Have you seen http://www.swa.hpi.uni-potsdam.de/seaside/tutorial lately? Every page link goes through about 8 URL changes. It's amusing the first time. Interesting the second time. Quite annoying the fifth or sixth time. What would cause this, so I can prevent it happening, and so that the HPI can also fix it? -- Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 <[hidden email]> <URL:http://www.stonehenge.com/merlyn/> Perl/Unix/security consulting, Technical writing, Comedy, etc. etc. See PerlTraining.Stonehenge.com for onsite and open-enrollment Perl training! _______________________________________________ seaside mailing list [hidden email] http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside |
On Dec 25, 2007 11:35 PM, Randal L. Schwartz <[hidden email]> wrote:
> Have you seen http://www.swa.hpi.uni-potsdam.de/seaside/tutorial lately? > Every page link goes through about 8 URL changes. It is pretty bad, isn't it? I made a program to show what's going on when I followed a link from that page, and this is what it said, with apologies for readability: Starting with http://www.hpi.uni-potsdam.de/swa/seaside/tutorial?1&_k=lRXaakpk&_s=ekNqiioEEPonTpbQ ------------------------------- ### 302 Found -> server: Apache -> location: http://www.hpi.uni-potsdam.de/hirschfeld/seaside/tutorial?1&_k=lRXaakpk&_s=ekNqiioEEPonTpbQ ------------------------------- ### 302 Found -> server: Apache -> location: http://www.swa.hpi.uni-potsdam.de/seaside/tutorial?1&_k=lRXaakpk&_s=ekNqiioEEPonTpbQ ------------------------------- ### 302 Moved Temporarily -> server: Comanche/6.2 (unix) -> location: http://www.hpi.uni-potsdam.de/swa/seaside/tutorial?_k=alidjmdR&_s=ekNqiioEEPonTpbQ ------------------------------- ### 302 Found -> server: Apache -> location: http://www.hpi.uni-potsdam.de/hirschfeld/seaside/tutorial?_k=alidjmdR&_s=ekNqiioEEPonTpbQ ------------------------------- ### 302 Found -> server: Apache -> location: http://www.swa.hpi.uni-potsdam.de/seaside/tutorial?_k=alidjmdR&_s=ekNqiioEEPonTpbQ ------------------------------- ### 200 OK -> server: Comanche/6.2 (unix) It took six requests to retrieve that page. Four of those URLs use the server name www.hpi.uni-potsdam.de, but two use www.swa.hpi.uni-potsdam.de. Let's call them Happy and Sappy for short. Happy is the Apache server; Sappy is Comanche, presumably Seaside. Sappy served the page that gave the initial link. Happy seemingly redirects /swa uris to /hirschfeld, but then re-redirects /hirschfeld ones back to Sappy. Sappy does a redirect of its own (notice that the _k parameter changes, but the _s session parameter is preserved), but since it goes back to /swa on Happy, it takes Apache two more tries to get back to Comanche. > What would cause this, so I can prevent it happening, and so that the HPI can > also fix it? My guess: Somebody tried moving this server to a new hostname, from Happy to Sappy, but wasn't able to tell Seaside about it. They tried asking Seaside (or Comanche?) to redirect (??) but that just changed the _k parameter (and maybe some other internal state?). Then they discovered that adding a few redirections to the Apache configuration on Happy got things to work, for a suitably low value of "work", and that's where things stand today. Conceivably, if it was somebody who didn't have the Seaside config URL and password, this may have been the best solution they could manage. Another piece of the puzzle may be that somebody, possibly the same person or persons, wanted to make it possible to bookmark the tutorial's URLs. (The Apache redirects may be an attempt to keep bookmarked URLs from failing, in fact.) The tutorial in question has a section on "Static Url" in Chapter 10 that seems partially relevant. Happy's admin should be able to do in one redirect what the Apache server is now doing in two, without even breaking old bookmarks. But the Sappy content, coming presumably from Seaside, is the real culprit; it seems to be misconfigured such that it thinks it is still on the /swa path at Happy's hostname. If that alone were fixed, I think it would take normal clicks down to a single redirection; that last redirection can either be fixed, justified, or tolerated, depending upon the phase of the moon, I suppose. Of course, that's just my guess. I hope that someone else can read more clearly the story of the evidence, or can confirm that my interpretation is plausible. Cheers! --Tom Phoenix _______________________________________________ seaside mailing list [hidden email] http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside |
Tom,
Just a hint, you can use curl to trace the redirects in the future, $ curl -IL "http://www.hpi.uni-potsdam.de/swa/seaside/tutorial?2&_k=qCjOXJzQ&_s=WPd jPhAivkXJfbuj" HTTP/1.1 302 Found Date: Thu, 27 Dec 2007 19:17:49 GMT Server: Apache Location: http://www.hpi.uni-potsdam.de/hirschfeld/seaside/tutorial?2&_k=qCjOXJzQ& _s=WPdjPhAivkXJfbuj Content-Type: text/html; charset=iso-8859-1 HTTP/1.1 302 Found Date: Thu, 27 Dec 2007 19:17:49 GMT Server: Apache Location: http://www.swa.hpi.uni-potsdam.de/seaside/tutorial?2&_k=qCjOXJzQ&_s=WPdj PhAivkXJfbuj Content-Type: text/html; charset=iso-8859-1 HTTP/1.1 302 Moved Temporarily Date: Thu, 27 Dec 2007 19:12:04 GMT Server: Comanche/6.2 (unix) Cache-Control: no-cache Location: http://www.hpi.uni-potsdam.de/swa/seaside/tutorial?_k=zBhvczNC&_s=WPdjPh AivkXJfbuj Content-type: text/html;charset=utf-8 Via: 1.1 swa.hpi.uni-potsdam.de HTTP/1.1 302 Found Date: Thu, 27 Dec 2007 19:17:50 GMT Server: Apache Location: http://www.hpi.uni-potsdam.de/hirschfeld/seaside/tutorial?_k=zBhvczNC&_s =WPdjPhAivkXJfbuj Content-Type: text/html; charset=iso-8859-1 HTTP/1.1 302 Found Date: Thu, 27 Dec 2007 19:17:50 GMT Server: Apache Location: http://www.swa.hpi.uni-potsdam.de/seaside/tutorial?_k=zBhvczNC&_s=WPdjPh AivkXJfbuj Content-Type: text/html; charset=iso-8859-1 HTTP/1.1 200 OK Date: Thu, 27 Dec 2007 19:12:05 GMT Server: Comanche/6.2 (unix) Cache-Control: no-cache Content-type: text/html;charset=utf-8 Via: 1.1 swa.hpi.uni-potsdam.de -Boris -- +1.604.689.0322 DeepCove Labs Ltd. 4th floor 595 Howe Street Vancouver, Canada V6C 2T5 http://tinyurl.com/r7uw4 [hidden email] CONFIDENTIALITY NOTICE This email is intended only for the persons named in the message header. Unless otherwise indicated, it contains information that is private and confidential. If you have received it in error, please notify the sender and delete the entire message including any attachments. Thank you. > -----Original Message----- > From: [hidden email] [mailto:seaside- > [hidden email]] On Behalf Of Tom Phoenix > Sent: Thursday, December 27, 2007 11:13 AM > To: Seaside - general discussion > Subject: Re: [Seaside] the URL dance forhttp://www.swa.hpi.uni- > potsdam.de/seaside/tutorial > > On Dec 25, 2007 11:35 PM, Randal L. Schwartz <[hidden email]> > wrote: > > > Have you seen http://www.swa.hpi.uni-potsdam.de/seaside/tutorial > > Every page link goes through about 8 URL changes. > > It is pretty bad, isn't it? I made a program to show what's going on > when I followed a link from that page, and this is what it said, with > apologies for readability: > > Starting with > http://www.hpi.uni- > potsdam.de/swa/seaside/tutorial?1&_k=lRXaakpk&_s=ekNqiioEEPonTpbQ > ------------------------------- > ### 302 Found > -> server: Apache > -> location: > http://www.hpi.uni- > > ------------------------------- > ### 302 Found > -> server: Apache > -> location: > http://www.swa.hpi.uni- > potsdam.de/seaside/tutorial?1&_k=lRXaakpk&_s=ekNqiioEEPonTpbQ > ------------------------------- > ### 302 Moved Temporarily > -> server: Comanche/6.2 (unix) > -> location: > http://www.hpi.uni- > potsdam.de/swa/seaside/tutorial?_k=alidjmdR&_s=ekNqiioEEPonTpbQ > ------------------------------- > ### 302 Found > -> server: Apache > -> location: > http://www.hpi.uni- > potsdam.de/hirschfeld/seaside/tutorial?_k=alidjmdR&_s=ekNqiioEEPonTpbQ > ------------------------------- > ### 302 Found > -> server: Apache > -> location: > http://www.swa.hpi.uni- > potsdam.de/seaside/tutorial?_k=alidjmdR&_s=ekNqiioEEPonTpbQ > ------------------------------- > ### 200 OK > -> server: Comanche/6.2 (unix) > > It took six requests to retrieve that page. Four of those URLs use the > server name www.hpi.uni-potsdam.de, but two use > www.swa.hpi.uni-potsdam.de. Let's call them Happy and Sappy for short. > Happy is the Apache server; Sappy is Comanche, presumably Seaside. > > Sappy served the page that gave the initial link. Happy seemingly > redirects /swa uris to /hirschfeld, but then re-redirects /hirschfeld > ones back to Sappy. Sappy does a redirect of its own (notice that the > _k parameter changes, but the _s session parameter is preserved), but > since it goes back to /swa on Happy, it takes Apache two more tries to > get back to Comanche. > > > What would cause this, so I can prevent it happening, and so that > HPI can > > also fix it? > > My guess: Somebody tried moving this server to a new hostname, from > Happy to Sappy, but wasn't able to tell Seaside about it. They tried > asking Seaside (or Comanche?) to redirect (??) but that just changed > the _k parameter (and maybe some other internal state?). Then they > discovered that adding a few redirections to the Apache configuration > on Happy got things to work, for a suitably low value of "work", and > that's where things stand today. Conceivably, if it was somebody who > didn't have the Seaside config URL and password, this may have been > the best solution they could manage. > > Another piece of the puzzle may be that somebody, possibly the same > person or persons, wanted to make it possible to bookmark the > tutorial's URLs. (The Apache redirects may be an attempt to keep > bookmarked URLs from failing, in fact.) The tutorial in question has a > section on "Static Url" in Chapter 10 that seems partially relevant. > > Happy's admin should be able to do in one redirect what the Apache > server is now doing in two, without even breaking old bookmarks. But > the Sappy content, coming presumably from Seaside, is the real > culprit; it seems to be misconfigured such that it thinks it is still > on the /swa path at Happy's hostname. If that alone were fixed, I > think it would take normal clicks down to a single redirection; that > last redirection can either be fixed, justified, or tolerated, > depending upon the phase of the moon, I suppose. > > Of course, that's just my guess. I hope that someone else can read > more clearly the story of the evidence, or can confirm that my > interpretation is plausible. > > Cheers! > > --Tom Phoenix > _______________________________________________ > seaside mailing list > [hidden email] > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside seaside mailing list [hidden email] http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside |
Free forum by Nabble | Edit this page |