Hi,

Well, there is a directive called "dontlog-normal" to log only errors
You should give it a try.

cheers

On Thu, Apr 5, 2012 at 6:30 PM, Guy Knights <[email protected]> wrote:
> Thanks Baptiste. I turned on HTTP logging and I'm now getting the actual
> HTTP request path in the log output. It looks like it's just logging all
> calls from the Nginx reverse proxy we have on the box that's doing SSL
> termination. Is there a way to disable this output to the log? Obviously I
> still want errors to show up but there's no need to log every request that's
> sent from Nginx to Haproxy.
>
> Thanks,
> Guy
>
>
> On Wed, Apr 4, 2012 at 9:45 PM, Baptiste <[email protected]> wrote:
>>
>> Hi,
>>
>> You're logging using the tcp log format (option tcplog from your
>> defaults).
>> You may turn on "option httplog" in your http listen section to get
>> more information about this request.
>>
>> It may be some monitoring systems.
>>
>> Regards
>>
>>
>> On Thu, Apr 5, 2012 at 4:07 AM, Guy Knights <[email protected]>
>> wrote:
>> > I happened to check the haproxy logs today on our load balancer and I'm
>> > seeing a ton of weird logs showing
>> >
>> > localhost.localdomain haproxy[18309]: 127.0.0.1:41386
>> > [04/Apr/2012:21:03:31.942] example.com example.com/example-server 0/0/25
>> > 238
>> > -- 0/0/0/0/0 0/0
>> >
>> > The port number for 127.0.0.1 seems to change randomly within a pretty
>> > small
>> > range in every time the above error is logged.
>> >
>> > Here's my haproxy.cfg file:
>> >
>> > global
>> >         log 127.0.0.1 local0 info
>> >         maxconn 20000
>> >         user haproxy
>> >         stats socket /tmp/haproxy.socket uid haproxy mode 770 level
>> > admin
>> >         daemon
>> >
>> > defaults
>> >         log global
>> >         mode tcp
>> >         option tcplog
>> >         option dontlognull
>> >         retries 3
>> >         option redispatch
>> >         maxconn 15000
>> >         timeout connect 5s
>> >         timeout client 120s
>> >         timeout server 120s
>> >
>> > listen example.com 127.0.0.1:81
>> >         mode http
>> >         balance roundrobin
>> >         option httpclose
>> >         option forwardfor
>> >         server example-server 10.20.20.1:80 check
>> >
>> > listen private_monitoring :8100
>> >     mode http
>> >     option httplog
>> >     stats enable
>> >     stats uri /stats
>> >     stats refresh 5s
>> >     stats auth xxxxx:xxxxxxxxx
>> >
>> > Does anyone have any idea what the problem is here?
>> >
>> > Thanks,
>> > Guy
>
>
>
>
> --
> Guy Knights
> Systems Administrator
> Eastside Games
> www.eastsidegamestudio.com
> [email protected]

Reply via email to