information is power. (21:4/122)

Something I havn't really thought about until recently....

What do you do if your feed goes down?

I'll have a cup of tea, grab a book and relax. It will probably fix itself eventually and if not I'll deal with it then.


--- NiKom v2.5.0
* Origin: Delta City (deltacity.se, Vallentuna, Sweden) (21:2/130.0)

Which reminds me of the time I switched over to the Dvorak keyboard for a while. I only got up to 25wpm or something, and am unconvinced that it'd be measurably different from my typing speed with QWERTY, even under
ideal conditions.

The IIgs came with a dvorak setting for its keyboard in its "rom". I only ever came across one person that used it. I don't know that he was a lot faster, although he felt it was easier to use. He certainly couldn't drive a regular qwerty though.

Spec


*** THE READER V4.50 [freeware]
--- SuperBBS v1.17-3 (Eval)
* Origin: Scrawled in haste at The Lower Planes (21:3/101)

On 17 Sep 2020, Spectre said the following...

The IIgs came with a dvorak setting for its keyboard in its "rom". I
only ever came across one person that used it. I don't know that he
was a lot faster, although he felt it was easier to use. He certainly couldn't drive a regular qwerty though.

Dvorak is certainly more efficient in that movement is minimized and your fingers leave the home row less often, at least when typing English, but that doesn't mean learning it will turn you into a speed demon.

--- Mystic BBS v1.12 A46 2020/08/26 (Linux/64)
* Origin: Subcarrier BBS (21:2/140)

The IIgs came with a dvorak setting for its keyboard in its "rom". I
only ever came across one person that used it. I don't know that he
was a lot faster, although he felt it was easier to use. He certainly couldn't drive a regular qwerty though.

My most entertaining instance of someone else who used it was a friend who
had a computer set up for Dvorak, but on a keyboard where none of the keys
were labeled. So it was basically impossible for anyone else to use his computer.

But, yeah, on speed... Still, once someone gets past 50wpm or so, the
limiting factor is probably more about how quickly you can compose than how quickly you can type.

Though I do still kinda like how much my hands type on their own -- there
must be a ton of muscle memory involved, since there's a variety of times
where I'll be typing a sentence, and my hands do an autocomplete. And
sometimes get it right even though it wasn't what I was thinking.

--- Mystic BBS v1.12 A46 2020/08/26 (Linux/64)
* Origin: Storm BBS (21:2/108)

Dvorak is certainly more efficient in that movement is minimized and your fingers leave the home row less often, at least when typing English, but that doesn't mean learning it will turn you into a speed demon.

Yeah, true, though QWERTY is good at alternating between the hands, which
might have some sort of speed advantage.

I'm not sure which one uses the pinkies less, though.

But I'm no purist. The difference just seems pretty minimal.

I suppose kinda like learning Esperanto or Lojban. Such endeavours can still
be fun, though.

--- Mystic BBS v1.12 A46 2020/08/26 (Linux/64)
* Origin: Storm BBS (21:2/108)