donc ce comportement s’est empiré ces derniers jours ?
On my case, I would say it became noticeable in the last months. Before January, I never noticed the impact of Jami on the network. One or two months ago, my wife started complaining that the network was unusable at time, particularly when I started my laptop, and after a few days investigation Jami was spotted as the culprit. Presently, when Jami starts, it floods my router and no other software (web browser, email client, ssh, etc.) can reach the Internet. I can’t even reach other equipment on my LAN like the email server. When I quit Jami, the Wi-Fi becomes usable instantaneously.
Demonstration with a single ping to the router. Note the high latency and packets drop before packet 79. Between packets 76 and 79, I quit Jami: packet latency is divided by 100 and immediately the Web page shows and emails are received.
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=50 ttl=64 time=208 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=52 ttl=64 time=170 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=54 ttl=64 time=170 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=55 ttl=64 time=158 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=59 ttl=64 time=143 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=62 ttl=64 time=154 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=64 ttl=64 time=182 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=65 ttl=64 time=165 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=66 ttl=64 time=151 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=67 ttl=64 time=148 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=68 ttl=64 time=155 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=70 ttl=64 time=182 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=76 ttl=64 time=210 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=79 ttl=64 time=1.80 ms <== Quit Jami
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=80 ttl=64 time=1.87 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=81 ttl=64 time=0.979 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=82 ttl=64 time=2.02 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=83 ttl=64 time=1.38 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=84 ttl=64 time=0.956 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=85 ttl=64 time=1.76 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=86 ttl=64 time=4.57 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=87 ttl=64 time=1.41 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=88 ttl=64 time=1.93 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=89 ttl=64 time=3.61 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=90 ttl=64 time=3.63 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=91 ttl=64 time=3.47 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=92 ttl=64 time=3.62 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=93 ttl=64 time=3.80 ms
64 bytes from walter.alterna.tv (26**********::1): icmp_seq=94 ttl=64 time=3.81 ms
To answer precisely the question, I can’t answer in days. For the last two months, for sure it degraded.
Well, I mean the switch in Jami’s settings. I don’t have UPnP in my network, but the log shows it still tries a few times.
I left a couple of comments (and questions) on your GitLab issue. In short, it seems likely that the problem is caused by the DHT operations performed by Jami on startup. However, I haven’t found anything so far that changed on Jami’s side in the last few months and could explain a sudden increase in the amount of generated UDP traffic (or number of distinct addresses contacted).
Can you think of anything that changed on your side around the time where Jami started interfering with your Wi-Fi? New Jami accounts created, contacts added, group conversations joined, router configuration changed, …?
I’ve updated the ticket with all details I thought of. No special changes recently, apart from the SQM plugin enabled on the router.