I would then expect to see a similar problem when connecting from my Linux workstation to the same x2go host over the same network, but that client does not seem to have the same issue. Not sure that UDP is the culprit, but as I have not seen the code, I cannot speak authoritatively on that.
Timm
- *"It's easier to fix bugs the same day you wrote the code. It can be very hard and time-consuming to fix bugs a month later when you've forgotten exactly how the code works." - Joel Spolsky, CEO/Founder of Stack Overflow*
On Wed, Jul 17, 2024 at 3:24 PM Michael Ashley <m.ashley@unsw.edu.au> wrote:
You don't need the particular equipment I mentioned to see the problem. If x2go is using UDP for keystrokes, then there is no guarantee that the data won't be missing/duplicated/out-of-order. The rate at which you see problems will depend on the network hardware, cables, etc, as well as how close to saturation the connection is.
On Wed, Jul 17, 2024 at 02:34:24PM -0700, Uranus Hertz wrote:
I have none of that equipment in place in either situation. At home, over the VPN, everything is Ubiquti. At the office, over LAN everything is Cisco. they can’t use those resources for other things that truly, deeply
- "If the result of your work consumes someone’s cognitive resources,
matter. This is NOT about consuming their time and attention while they're using your app. This is about draining their ability for logical thinking, problem-solving, and willpower after the clicking/swiping/gesturing is done." - Kathy Sierra On Wed, Jul 17, 2024 at 1:06 PM Michael Ashley <[2]m.ashley@unsw.edu.au> wrote:
I have seen massive problems with missing and re-ordered keystrokes when I was using an EnGenius ENH500v3 wifi bridge. This bridge had some sort of networking problem where pings would normally take a few milliseconds, but every 30 seconds or so they would take over a second. I presumed that TCP/IP would take care of any missing/duplicated/re-ordered packets. Does x2go use UDP? I replaced the ENH500v3 with a Ubquiti Nanostation Loco (a *far* better product) and the problem immediately went away. Regards, Michael-- Professor Michael Ashley School of Physics University of New South Wales http://www.phys.unsw.edu.au/~mcba