Perl 6 - the future is here, just unevenly distributed

IRC log for #openam, 2017-09-08

| Channels | #openam index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:55 ilbot3 joined #openam
04:21 aldaris joined #openam
06:02 aldaris joined #openam
09:24 FireBurn|Work joined #openam
09:25 FireBurn|Work Hi aldaris, thanks for the mention, does spareyc hang around irc?
09:26 aldaris nope
09:27 FireBurn|Work I'm certain it's the openam apache module, if I login and get the blank page comment out openam in the apache configs and refresh it loads perfectly everytime
09:27 FireBurn|Work Using the cookies it got from OpenAM the first time
09:29 FireBurn|Work It also works with the old 3.3.4 module except it applies over both virtual hosts, so the openam ones goes into a loop
10:09 Shra joined #openam
10:12 FireBurn|Work I'm wondering if this: ap_hook_translate_name(amagent_preserve_url, NULL, NULL, APR_HOOK_FIRST - 2) in the openam module is clashing with proxy/mod_proxy.c:    ap_hook_translate_name(proxy_trans, aszSucc, NULL, APR_HOOK_FIRST);
11:12 aldaris joined #openam
11:33 FireBurn|Work Na, not that, I've tried all combos of hooks before and after mod_proxy.c and mod_proxy_http.c
12:19 FireBurn|Work So when it's it a broken state, we can retrieve headers just fine, it's the body that's not coming down the line
12:47 FireBurn|Work Hmm so I commeded out the case where F_SKIPBODY is used in the HTTP_PARSER
12:48 FireBurn|Work I'm wondering if it's being set but then not being reset for the GET or something weird like that
13:09 FireBurn|Work Nope wasn't that wither
13:09 FireBurn|Work *either
14:16 FireBurn|Work So I can definately see GET requests being downgraded to HEAD requests by the Agent
14:16 FireBurn|Work I'm not a "C" guy but I'm slowly going through the bit of code that does this to see where it's going wrong
14:16 FireBurn|Work It doesn't help there's a few variables with very similar names
14:38 FireBurn|Work 2017-09-08 15:38:13.595 +0100   DEBUG [0x7f2502e46700:29476][source/apache/agent.c:579] get_method_num(): method GET (HEAD, 0)
14:39 FireBurn|Work 2017-09-08 15:38:13.595 +0100   DEBUG [0x7f2502e46700:29476][source/apache/agent.c:588] get_method_num(): number corresponds to HEAD method
14:39 FireBurn|Work 2017-09-08 15:38:13.595 +0100   DEBUG [0x7f2502e46700:29476][source/apache/agent.c:607] get_method_num(): set method to HEAD
14:39 FireBurn|Work 2017-09-08 15:38:13.595 +0100   DEBUG [0x7f2502e46700:29476][source/apache/agent.c:609] get_method_num(): This is method_num before return: 3
14:51 FireBurn|Work and when it's working:
14:51 FireBurn|Work 2017-09-08 15:50:16.939 +0100   DEBUG [0x7f24b37fe700:30733][source/apache/agent.c:579] get_method_num(): method GET (GET, 0)
14:51 FireBurn|Work 2017-09-08 15:50:16.939 +0100   DEBUG [0x7f24b37fe700:30733][source/apache/agent.c:588] get_method_num(): number corresponds to GET method
14:51 FireBurn|Work 2017-09-08 15:50:16.939 +0100   DEBUG [0x7f24b37fe700:30733][source/apache/agent.c:609] get_method_num(): This is method_num before return: 1
15:30 FireBurn|Work aldaris: That does look like the issue
15:39 FireBurn|Work Mind if I get the offical fix rather than that hacky mess I've put together
19:27 aldaris joined #openam
19:43 aldaris joined #openam
20:34 aldaris joined #openam
22:03 aldaris joined #openam
22:20 aldaris joined #openam

| Channels | #openam index | Today | | Search | Google Search | Plain-Text | summary