daniels220

Okay, this turns out to be a bad interaction between libuv and [this script](http://snippets.scripts.mit.edu/gitweb.cgi/.git/blob/HEAD:/bash/newline-fix.bash). Something is still _different_ about `uv`s behavior depending on whether or not I close the tty...

Note also per the (duplicate, oops) #13159 that I reported that this is a regression since Arachne Beta 2 (and Cura 4). These older versions not only avoided many separate...

Okay, so I think you're technically right that the problem is due to the involvement of an Arachne beta. However, I think Cura is, first, doing an _awful_ job of...

I just tried the reproduction in Pharo 11[^1] on Linux. Message size 1MB ±16kB, 4 clients, 10k reps each. Three of the four clients finished their run normally, but with...

I made some improvements to the socket load-testing code from #10980—pulled the data sizes to class variables, added similar configuration for send/receive buffer sizes, and attempted to improve detection of...

I am pretty familiar with Dolphin Smalltalk's UI code, and it indeed uses a very similar core approach of exiting the message loop when that particular process is no longer...

This is still the case in Pharo Launcher 3.0.1. Of note, trying to quit Pharo Launcher by other methods like right-clicking on the Dock icon, pressing Cmd+Q with it selected...

I would say, equally common as noticing a typo in the commit message right after hitting commit, I notice a _change_ that should have been part of that commit that...

We would _like_ to deploy using a 32-bit VM to save memory overhead (we run many separate servers each with moderate memory use, such that they will never individually need...

I can't fully follow most of this, but thought I'd point something out. I noticed in digging around to deal with [pharo#11723](https://github.com/pharo-project/pharo/issues/11723) that there already exist separate `primSocket:bindTo:port:` and `primSocket:listenWithBacklog:`...