:) How about WinXP + VirtualPC + Linux + Pound = SUCCESS?
[...]

That's what we've decided on for our standard.

( VMware ESXi Or Windows + Oracle/Sun Virtualbox) +
 Debian Lenny +
 Pound

The cygwin solution in our testing was actually less stable than going the
Hardware virtualization route :)
Joe

[...]

In our trial period we used Pound (which is why I am on this list).  We
chose to put a BlueCoat in production and point its logs to Splunk.
 
John
 

John H. Folkers
Sr. Network Architect, CCNP, NCTS
UGI Utilities, Inc.
225 Morgantown Road
Reading, PA 19612-3009
610.736.5413


>>> On 6/16/2010 at 12:16 pm, in message
<MailBoxer.1603.1276706988.42.pound(at)apsis.ch>,
<kborn(at)kcp.com>
wrote:
What solutions have you all come across for analyzing the pound log
files 
in its various formats?
I have looked at using AWStats a bit, but was curious if anyone has a 
different setup worth sharing...

Thanks,
Kenton

--
To unsubscribe send an email with subject unsubscribe to
pound(at)apsis.ch.
Please contact roseg(at)apsis.ch for questions.
[...]

----- Puvodní zpráva ----- 
Od: <kborn(at)kcp.com>

[...]

You could use EXTRA SECTIONS to boost your stats (bottom of the configfile):

#PDF download
ExtraSectionName2="Downloads - PDF"
ExtraSectionCodeFilter2="200 304"
ExtraSectionCondition2="URL,(.*((\.pdf)))"
ExtraSectionFirstColumnTitle2="Download"
ExtraSectionFirstColumnValues2="URL,(.*)"
ExtraSectionFirstColumnFormat2="%s"
ExtraSectionStatTypes2=HBL
ExtraSectionAddAverageRow2=0
ExtraSectionAddSumRow2=1
MaxNbOfExtra2=50
MinHitExtra2=1

# ExtraSectionCondition for PDF-Downloads being in a subsubfolder
#ExtraSectionName3="Downloads"
#ExtraSectionCodeFilter3="200 304"
#ExtraSectionCondition3="URL,(^\/uploads\/tx_ffbmatlas\/.*\.pdf)"
#ExtraSectionFirstColumnTitle3="Downloads"
#ExtraSectionFirstColumnValues3="URL,(.*\.pdf)"
#ExtraSectionFirstColumnFormat3="<a href='http://www.beton.org%s' 
target='_blank'>http://www.beton.org%s</a>"
#ExtraSectionStatTypes3=HL
#ExtraSectionAddAverageRow3=1
#ExtraSectionAddSumRow3=1
#MaxNbOfExtra3=2000
#MinHitExtra1=3


#http://www.antezeta.com/blog/awstats
#http://www.antezeta.com/d/AWStats_ExtraSection_Samples.txt
# Assumes default page is "/" and is always referenced as /, not index.html 
etc.
# Assumes default page extension is html.  This will thus exclude directory 
pages which appear as \
# Change html to your page suffix if different, i.e. htm.

ExtraSectionName3="Navigatoins from homepage"
ExtraSectionCodeFilter3="200 304"
ExtraSectionCondition3="REFERER,http:\/\/domain\.ath\.cx\/"
ExtraSectionFirstColumnTitle3="URL"
#ExtraSectionFirstColumnValues3="URL,(.*html$)"
ExtraSectionFirstColumnValues3="URL,(.*$)"
ExtraSectionFirstColumnFormat3="<a href='%s' title='%s' 
target='_blank'>%s</a>"
ExtraSectionStatTypes3=PHBL
ExtraSectionAddAverageRow3=0
ExtraSectionAddSumRow3=1
MaxNbOfExtra3=25
MinHitExtra3=1

# Assumes default page is always linked to as "/".  Some sites need to add 
index.html or default.asp as the case may be.

ExtraSectionName4="Access to homepage from another pages of this website"
ExtraSectionCodeFilter4="200 304"
ExtraSectionCondition4="URL,(^\/$)"
ExtraSectionFirstColumnTitle4="REFERER"
ExtraSectionFirstColumnValues4="REFERER,^http:\/\/domain\.ath\.cx\/(.*)"
ExtraSectionFirstColumnFormat4="<a href='%s' title='%s' 
target='_blank'>%s</a>"
ExtraSectionStatTypes4=PHBL
ExtraSectionAddAverageRow4=0
ExtraSectionAddSumRow4=1
MaxNbOfExtra4=25
MinHitExtra4=1



# Statistiky
ExtraSectionName5="Statistiky"
ExtraSectionCodeFilter5="200 304"
ExtraSectionCondition5="URL,\/_stat_||URL,\/awstats||URL,\/cgi\-bin\/order2\.cgi"
ExtraSectionFirstColumnTitle5="Statistika"
ExtraSectionFirstColumnValues5="QUERY_STRING,config=([^&]+)"
ExtraSectionFirstColumnFormat5="%s"
#        U = Unique visitors
#        V = Visits
#        P = Number of pages
#        H = Number of hits (or mails)
#        B = Bandwith (or total mail size for mail logs)
#        L = Last access date
#        E = Entry pages
#        X = Exit pages
#        C = Web compression (mod_gzip,mod_deflate)
#        M = Average mail size (mail logs)
ExtraSectionStatTypes5=UVPBL
ExtraSectionAddAverageRow5=1
ExtraSectionAddSumRow5=1
MaxNbOfExtra5=30
MinHitExtra5=1


ExtraSectionName6="Pictures"
ExtraSectionCodeFilter6="200 304"
ExtraSectionCondition6="URL,(.*((\.jpg)))||URL,(.*((\.gif)))||URL,(.*((\.png)))"
ExtraSectionFirstColumnTitle6="Obrazek"
ExtraSectionFirstColumnValues6="URL,(.*)"
ExtraSectionFirstColumnFormat6="<a href='%s' target='_blank'><img
src='%s' 
width=111 border=1 align=middle> %s</a>"
ExtraSectionStatTypes6=HLB
ExtraSectionAddAverageRow6=0
ExtraSectionAddSumRow6=1
MaxNbOfExtra6=50
MinHitExtra6=1

# download sekce
ExtraSectionName6="Download sekce"
ExtraSectionCodeFilter6="200 304"
ExtraSectionCondition6="URL,\/_download\/(\.*)"
ExtraSectionFirstColumnTitle6="Download"
ExtraSectionFirstColumnValues6="URL,(.*)"
ExtraSectionFirstColumnFormat6="%s"
ExtraSectionStatTypes6=HBL
ExtraSectionAddAverageRow6=0
ExtraSectionAddSumRow6=1
MaxNbOfExtra6=100
MinHitExtra6=1

(collected from few configfiles, so do not copy/paste)

I've used a combination of AWstats and webalizer in the past, with perl scripts
to split things up.  (I.e. a perl script to create a file per backend which can
then be tracked in separate stats folders)

Joe
[...]

I should have been a little clearer in my post :)

1) Distribute = bad choice of words... pound will be behind three remote
links as "backups". Their configs will be identical. Distribution is the
word I chose for getting replicative setups (pound.cfg) on each remote
link. I didn't mean "distributed-as-in-clustering". Sorry.
 
2) I run vSphere in production and DR now - and there is a separate
project to P2V the remote sites to vSphere... but unfortunately this
pound-project has been bumped up the list - chicken-or-egg scenario. The
Windows-based solution I seek is only temporary, until I get those sites
virtualised. (BTW: I did consider VMware server/workstation at the sites
- but they are just too underpowered - the virtualises project is part
of a hardware refresh).

3) after I sent the post I started to think about CYGWIN. That's my
project for today. 
I eventually noticed (once I discredited all of spurious google results)
that this is well-trodden ground (eventually found someone posted a
W32-Pound-2.3-no-ssl on the 'net).. I'll see how I go with this today
and it might be the perfect temporary solution (yes, I do understand it
will be slow and sluggish... but better than what I have now!)

Just so you understand where I come from here: we had a 6 hour outage
recently that even a BGP re-route could fix (the outage was in the ISP's
core and took out all the BGP routes - as you can imagine - a big
problem affecting a LOT of companies and a 1 in a million...). it was an
outage that was "never suppose to ever happen" - but it did.
Of course, this all came about whilst I was concurrently doing an
ISO270002 audit... so is now in the register and thus - "urgent" ;) 
That's where I came across pound. With it I can keep provision of
webservices via a few other (different ISP) links.. and mitigate the
risks. 

Thanks to you all for your comments.. my apologies for not being clearer
in my post.

While I'm here:
Does anyone know a trick to get DNS "IN A" records to have a preference
(like "IN MX Preference=10")?
My next hurdle is - now I have pound at each of the sites - I only want
the remote sites involved during an outage.
If I put the IP's in to the A record (or CNAME them) - I will
essentially be doing round-robin or "first reply"..
The only way I can think to do this is to manually manipulate the NS
records, wait for DNS cache to refresh to pick up the new sites.
I'm concerned that the backup links will be slow so I don't want any
client to use them unless the core link is down.
If anyone has an idea to that I would appreciate feedback!



[...]

Thanks for all the comments and suggestions.. just an update that I
managed to build with CYGWIN without too many problems (surprise!)

If anyone wants to know how to build under cygwin in the future:
1. Install CYGWIN
2. Install modules: gcc, openssl-devel, make, gcd, pcre (I think that
was all ;)
3. Decompress the pound-2.5.tgz tar/gz
4. build the platform
(I wanted to have the files in c:\pound\bin and c:\pound\etc thus I used
the "--prefix=c:/pound" - note the slash direction CYGWIN needs the c:
to understand what you are trying to do!)
./configure --prefix=c:/pound --with-ssl --disable-super
make

5. make the directory structure under windows
mkdir /cygdrive/c/pound
mkdir /cygdrive/c/pound/bin
mkdir /cygdrive/c/pound/etc

6. copy the cygwin running environment
cp pound*.exe /cygdrive/c/pound/bin
cp /usr/bin/cygcrypto*.dll /cygdrive/c/pound/bin
cp /usr/bin/cygwin1*.dll /cygdrive/c/pound/bin
cp /usr/bin/cyggcc*.dll /cygdrive/c/pound/bin
cp /usr/bin/cygpcre*.dll /cygdrive/c/pound/bin
cp /usr/bin/cygssl*.dll /cygdrive/c/pound/bin

Go back to windows, cd into c:\pound\bin and do a "pound.exe -v"
(remember - any posix folder structures in your pound.cfg (ie. You .pem
file location) will need changing.)

I am testing this as a non-daemon and appears to work fine right now...
very little impact to CPU etc.. certainly looks promising!
Hope this helps someone else in the future.
Mike
-----Original Message-----
From: Michael Weinbergs 
Sent: Thursday, 17 June 2010 9:23 AM
To: pound(at)apsis.ch
Subject: RE: [Pound Mailing List] RE: Ported to Windows?

I should have been a little clearer in my post :)

1) Distribute = bad choice of words... pound will be behind three remote
links as "backups". Their configs will be identical. Distribution is the
word I chose for getting replicative setups (pound.cfg) on each remote
link. I didn't mean "distributed-as-in-clustering". Sorry.
 
2) I run vSphere in production and DR now - and there is a separate
project to P2V the remote sites to vSphere... but unfortunately this
pound-project has been bumped up the list - chicken-or-egg scenario. The
Windows-based solution I seek is only temporary, until I get those sites
virtualised. (BTW: I did consider VMware server/workstation at the sites
- but they are just too underpowered - the virtualises project is part
of a hardware refresh).

3) after I sent the post I started to think about CYGWIN. That's my
project for today. 
I eventually noticed (once I discredited all of spurious google results)
that this is well-trodden ground (eventually found someone posted a
W32-Pound-2.3-no-ssl on the 'net).. I'll see how I go with this today
and it might be the perfect temporary solution (yes, I do understand it
will be slow and sluggish... but better than what I have now!)

Just so you understand where I come from here: we had a 6 hour outage
recently that even a BGP re-route could fix (the outage was in the ISP's
core and took out all the BGP routes - as you can imagine - a big
problem affecting a LOT of companies and a 1 in a million...). it was an
outage that was "never suppose to ever happen" - but it did.
Of course, this all came about whilst I was concurrently doing an
ISO270002 audit... so is now in the register and thus - "urgent" ;) 
That's where I came across pound. With it I can keep provision of
webservices via a few other (different ISP) links.. and mitigate the
risks. 

Thanks to you all for your comments.. my apologies for not being clearer
in my post.

While I'm here:
Does anyone know a trick to get DNS "IN A" records to have a preference
(like "IN MX Preference=10")?
My next hurdle is - now I have pound at each of the sites - I only want
the remote sites involved during an outage.
If I put the IP's in to the A record (or CNAME them) - I will
essentially be doing round-robin or "first reply"..
The only way I can think to do this is to manually manipulate the NS
records, wait for DNS cache to refresh to pick up the new sites.
I'm concerned that the backup links will be slow so I don't want any
client to use them unless the core link is down.
If anyone has an idea to that I would appreciate feedback!



[...]

Windows is not a professional and serious Operative System  :P

Sorry!

Best regards

2010/6/17 Michael Weinbergs <Michael.Weinbergs(at)wridgways.com.au>
[...]

I Think Awstats is a good solution. You can modify the perl code easily and
add new own modules for your use, or change the web design, change the pound
logs to apache format or modify the file config in awstats to configure your
own structure log.

I use it and I suggest because is flexible and  easily modifiable

Regards!

2010/6/16 Jaroslav Lukesh <lukesh(at)seznam.cz>
[...][...][...]

2010/6/16 Joe Gooch <mrwizard(at)k12system.com>:[...]

... anyone tried MinGW + pthreads-win32 (
http://sourceware.org/pthreads-win32/
) ?

Concur... 
but when you have limited resources to solve immediate problems.. 
all you can do is "work the problem"..
at least I have a solution to my immediate problem, now I can start to
deal with improving it.
-----Original Message-----
From: Emilio Campos [mailto:emilio.campos.martin(at)gmail.com] 
Sent: Thursday, 17 June 2010 4:25 PM
To: pound(at)apsis.ch
Subject: Re: [Pound Mailing List] RE: Ported to Windows?

Windows is not a professional and serious Operative System  :P

Sorry!

Best regards

#####################################################################################
This e-mail message has been scanned for Viruses and Content and cleared 
by NetIQ MailMarshal
#####################################################################################

#####################################################################################
The information contained in this email communication may be confidential.
You should only read, disclose, re-transmit, copy, distribute, act in reliance
on or commercialise the information if you are authorised to do so. If you are
not
the intended recipient of this email communication, please notify us
immediately by
email to administrator(at)wridgways.com.au or reply by email direct to the
sender and 
then destroy any electronic or paper copy of this message. Any views expressed
in 
this email communication are those of the individual sender, except where the
sender 
specifically states them to be the views of Wridgways The Removalists.  Any
personal 
information in this email must be handled in accordance with the Privacy Act
1988 
(Cth). Wridgways The Removalists does not represent, warrant or guarantee that
the 
integrity of this communication has been maintained nor that the communication
is 
free of errors, virus or interference.
#####################################################################################

Hiya

Does nobody have any experience of using SAN/UC certificates with pound?
[...]

thanks

Clint

I don't (nor do I remember seeing it on the list before); however, your thought
that Pound will hand the cert to the browser and the browser will deal with it
is pretty much correct.  I would guess that it would work fine.

You can test it yourself, just use openssl to create a certificate with the
appropriate attributes, hand it to pound and test it out. 

Joe
[...]

Responses inline
[...]

I didn't think you did... I meant if you try to zip up your c:\pound directory
and put it on another server, one you might not fully control that might have
different cygwin dll versions, without doing a full cygwin install/compile, you
might run into issues.

Seems like you have some type of control over the OS so you're probably ok :)

[...]

Good to hear!
 [...]

It was ok... performance was "ok"... But the thing that got me was Windows'
memory management, and some issues where it would lock for no reason... Maybe
something in cygwin pthreads causes a deadlock; I never really figured it out.
 [...]

I think the best thing you can do is set the TTL of the A record low... like
300 seconds.  (You don't have to adjust the SOA)
Cons:
Every nameserver will re-ask for your A record every 5 minutes.

Pros:
Every nameserver will re-ask for your A record every 5 minutes.

Then, at least, you can change the record and it'll take at most 5min to
failover.

Or you could get tricky, if your name server lets you do it, and give different
responses based on source ips... I used the TTL method above to deal with our
multi-ISP non-portable IPs situation.  It was manual failover (I had to change
the A) but it was better than nothing.

Joe

Unfortunately it's a convenient operating system and sometimes you don't have
control over the original OS choices :)

Joe
[...]

Hi all

On Thu, 2010-06-17 at 13:32 +0000, Joe Gooch wrote:[...]

And your guess was correct :)  

By the way, if anybody wants to know how to go about creating a
subjectAltname (SAN/UCC) CSR, this link is very helpful:

http://langui.sh/2009/02/27/creating-a-subjectaltname-sanucc-csr/

thanks Joe

Clint

>I think the best thing you can do is set the TTL of the A record low...
like 300 seconds.  (You don't have to adjust the SOA)[...]
to failover.[...]
different responses based on source ips... I used the TTL method above
to deal >with our multi-ISP non-portable IPs situation.  It was manual
failover (I had to change the A) but it was better than nothing.\[...]

I was hoping for a "simple" (aka. Like the MX records) trick.. but I
suspect that this is just not possible.
It appears the only way to implement this is with either an "manual NS
change" (ie low TTLs) or by putting in faster backend links...
For now, I will head down the manual path - but I can't be alone in this
config.

Short of outsourcing my DNS to a "failover" company that monitors my
links and sets the NS for me (although that means delegating my domain
to them - which is a risk (to me) by itself ;))
Alternatively, I suppose I could set a html redirect on the "slow link"
pound host (ie. Pound becomes active when the links go down?! 
As my link speed at the "backup sites" is only slow between pound and
iis - not the internet to the pound. Until I put webserver clones at the
backup sites (see my vmware note pervious).. I guess I will have to
compromise something! 
I appreciate your comments, and thank you for your ideas.
#####################################################################################
This e-mail message has been scanned for Viruses and Content and cleared 
by NetIQ MailMarshal
#####################################################################################

#####################################################################################
The information contained in this email communication may be confidential.
You should only read, disclose, re-transmit, copy, distribute, act in reliance
on or commercialise the information if you are authorised to do so. If you are
not
the intended recipient of this email communication, please notify us
immediately by
email to administrator(at)wridgways.com.au or reply by email direct to the
sender and 
then destroy any electronic or paper copy of this message. Any views expressed
in 
this email communication are those of the individual sender, except where the
sender 
specifically states them to be the views of Wridgways The Removalists.  Any
personal 
information in this email must be handled in accordance with the Privacy Act
1988 
(Cth). Wridgways The Removalists does not represent, warrant or guarantee that
the 
integrity of this communication has been maintained nor that the communication
is 
free of errors, virus or interference.
#####################################################################################

On 04/06/10 18:42, Will Tatam wrote:[...]

Nobody got any ideas why the content would be garbled ?

Will

Hi,

On Jun 26, 2010, at 2:13 PM, Adrian Padilla wrote:
[...]

That depends. If the underlying system is configured the same (same software)
it makes sense to run all sites on all servers, and allow a carefully-designed
Pound setup to spread the load across them.

However if the websites are very different (i.e. one php, one perl and one
ruby-driven), it might be better to keep them on separate systems.

/Eirik

basically all my sites are pretty much business sites and basic php 
websites, and some shopping carts,

here is what i have,

i have all three ubuntu servers running apache 2.0

all running php, perl, pretty much LAMP servers,

would i just duplicate all the same data across all the servers, and 
have pound deligate what servers is being used



Eirik ?verby wrote:[...][...][...]

On 6/26/2010 10:29 AM, Adrian Padilla wrote:[...]

Eirik has a point that sometimes it is better to separate your 
applications.  I am in the camp that unless you have some specific 
reason to split things up, reliability & redundancy trump most other 
goals, so I say use all 3 servers and split things up with pound.

Assuming you don't have any sort of shared storage (i.e. NFS, etc), you 
would need to copy all the application code across the servers.  You do 
need a method to share session data (maybe a DB or something) if you 
intend to allow sessions to migrate between backends.

Basically if you're going to use 3 servers in a round-robin format, you 
need to configure the servers so that at any moment any one can serve 
any request.  Every server must have access to the sessions and any 
other code or data the other servers have.  The advantage here over 
using sticky-sessions is that problems with one backend become obvious 
quickly.  With sticky sessions, if you want to figure out why a certain 
backend is doing something, you first have to either setup pound to 
isolate the backend or figure out some other way of hitting it directly. 
  With round-robin, you just hit reload in your browser a few times.

Regards,[...]

gotcha,

so on the config i have a question on the setup, this is what i think i 
should have

# Main listening ports
               ListenHTTP
                   Address 192.168.3.120
                   Port    80
               End
               Service
                   BackEnd
                       Address 192.168.3.120
                       Port    8080
                       Priority 5
                   End
                       Address 192.168.3.118
                       Port    80
                       Priority 4
                   End
                      Address 192.168.3.119
			Port 80
			Priority 3
		   End
		      Address 192.168.3.102
		        Port 80
			Priority 2
		   End
               End

i have pound on a machine that is also a webserver, so that way i can utilize
apache on that same machine that is why i have one of those set to port 8080,
will this work




Dave Steinberg wrote:[...][...][...]

On 6/26/2010 10:58 AM, Adrian Padilla wrote:[...]

Looks good to me.

Regards,[...]

On Jun 26, 2010, at 4:58 PM, Adrian Padilla wrote:
[...]

Looks like you're using one of the servers as your front-end. That makes me
wonder - what's the point having Pound running on only a single machine? It'll
be a single point of failure, regardless of how many webservers you have.

However, if capacity is your concern, and you can run all the websites on all
the servers, why not dedicate one as Pound front-end and use the other two for
the applications? Chances are two servers will provide all the performance and
redundancy you'll need.

Also, having only two back-ends makes it easier to replicate data between them
if necessary. You could even run a local mysqld on each, which replicates to
the other - but two-way replication only works in pairs, so with three servers
you'd have to run a mysql cluster - which is a whole different story.

/Eirik
[...][...]
>>> basically all my sites are pretty much business sites and basic
php
>>> websites, and some shopping carts,
>>> 
>>> here is what i have,
>>> 
>>> i have all three ubuntu servers running apache 2.0
>>> 
>>> all running php, perl, pretty much LAMP servers,
>>> 
>>> would i just duplicate all the same data across all the servers,
and
>>> have pound deligate what servers is being used[...][...]

Eirik,

what you said made me think so i deployed another box, and i have 
dedicated it to just purely Pound. Now i have also deployed a dedicated 
mysql Database that i am going to configure all the webservers to use 
for a dedicated DB server,

now my question,

in this scenario

server 1
server 2
server 3

when pound decides that it wants to send requests to server 3 will it 
keep say in a case where a customer logs onto a shopping cart and 
decides to purchase a product, will it keep it on server 3 until the 
session is over or will it flip flop from server 1, 2 or 3?

Eirik ?verby wrote:[...][...][...][...]
>>> On 6/26/2010 10:29 AM, Adrian Padilla wrote:
>>>       
>>>> basically all my sites are pretty much business sites and
basic php
>>>> websites, and some shopping carts,
>>>>
>>>> here is what i have,
>>>>
>>>> i have all three ubuntu servers running apache 2.0
>>>>
>>>> all running php, perl, pretty much LAMP servers,
>>>>
>>>> would i just duplicate all the same data across all the
servers, and
>>>> have pound deligate what servers is being used
>>>>         
>>> Eirik has a point that sometimes it is better to separate your
applications.  I am in the camp that unless you have some specific reason to
split things up, reliability & redundancy trump most other goals, so I say
use all 3 servers and split things up with pound.
>>>
>>> Assuming you don't have any sort of shared storage (i.e. NFS,
etc), you would need to copy all the application code across the servers.  You
do need a method to share session data (maybe a DB or something) if you intend
to allow sessions to migrate between backends.
>>>
>>> Basically if you're going to use 3 servers in a round-robin
format, you need to configure the servers so that at any moment any one can
serve any request.  Every server must have access to the sessions and any other
code or data the other servers have.  The advantage here over using
sticky-sessions is that problems with one backend become obvious quickly.  With
sticky sessions, if you want to figure out why a certain backend is doing
something, you first have to either setup pound to isolate the backend or
figure out some other way of hitting it directly.  With round-robin, you just
hit reload in your browser a few times.
>>>
>>> Regards,
>>>       [...][...]

Hi,

On Jun 27, 2010, at 1:42 AM, Adrian Padilla wrote:
[...]

Sounds reasonable.

[...]

That depends how you configure pound, I think. 'man pound' should help you,
check session management. It supports various kinds of session handling.

/Eirik

[...][...]
>>> gotcha,
>>> 
>>> so on the config i have a question on the setup, this is what i
think i should have
>>>    [...]
>>> # Main listening ports
>>>             ListenHTTP
>>>                 Address 192.168.3.120
>>>                 Port    80
>>>             End
>>>             Service
>>>                 BackEnd
>>>                     Address 192.168.3.120
>>>                     Port    8080
>>>                     Priority 5
>>>                 End
>>>                     Address 192.168.3.118
>>>                     Port    80
>>>                     Priority 4
>>>                 End
>>>                    Address 192.168.3.119
>>> 			Port 80
>>> 			Priority 3
>>> 		   End
>>> 		      Address 192.168.3.102
>>> 		        Port 80
>>> 			Priority 2
>>> 		   End
>>>             End
>>> 
>>> i have pound on a machine that is also a webserver, so that way i
can utilize apache on that same machine that is why i have one of those set to
port 8080, will this work
>>> 
>>> 
>>> 
>>> 
>>> Dave Steinberg wrote:
>>>    
>>>> On 6/26/2010 10:29 AM, Adrian Padilla wrote:
>>>>      
>>>>> basically all my sites are pretty much business sites and
basic php
>>>>> websites, and some shopping carts,
>>>>> 
>>>>> here is what i have,
>>>>> 
>>>>> i have all three ubuntu servers running apache 2.0
>>>>> 
>>>>> all running php, perl, pretty much LAMP servers,
>>>>> 
>>>>> would i just duplicate all the same data across all the
servers, and
>>>>> have pound deligate what servers is being used
>>>>>        
>>>> Eirik has a point that sometimes it is better to separate your
applications.  I am in the camp that unless you have some specific reason to
split things up, reliability & redundancy trump most other goals, so I say
use all 3 servers and split things up with pound.
>>>> 
>>>> Assuming you don't have any sort of shared storage (i.e. NFS,
etc), you would need to copy all the application code across the servers.  You
do need a method to share session data (maybe a DB or something) if you intend
to allow sessions to migrate between backends.
>>>> 
>>>> Basically if you're going to use 3 servers in a round-robin
format, you need to configure the servers so that at any moment any one can
serve any request.  Every server must have access to the sessions and any other
code or data the other servers have.  The advantage here over using
sticky-sessions is that problems with one backend become obvious quickly.  With
sticky sessions, if you want to figure out why a certain backend is doing
something, you first have to either setup pound to isolate the backend or
figure out some other way of hitting it directly.  With round-robin, you just
hit reload in your browser a few times.
>>>> 
>>>> Regards,
>>>>      
>>> --
>>> To unsubscribe send an email with subject unsubscribe to
pound(at)apsis.ch.
>>> Please contact roseg(at)apsis.ch for questions.
>>>    [...][...]