real life test next tuesday 21:00 berlin time (utc+1)

[luckow]

@humptydumpty to be fair. I will wait for a 24/7 instance of jitsi on Cloudron until the LDAP settings work. I need my "shared" Cloudron instance for so many projects that I'm not ready to abandon it to the world. A reliable public instance for jitsi is https://meet.jit.si/ or https://meet.ffmuc.net/.

20 minutes before the test starts, I "fire up" the jitsi app. And yes: you only require a "modern" browser (in case you missed it: shame on you ) and ideally a headset.

[nebulon]

Great idea indeed, happy to meet you all there

[luckow]

Reminder: 52 minutes to go.

[timconsidine]

@luckow it will be interesting
I have another call at that time so I may join on mute, and may not stay for very long. I hope that does not seem rude !

[luckow]

@timconsidine I'm fine with it. Maybe next time with a bit more planning. For today, we need load. Every attendee counts. Bots are welcome too

[girish]

I could only be for a short time since I am driving to a meeting! But it worked well for the short time was around

[luckow]

@girish we are here

[BrutalBirdie]

~10 Users Network statistics for the whole server NIC

Bits Sent:

caption: min 3.5 Mbps, avg 9.56 Mbps, max 12.16 Mbps

Bit Received:

caption: min 3.7 Mbps, avg 6.49 Mbps, max 8.21 Mbps

[luckow]

So to me, it looks like a reliable first app package. IMHO, up to 15 or 20 participants should not have any relevant problems with network, RAM or CPU.
Thank you for taking part in this little experiment

[timconsidine]

@luckow thank you for for organising
I tried to join but only managed 10 seconds because I messed up the muting and it disturbed my other meeting.
Looks like it performed well

[ei8fdb]

@luckow Sorry I couldn't join I'm just home now.

I'm curious to know the resources available for the jitsi server - CPU, RAM as configured in the cloudron UI.

My (very ad-hoc) testing so far has not been too successful, but I've not had a chance to dedicate time.

[BrutalBirdie]

@ei8fdb default settings from the install.

[ei8fdb]

@BrutalBirdie Oh! Then I need to look again at things. Thanks.

[timconsidine]

@ei8fdb default install works fine for me
not load tested but for 1-to-1 calls, (my 90% use case), it works fine

[nebulon]

@ei8fdb currently the manifest has a default memory limit set of 1.5Gb. However that value is really just a wild guess on my side.

[scooke]

A few others ought to try, but with lower spec'ed machines, possibly running other apps on the Cloudron too. Why? I'd like to use this but I can't afford a 29 euro/month VPS (looks like a RS 4000 G9)(yet). How good is it with less?

[ei8fdb]

@scooke said in real life test next tuesday 21:00 berlin time (utc+1):

A few others ought to try, but with lower spec'ed machines, possibly running other apps on the Cloudron too. Why? I'd like to use this but I can't afford a 29 euro/month VPS (looks like a RS 4000 G9)(yet). How good is it with less?

This is exactly what I'm trying to do at the moment with my self-hosted Cloudron.

I'm not sure I yet have networking side working, but when I do I'll happily do a test with others. Give me a few more days to find the time to test more.

I assume my Internet bandwidth will be the limiting factor before my home server.

[jdaviescoates]

I've done a couple of tests with Jitsi on my VPS.

The first one was with default RAM/ CPU resources with my Dad in a village in Spain on an iffy connection (sometimes it's good, often not): didn't really work at all. We couldn't see nor hear eachother.

I then upped the RAM and CPU just for good measure and did a test with my wife on our Wifi and my one my 4G mobile. Worked great.

Also did a test with someone in the same country as me (UK), but both in a different country as the server (Germany). Mostly worked well, but sometimes the audio and video went out of sync.

So far not managed to test a meeting with more than 2 people in though...

I'm looking forward to LDAP integration and being able to record meetings...

[nebulon]

@jdaviescoates just for information. Jitsi will do a direct p2p connection with two meeting participants. Once a 3rd person joins, it will use an SFU. So the connectivity specifics are quite different between 2 and more meeting participants.