Subj : Re: Rougue bot
To   : Atreyu
From : apam
Date : Tue Nov 09 2021 01:05 pm

 > Then Avon should be the one to pull the plug on that system if in fact
 > its
 > causing problems for everyone. You are not allowed to have a feed
 > until you 
 > confirm you fixed your broken system. Harsh? Maybe... but an effective
 > "fix".

Yes, but what if Avon is away?

 > I'm the author of a major mailer/tosser and had added this
 > functionality in
 > my product recently. It wasn't because of this incident but due to the
 > stubbornness of some insisting on running software that strips Seen-By
 > lines.

So... you've already added the feature, but suggesting no one else should?

Why didn't you just cut the feeds of Seen-By stripping nodes?

I get that it doesn't happen very often, but when it does, what's the
harm in sending a few identical messages to /dev/null? Or at least having
the option to?

It's not that different from dupe checking. Sure people want to post
identical messages periodically, that's what the time out is for..

 > I wouldn't press Next 8000 times. I would just tell my BBS to mark the
 > lastread pointer ahead 8000 messages. Its not a problem for me, is my
 > point.

Yeah I get it. So it's not a problem for you, it's not a problem for
anyone else? Granted it happened in the bot echo, so nothing particularly
important in there, but what if it was in FSX_GEN, especially if you have
message base pruning on? And it's not like 8000 messages didn't have odd
normally posted messages inbetween.

 > But I'm not a fan of adding this level of checking for a few reasons
 > which
 > could morph into a whole other uninteresting topic altogether.

Ok, I see you're not a fan... I guess the whole other uninteresting topic
is performance degradation? If so, why not just have it as an option and
meant for end nodes? Hubs can spew out thousands of dupes, and people who
do actually care can toggle it on, and have some kind of protection?

 > Every couple of years we seem to have someone rescan/dump old messages
 > everywhere... often with new MSGID's, different header CRC, dates, you
 > name it... I'm actually used to it now.

Yes, I've seen it happen plenty of times too. I guess it doesn't happen
very often, I've only seen this one time with the 8000~ identical
messages, but that doesn't mean it won't happen again, and I don't see
the issue with having a discussion about ways to proactivley prevent it.

I certainly don't think it's an asinine suggestion.

Andrew

--
|03Andrew Pamment |08(|11apam|08)
|13Happy|10Land |14v2.0|08!|07

--- Talisman v0.30-dev (Linux/x86_64)
 * Origin: HappyLand v2.0 - telnet://happylandbbs.com:11892/ (21:1/182)

.