download.qt.io down?

[StevenNH]

My co-workers in Boston are seeing the same thing. The site is up and down. I'm not sure if there is a networking issue somewhere. The site was down for me at 5AM EST, and is back up now, but it's been going up and down. Our github runners have been failing off and on for the past few days (new problem).

Sorry, NH is New Hampshire, USA, Seeing the same in Boston USA.

[Axel Spoerl]

@StevenNH said in download.qt.io down?:

NH is New Hampshire

Oh, forgive my ignorance :-)
It seems like something along the way from Scandinavia to the US then.
That would explain the intermittent behavior.
The connection is pretty consistent and fast here in Norway.
To which IP address does download.qt.io resolve at your end?

axel@axelsqt:~> ping download.qt.io
PING download.qt.io (77.86.162.2) 56(84) bytes of data.
64 bytes from 77.86.162.2: icmp_seq=1 ttl=52 time=22.3 ms
64 bytes from 77.86.162.2: icmp_seq=2 ttl=52 time=20.8 ms
64 bytes from 77.86.162.2: icmp_seq=3 ttl=52 time=24.5 ms
64 bytes from 77.86.162.2: icmp_seq=4 ttl=52 time=21.0 ms
64 bytes from 77.86.162.2: icmp_seq=5 ttl=52 time=22.0 ms
64 bytes from 77.86.162.2: icmp_seq=6 ttl=52 time=20.9 ms
64 bytes from 77.86.162.2: icmp_seq=7 ttl=52 time=22.3 ms
64 bytes from 77.86.162.2: icmp_seq=8 ttl=52 time=21.0 ms
64 bytes from 77.86.162.2: icmp_seq=9 ttl=52 time=21.7 ms
64 bytes from 77.86.162.2: icmp_seq=10 ttl=52 time=24.6 ms
64 bytes from 77.86.162.2: icmp_seq=11 ttl=52 time=21.9 ms
64 bytes from 77.86.162.2: icmp_seq=12 ttl=52 time=22.5 ms
64 bytes from 77.86.162.2: icmp_seq=13 ttl=52 time=21.0 ms
^C
--- download.qt.io ping statistics ---
13 packets transmitted, 13 received, 0% packet loss, time 12013ms
rtt min/avg/max/mdev = 20.843/22.051/24.624/1.217 ms
``

[StevenNH]

Ping gets me 77.86.162.2. Right now I can ping, but the site won't load.

Pinging download.qt.io [77.86.162.2] with 32 bytes of data:
Reply from 77.86.162.2: bytes=32 time=112ms TTL=47
Reply from 77.86.162.2: bytes=32 time=118ms TTL=47
Reply from 77.86.162.2: bytes=32 time=117ms TTL=47
Reply from 77.86.162.2: bytes=32 time=111ms TTL=47

[david sastre]

Same issue for me here for the past 2 days. I am in Spain.

ping download.qt.io
PING download.qt.io (77.86.162.2) 56(84) bytes of data.
64 bytes from 77.86.162.2 (77.86.162.2): icmp_seq=1 ttl=48 time=73.6 ms

That is what I get, however accessing the url in browser or wget does not work. I noticed this because my github CI workflows were failing, I guess its servers are in the US, although I couldn't say for sure.

Thanks for the attention Alex.

[Axel Spoerl]

That seems to be depending on the location.
It's offline in Berlin as well.

[StevenNH]

@Axel-Spoerl said in download.qt.io down?:

That seems to be depending on the location.
It's offline in Berlin as well.

It's odd that ping works but the site does not. That would seem to imply something flaky on the server, or selective port blocking?

[PPRiphagen]

Just to add to the story that here in the Netherlands it's the same, ping is fine, website is down.
Our azure pipelines experience the same issue.

[Axel Spoerl]

Unfortunately confirm outages, which are caused by traffic overload. Gravity may vary from region to region.
This only affects opensource downloads. Switching mirrors solves the problem, unless those mirrors are also affected. We have no visibility on those, however.

Commercial downloads via https://account.qt.io/s/downloads are unaffected and work everywhere without issues.

[Axel Spoerl]

I have posted a current mirror list on the development mailing list (can't upload html files here).

[Pl45m4]

Just to report back:
download.qt.io up and stable
(Northrhine-Westphalia region, Germany, 60km from Dutch border)
(Weird that @PPRiphagen also reports outages)

@Axel-Spoerl how can the location make a difference whether the site/resources are reachable or not?!
If the host itself is ping-able there must be some webserver/service (or port) issue so the page isn't provided properly?!

[Axel Spoerl]

Seems like an unusually high number of bots were connecting. Maybe some of them flooded internet nodes.
We have shut down the web server and started it again, interrupting overboarding traffic.
That has put it online again. Our IT is on it, analyzing the failure and monitoring performance.

[Firewave]

Seems like an unusually high number of bots were connecting. Maybe some of them flooded internet nodes.

Part of this might be that the caching of the jurplel/install-qt-action GitHub action is currently broken because the underlying mechanism has been sunsetted.

There are also certificate issues with the https://ftp1.nluug.nl mirror.

And there is a single(!) report of potentially suspicious files being served: https://www.reddit.com/r/QtFramework/comments/1k16ijb/comment/mnk5p3u/

More details:
https://github.com/jurplel/install-qt-action/issues/283
https://github.com/jurplel/install-qt-action/issues/285
https://www.reddit.com/r/QtFramework/comments/1k16ijb/any_word_on_httpsdownloadqtio/

[cristian-adam]

https://web.archive.org/web/20250404053014/https://download.qt.io/static/mirrorlist/ should help when download.qt.io is down in the future.

[Steve Pritchard]

@Firewave that was my report re: malware. It could just be a false positive but it’s a strange coincidence that it happened at the same time as the network disruption. I’ve never had a problem before with my built packages

[Steve Pritchard]

@Firewave here is a helpful GitHub action fork (not mine) that fixes caching: https://github.com/Kidev/install-qt-action

[Steve Pritchard]

@Axel-Spoerl Thanks for confirming the hashes. I wonder though if a mirror could be compromised and the attacker is DDoSing the primary site to drive traffic to the compromised mirror.

Has anyone else noticed had any built packages identified as malware infected since the network disruption? I’m using Innosetup as my installer.