Danny McGrath dmcgrath
Danny McGrath commented on issue infrastructure/blender-projects-platform#24 2023-02-14 05:31:42 +01:00
Gitea: Poor page load and download times for some countries

oops, I realize I closed this earlier. I was wondering what happened when I misclicked that!

Danny McGrath commented on issue infrastructure/blender-projects-platform#24 2023-02-14 05:26:18 +01:00
Gitea: Poor page load and download times for some countries

I dug a little bit more. I tried different sysctl's on the host, checked the netstat -m nmbcluster values, mtu's, but no matter what I just can't seem to get the window size to go past around 256k…

Danny McGrath commented on issue infrastructure/blender-projects-platform#24 2023-02-14 03:21:46 +01:00
Gitea: Poor page load and download times for some countries

I did look more into the interface errors, but this time from the host pov. It would seem that, based on the total interface errors and uptime, we are receiving about 1.173 million interface…

Danny McGrath closed issue infrastructure/blender-projects-platform#24 2023-02-14 03:15:37 +01:00
Gitea: Poor page load and download times for some countries
Danny McGrath commented on issue infrastructure/blender-projects-platform#24 2023-02-14 03:15:36 +01:00
Gitea: Poor page load and download times for some countries

No networking expert so feel free to ignore if this is obvious or wrong.

To me it seems a small TCP window size as found by @LazyDodo is exactly the kind of thing that would lead to a…

Danny McGrath commented on issue infrastructure/blender-projects-platform#24 2023-02-14 00:56:11 +01:00
Gitea: Poor page load and download times for some countries

In terms of network statistics, I took another look at the switch that directly connects to the ISP. There are (still) tons of errors from the server where Phabricator/Git/Svn/Buildbot run, but…

Danny McGrath commented on issue infrastructure/blender-projects-platform#24 2023-02-13 23:38:07 +01:00
Gitea: Poor page load and download times for some countries

Last weekend when I took a quick peek into this problem, it was found that git checkouts gave results that were fastest in the same rack/network as the VM of gitea.

Performance at home was…