On Nov 8, 2011, at 3:24 PM, Richard Jones wrote: > On 9 November 2011 06:14, Brian Curtin <[email protected]> wrote: >> On Tue, Nov 8, 2011 at 13:09, anatoly techtonik <[email protected]> wrote: >>> http://www.isup.me/pypi.python.org >> >> Works for me, and that page agrees: "It's just you. >> http://pypi.python.org is up." > > Thanks for the report. Going by the times in the email headers > (2011-11-08 20:00 to 20:09 CET) I've looked through the logs and there > was a 6 minute period during which the request handling rate tapered > right off. Not sure what the cause could have been at this time. > There's a ton of what appear to be delayed requests finishing up > around that time, which I've seen before. Interleaved with the > requests timestamped above are some with timestamps of 5-10 minutes > earlier from a variety of clients. I've discussed this with Martin > before but with no real conclusion.
When I got to the machine I could tell from the load averages that it was just recovering from some kind of load spike. Something like 14, 50, 150. --Noah
signature.asc
Description: Message signed with OpenPGP using GPGMail
_______________________________________________ pydotorg-www mailing list [email protected] http://mail.python.org/mailman/listinfo/pydotorg-www
