On Saturday, 7 July 2018 10:07:39 PM AEST Raphael Hertzog wrote: > Well, mtr is certainly not a browser request. It's merely a ping on > intermediate hosts. So it's easy to reproduce. `mtr` output is the only hard evidence of a problem in Debian infrastructure. We are lucky to have something as prominent as `mtr`. Just black hole router would be much more difficult to diagnose. `mtr` may be giving us a symptom of a root cause. > So I'm not sure that you are on the right track. Could be but something dodgy seems to be happening between "po1.ar2.dc1.yo26.yrk.bytemark.co.uk" and "bm-bl1.debian.org". I think it is worth investigating. > For a more interactive discussion, they are at #debian-admin on OFTC. > But I have a hard time believing that the Debian infrastructure is at > fault here. Not too hard for me. We've thoroughly investigated everything on our site. Among hundreds of web site, tracker.debian.org is the only one exhibiting such problem. Believe me, if any site we've visited in Firefox made it hang for three minutes I would have been told let alone noticed myself. I also had hard time believing that the problem is on Debian side. Yet today I've got consistent evidence of packet loss in Debian infrastructure. Firefox problems aside, packet loss is probably worth reporting. > If you want to go further, you should probably also seek the help of > Firefox developers since it's behaving badly compared to chromium in the > same (possibly problematic) network environment. Of course. Sorry to bother you but unfortunately this is a quite a serious problem because of how much it affects my work... I'll see if I can get any hints. -- Cheers, Dmitry Smirnov.
Attachment:
signature.asc
Description: This is a digitally signed message part.