Re: Unable to use a GET url-param

2017-04-15 Thread Ajay Garg
Hi Francis. Thanks for your continued help. On Sat, Apr 15, 2017 at 8:50 PM, Francis Daly wrote: > On Sat, Apr 15, 2017 at 02:47:26PM +0530, Ajay Garg wrote: > > Hi there, > > > If there is no forwarded-port in listening state (port 5000 in this case) > > for the upstream-server, the request su

Re: Unable to use a GET url-param

2017-04-15 Thread Francis Daly
On Sat, Apr 15, 2017 at 02:47:26PM +0530, Ajay Garg wrote: Hi there, > If there is no forwarded-port in listening state (port 5000 in this case) > for the upstream-server, the request suitably returns a 502 error. More > importantly, the $arg_upstream_protocol does seem to be parsed properly ::

Re: Unable to use a GET url-param

2017-04-15 Thread Ajay Garg
Hi Francis. I tried the curl method, and I happened to land on an interesting observation. a) If there is no forwarded-port in listening state (port 5000 in this case) for the upstream-server, the request suitably returns a 502 error. More importantly, the $arg_upstream_protocol does seem to be p

Re: auth_basic and satisfy allowing all traffic

2017-04-15 Thread Francis Daly
On Fri, Apr 14, 2017 at 03:26:41PM -0400, daveyfx wrote: Hi there, > I tested the same server configuration as your example, but the testing VM > produced the same results. The satisfy/allow/deny directives allow > bypassing of the basic_auth. Once those entries have been commented out, > auth

Re: Unable to use a GET url-param

2017-04-15 Thread Francis Daly
On Fri, Apr 14, 2017 at 06:42:22PM +0530, Ajay Garg wrote: Hi there, > When I do the following call :: > > https://username:password@1.2.3.4?upstream_protocol=http > 2017/04/14 13:03:51 [error] 16039#16039: *1 invalid URL prefix in ":// > 127.0.0.1:5000", client: 182.69.5.226, server: , request