Hi! In my RoR application, hosted on textdrive there''s a redirect between pages. Somehow it might take up to minute to be proccessed. Here''s the part of my production.log: Processing XXXX(for a.b.c.d at Fri Sep 09 14:56:13 GMT 2005) ... Redirected to YYYY Completed in 0.13740 (7 reqs/sec) | DB: 0.11887 (86%) [XXXX] Processing YYYY(for a.b.c.d at Fri Sep 09 14:56:45 GMT 2005) ... Rendering within layouts/empty Rendering drill/item (200 OK) Completed in 0.00717 (139 reqs/sec) | Rendering: 0.00362 (50%) | DB: 0.00068 (9%) [YYYY] As you can see - processing of page XXXX started at 56:13 and took 0.1 seconds, but processing of YYYYstarted 32 seconds later without apparent reason! Any idea how can I figure out what''s going on? -- Regards, Mike
Mike Aizatsky wrote:> Hi! > > In my RoR application, hosted on textdrive there''s a redirect between > pages. Somehow it might take up to minute to be proccessed.I guess you''re running Lighttpd. v1.4.something has a bug where redirects take around 30 seconds. You can get around it by disabling Keep Alive: server.max-keep-alive-requests = 0 -- Jakob L. Skjerning - http://mentalized.net
On 9/9/05, Jakob L. Skjerning <jakob-pixy5vpirPnEueBKFXcDjA@public.gmane.org> wrote:> Mike Aizatsky wrote: > > Hi! > > > > In my RoR application, hosted on textdrive there''s a redirect between > > pages. Somehow it might take up to minute to be proccessed. > > I guess you''re running Lighttpd. v1.4.something has a bug where > redirects take around 30 seconds. You can get around it by disabling > Keep Alive: > > server.max-keep-alive-requests = 0Man, there seem to be a lot of problems with lighttpd 1.4.
Had this same issue - what are the performance implications of setting keep-alive to zero? I looked through the lighttpd docs but they are pretty thin. Thanks, - Derek On 9/9/05, Joe Van Dyk <joevandyk-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> On 9/9/05, Jakob L. Skjerning <jakob-pixy5vpirPnEueBKFXcDjA@public.gmane.org> wrote: > > Mike Aizatsky wrote: > > > Hi! > > > > > > In my RoR application, hosted on textdrive there''s a redirect between > > > pages. Somehow it might take up to minute to be proccessed. > > > > I guess you''re running Lighttpd. v1.4.something has a bug where > > redirects take around 30 seconds. You can get around it by disabling > > Keep Alive: > > > > server.max-keep-alive-requests = 0 > > Man, there seem to be a lot of problems with lighttpd 1.4. > _______________________________________________ > Rails mailing list > Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org > http://lists.rubyonrails.org/mailman/listinfo/rails >-- Derek Haynes HighGroove Studios - http://www.highgroove.com Keeping it Simple. 404.593.4879
> Man, there seem to be a lot of problems with lighttpd 1.4.What other problems? We''ve been running 1.4 for a bit, and the slow redirect issue has been the only one we''ve hit.
Derek Haynes wrote:> Had this same issue - what are the performance implications of setting > keep-alive to zero? I looked through the lighttpd docs but they are > pretty thin.I think the impact is small unless your site has significant load. I''ve opened a bug ticket for the issue: http://trac.lighttpd.net/trac/ticket/243
Hey, thanks! That really sped up things for me. It was very weird that it was taking so long for redirects. Thanks again, Adrian Madrid Jakob L. Skjerning wrote:> Mike Aizatsky wrote: > >> Hi! >> >> In my RoR application, hosted on textdrive there''s a redirect between >> pages. Somehow it might take up to minute to be proccessed. > > > I guess you''re running Lighttpd. v1.4.something has a bug where > redirects take around 30 seconds. You can get around it by disabling > Keep Alive: > > server.max-keep-alive-requests = 0 >-- Adrian Madrid HyperX Inc. Mobile: 801.815.1870 Office: 801.566.0670 aemadrid-kSB444ljgzMmlAP/+Wk3EA@public.gmane.org www.hyperxmedia.com 9000 S. 45 W. Sandy, UT 84070 _______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails
The only other problem I have with 1.4.3 is that my upload progress bar does not work anymore. As soon as I moved from SVN trunk to release the bar stopped working. Anybody else noticed the same problem? I went through the checklist and it is all there still. And I know I didn''t change the code so the only change has been lighttpd. Thanks in advance, Adrian Madrid Michael Schoen wrote:>> Man, there seem to be a lot of problems with lighttpd 1.4. > > > What other problems? We''ve been running 1.4 for a bit, and the slow > redirect issue has been the only one we''ve hit. > _______________________________________________ > Rails mailing list > Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org > http://lists.rubyonrails.org/mailman/listinfo/rails >-- Adrian Madrid HyperX Inc. Mobile: 801.815.1870 Office: 801.566.0670 aemadrid-kSB444ljgzMmlAP/+Wk3EA@public.gmane.org www.hyperxmedia.com 9000 S. 45 W. Sandy, UT 84070 _______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails
> What other problems? We''ve been running 1.4 for a bit, and the slow > redirect issue has been the only one we''ve hit.Well, looking at the list it seems like there''s three: * memory leak * slow redirects * upload bar problems That does seem like a lot. I tried installing 1.4.3 last night and ran into the slow redirect problem (I think that was it) myself. But I didn''t check the mailing list then, it was late, so I just downgraded to 1.3.16 and went back to being fast and happy. I''ll be looking for 1.4.4 before I mess with it again.