Hi,
I just finished to setup my production server. I'm using pound to provide HTTPS 
to the Varnish backend. 
Now that I put the sites on production, I detect a lot of errors on the pound 
logs file:
-------------------
Apr 11 11:32:00 kwak pound: (7f32c626f700) error read from 85.201.217.135: 
Connection timed out
Apr 11 11:32:01 kwak pound: (7f32c5702700) error copy server cont: Connection 
timed out
Apr 11 11:32:02 kwak pound: (7f32c55bd700) error flush headers to 89.96.213.46: 
Connection reset by peer
Apr 11 11:32:19 kwak pound: (7f32c6f21700) error read from 79.21.0.21: 
Connection timed out
Apr 11 11:32:24 kwak pound: (7f32c5a90700) error read from 78.23.240.176: 
Connection timed out
Apr 11 11:32:29 kwak pound: (7f32c6e9f700) error copy server cont: Broken pipe
Apr 11 11:32:31 kwak pound: (7f32c64b8700) error copy server cont: Connection 
reset by peer
Apr 11 11:32:31 kwak pound: (7f32c7812700) error copy server cont: Connection 
reset by peer
Apr 11 11:32:31 kwak pound: (7f32c5437700) error copy server cont: Connection 
reset by peer
Apr 11 11:32:39 kwak pound: (7f32c516c700) error copy server cont: Connection 
reset by peer
Apr 11 11:32:42 kwak pound: (7f32c9bf9700) error read from 2.180.36.132: 
Connection reset by peer
Apr 11 11:32:50 kwak pound: (7f32c6f21700) error read from 91.178.141.219: 
Connection timed out
-------------------

I can't reproduce the problem by my self. Can you tell me if I have something 
wrong in my configuration. Here after the software versions and a piece of the 
pound configuration file.


Debian's version: 6 
Varnish's version:  3

Pound's version: 2.5

## Logging: (goes to syslog by default)
##      0       no logging
##      1       normal
##      2       extended
##      3       Apache-style (common log format)
LogLevel        1

## check backend every X secs:
Alive           30

## use hardware-accelleration card supported by openssl(1):
#SSLEngine      "<hw>"

# poundctl control socket
Control "/var/run/pound/poundctl.socket"

##Increase the timeout from 15 to 60 seconds - some requests are too long on 
the TA backend
#
TimeOut 60
######################################################################
## listen, redirect and ... to:
ListenHTTPS
        Address XXX.XXX.XXX.XXX
        Port    443
        Cert "/etc/ssl/certs/sequentiel/seqfr.pem"

        ## allow PUT and DELETE also (by default only GET, POST and HEAD)?:
        xHTTP           0

        Service
                BackEnd
                        Address 127.0.0.1
                        Port    8080
                End
        End
End




Thanks in advance for your support.
Regards,
Frederic

Reply via email to