0

hello I have a drupal multisite setup, with the same modules and paths enabled ( services with REST endpoints). I have some weird behavior:

on site1 I call a URL and get the HTTP status 401 ( which is normal as the test has the not signed request). I am not signing for the purpose of this test.

on site1 it works no matter which browser I use.

now site2, all is setup, and when I test it from chrome/firefox/IE/safari it all works (rest client, rest console etc...), when I say it works it means that I get the 401 back. BUT when I test with safari mobile, or webview I get a 404. same request, same everything.

here are some log entries:

site1.net ###.##.180.109 - - [13/Aug/2013:16:07:01 +0000] "GET /REST/myapp/sendPassReset/oscar4/123 HTTP/1.1" **401** 157 "-" "Mozilla/5.0 (iPhone; CPU iPhone OS 6_1 like Mac OS X) AppleWebKit/536.26 (KHTML, like Gecko) Version/6.0 Mobile/10B141 Safari/8536.25" [smartphones] [0.671] [MISS] [nil]

site2.net ###.##.180.109 - - [13/Aug/2013:16:07:15 +0000] "GET /REST/myapp/sendPassReset/oscar4/123 HTTP/1.1" **404** 5 "-" "Mozilla/5.0 (iPhone; CPU iPhone OS 6_1 like Mac OS X) AppleWebKit/536.26 (KHTML, like Gecko) Version/6.0 Mobile/10B141 Safari/8536.25" [smartphones] [0.219] [-] [nil]

site2.net ###.##.180.109 - - [13/Aug/2013:16:54:36 +0000] "GET /REST/myapp/sendPassReset/oscar4/123 HTTP/1.1" **401** 180 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/28.0.1500.95 Safari/537.36" [-] [0.285] [-] [k-xae2qQMbNdjU4VbQoTpoTh1c2gfnvp-8cz7505Ajk;]

both sites are served by nginx, and PHP-FPM, with the same configuration. same code base.

any idea as to what can be happening?

  • 1
    are there any cache differences? eg the [MISS] vs [-] seem to be the only differences apart from the response between the first two log entries. – Jimajamma Aug 13 '13 at 22:14
  • Finally tracked down the isse to broken/partially updated menu routers in some templates... they were causing the issue. have not been able to close the issue though. – cromestant Aug 13 '13 at 22:53
0

Finally tracked down the isse to broken/partially updated menu in some templates... they were causing the issue. have not been able to close the issue though.

hard reset of memcached fixed the issue... not nice to do in production site on busy day. we had to look through the whole bootstrap to finally see this.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.