So, just posted from my mastodon.social account - with a mention of this account, toot showed up here 3 seconds later.

I also posted with no mention - that toot still has not arrived after 3 minutes.

@djsundog Yeah I can only speculate as to what, based on my knowledge of OStatus and general Ruby knowledge, but *something* isn't working quite right there anyways.
@maiyannah I mostly want to make sure it's localized to their queues and isn't also affecting the lab, but yeah, I'm thinking there's a stuffed queue somewhere over yonder - the sysadmin in me would be log spelunking right about now were those my boxen :)
@djsundog Are you also on the head revision of master?  Because the relevant change is like ... today ish.  Maybe yesterday.
@maiyannah no, as soon as tagged releases hit his repo I moved to the tags, and have been holding off for about a day after the first wave of folk install the tagged release. Staying on the bleeding edge of that codebase was taking too much post-upgrade troubleshooting time for my current tastes ;)
@djsundog @maiyannah Given the rate of recent change (and the apparent misunderstanding of how semver works), I'd suggest waiting a week. Supposedly they're running dotsocial on release candidates for a day before public release. That seems to be not enough time for issues to show up.
@sungo @djsundog No it really isn't.  I run new changes at least a couple weeks on HLA and now Plateia before they rotate from Nightly to Master.
@maiyannah @sungo you'll never win the race to move fast and break things like that, pal ;)
@djsundog @maiyannah I very much appreciate how not-broken pA is :) Even nightly is more stable and not-broken than *cough* other products.
@sungo @djsundog You know I often get a bit embarassed because the bugs that HAVE slipped the net have been the "embarassingly simple" kinds of things.
@maiyannah @sungo those are just the times you let the younger generation see that it's okay to let one slip once in a while, even the silly ones :)
@djsundog Uh. To us, @maiyannah *is* the younger generation :)
@sungo @maiyannah true dat :) but I've got grandkids that need good role models, gotta keep the role model pipeline full, can't have generational role model starvation... ;)
@djsundog @maiyannah Do me a favor and starting teaching those grandkids the fundamentals of TCP/IP. By the time they're ready to take over the industry, they'll be the only ones who understand it.
@sungo @djsundog @maiyannah Ugh, ain't that the truth. All the kids these days showing up have no conception of the underlying tech - in another ten years, those of use who actually know how non-layer-7 stuff works will be the very very small minority.
@munin @maiyannah @djsundog I'm gonna get a tat that reads IT IS NOT THE FUCKING NETWORK YOUR APP IS SHIT

@sungo @djsundog @maiyannah

The app's bad performance is because the app is too fragile to handle the reality that no network is perfectly functioning. :-/

@munin @maiyannah @djsundog As a person who has done lots of development, sysadmin, and network, I am often called upon to sit in a room and mediate while we all tell the developers NO REALLY THIS IS YOUR FUCKING FAULT
@sungo @munin @djsundog IME 99% of network errors are three actual kinds of errors:

1:  PEBKAC
2: The app is shit
3: Someone is cutting corners in the server budget and the server cant handle the load
@maiyannah @djsundog (4) A sysadmin decided to get cute with a server configuration. "IPv6 isn't working!" You're blocking all ICMP. "StackOverflow said I should!" You're fired.
@sungo @maiyannah okay, you've convinced me that the "Young Person's Illustrated Guide to the IETF RFCs" should be a thing.