From jfrias at gmail.com  Tue Sep  1 02:58:50 2009
From: jfrias at gmail.com (Javier Frias)
Date: Mon, 31 Aug 2009 22:58:50 -0400
Subject: Saint mode
Message-ID: <22964b960908311958r211f8ac9g1b6eee46ed92d158@mail.gmail.com>

Can I achieve the same thing as "saint mode" on varnish 2.0.4? If not, when
will there be a release with this feature?

Specifically, we are looking to not store bad documents unless we have to,
iow, not replace a known good document ( status 200 or 404 ), with a bad
document ( status 503 ), if it would otherwise would be served from grace.


Saint Mode Change Set:

http://varnish.projects.linpro.no/changeset/4208

if (beresp.status == 500) {

 set beresp.saintmode = 20s;
restart;

}

-Javier
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.varnish-cache.org/lists/pipermail/vinyl-dev/attachments/20090831/5b7f4cbd/attachment.html>

From phk at phk.freebsd.dk  Tue Sep  1 08:58:21 2009
From: phk at phk.freebsd.dk (Poul-Henning Kamp)
Date: Tue, 01 Sep 2009 08:58:21 +0000
Subject: Saint mode 
In-Reply-To: Your message of "Mon, 31 Aug 2009 22:58:50 -0400."
	<22964b960908311958r211f8ac9g1b6eee46ed92d158@mail.gmail.com> 
Message-ID: <47314.1251795501@critter.freebsd.dk>

In message <22964b960908311958r211f8ac9g1b6eee46ed92d158 at mail.gmail.com>, Javie
r Frias writes:

>Specifically, we are looking to not store bad documents unless we have to,
>iow, not replace a known good document ( status 200 or 404 ), with a bad
>document ( status 503 ), if it would otherwise would be served from grace.

503's are never cached to my knowledge.

the rest can be obtained by:

	sub vcl_fetch {
		if (beresp.status != 200 && beresp.status != 404) {
			set beresp.ttl = 0;
			return (pass);
		}
	}

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.


From stockrt at gmail.com  Tue Sep  1 14:44:38 2009
From: stockrt at gmail.com (=?ISO-8859-1?Q?Rog=E9rio_Schneider?=)
Date: Tue, 1 Sep 2009 11:44:38 -0300
Subject: Saint mode
In-Reply-To: <47314.1251795501@critter.freebsd.dk>
References: <22964b960908311958r211f8ac9g1b6eee46ed92d158@mail.gmail.com>
	<47314.1251795501@critter.freebsd.dk>
Message-ID: <100657c90909010744r208ecec1t6eee4ee0cc407cc6@mail.gmail.com>

I believe this functionality is great and I think in start using it.

We have this need to continue serving correct (!= 5xx) documents if they are
cached and the backend responses are in that range of failure codes. This
patch appears to solve this need.

I also believe it is correlated to
http://varnish.projects.linpro.no/ticket/369

Best regards,

> --
Rog?rio Schneider

MSN: stockrt at hotmail.com
GTalk: stockrt at gmail.com
TerraVoip: stockrt
Skype: stockrt
http://stockrt.github.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.varnish-cache.org/lists/pipermail/vinyl-dev/attachments/20090901/1ca8129c/attachment.html>

From jfrias at gmail.com  Tue Sep  1 17:35:43 2009
From: jfrias at gmail.com (Javier Frias)
Date: Tue, 1 Sep 2009 13:35:43 -0400
Subject: Saint mode
In-Reply-To: <47314.1251795501@critter.freebsd.dk>
References: <22964b960908311958r211f8ac9g1b6eee46ed92d158@mail.gmail.com>
	<47314.1251795501@critter.freebsd.dk>
Message-ID: <22964b960909011035y10159263j1a82751ccb592db6@mail.gmail.com>

On Tue, Sep 1, 2009 at 4:58 AM, Poul-Henning Kamp <phk at phk.freebsd.dk>wrote:

>
>
>
> 503's are never cached to my knowledge.
>
> the rest can be obtained by:
>
>        sub vcl_fetch {
>                if (beresp.status != 200 && beresp.status != 404) {
>                        set beresp.ttl = 0;
>                        return (pass);
>                }
>        }
>
>

So my setup is a bit different, since I'm one of those that want to server
stuff pass the grace period, if my backends are down. I also want to protect
myself against a self inflicted DoS due to errors. So ideally, saint mode is
what i'd like ( ie, dont store this unless you are a better result )

here's the pertinent configs...

sub vcl_recv {
 ....
        # even though an object may be expired, we want to serve it up for
this long to prevent thread pile up
        # we also want to be somewhat aggresive if the backend is
up(healthy)
       if (req.backend.healthy) {
               set req.grace = 5m;
       } else {
               set req.grace = 600m;
       }
...
}

sub vcl_fetch {
       # here, we cache for more time codes that are not server errors
        if ( obj.status <500) {
                set obj.grace = 600m;
                set obj.ttl = 600s;
                set obj.cacheable = true;
        } else {
                set obj.grace = 5s;
                set obj.ttl = 60s;
                set obj.cacheable = true;
        }

}



Before, i use to just have a blanket vcl_fetch setting my grace and ttl,
which is probably the reason why i was caching the 500's



-Javier
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.varnish-cache.org/lists/pipermail/vinyl-dev/attachments/20090901/b0cfe570/attachment.html>

From stockrt at gmail.com  Wed Sep  2 03:05:08 2009
From: stockrt at gmail.com (=?ISO-8859-1?Q?Rog=E9rio_Schneider?=)
Date: Wed, 2 Sep 2009 00:05:08 -0300
Subject: behaviour when client drops the connection?
In-Reply-To: <4A9B2345.3030402@vuw.ac.nz>
References: <4A9B2345.3030402@vuw.ac.nz>
Message-ID: <100657c90909012005x74cfb04dhd08908e17744e955@mail.gmail.com>

Stuart, I think 60 minutes is too much time even for Varnish to wait.
In default configuration Varnish would wait only for, if I remember, 60
seconds and then give up on that request.

This can be tuned with -p params, like "first_byte_timeout". You can read
more here:

http://varnish.projects.linpro.no/browser/trunk/varnish-cache/bin/varnishd/mgt_param.c

Best Regards,

On Sun, Aug 30, 2009 at 10:11 PM, stuart yeates <stuart.yeates at vuw.ac.nz>wrote:

> Hello
>
> I'm looking to use varnish in front of an XSLT/PostgreSQL/Tomcat/Java
> site and have a question about varnish's behaviour when a client drops a
> connection.
>
> The problem is that there are a relatively small number of pages which
> are _very_ slow to generate (>60 minutes) and a large number that are
> relatively fast. Clients tend to abort the slow pages, so (using our
> current reverse proxy) they never make it into the cache.
>
> Ideally I'd like to make varnish continue the request and cache the
> result. Is that possible?
>
> Is there a way to make these slow-to-generate pages have a much longer
> cache-time?
>
> cheers
> stuart
>
> --
> Stuart Yeates
> http://www.nzetc.org/       New Zealand Electronic Text Centre
> http://researcharchive.vuw.ac.nz/     Institutional Repository
> _______________________________________________
> varnish-misc mailing list
> varnish-misc at projects.linpro.no
> http://projects.linpro.no/mailman/listinfo/varnish-misc
>


-- 
Rog?rio Schneider

MSN: stockrt at hotmail.com
GTalk: stockrt at gmail.com
TerraVoip: stockrt
Skype: stockrt
http://stockrt.github.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.varnish-cache.org/lists/pipermail/vinyl-dev/attachments/20090902/945c0b6c/attachment.html>

From rob.halff at gmail.com  Wed Sep  2 07:15:22 2009
From: rob.halff at gmail.com (Rob Halff)
Date: Wed, 2 Sep 2009 09:15:22 +0200
Subject: behaviour when client drops the connection?
In-Reply-To: <100657c90909012005x74cfb04dhd08908e17744e955@mail.gmail.com>
References: <4A9B2345.3030402@vuw.ac.nz>
	<100657c90909012005x74cfb04dhd08908e17744e955@mail.gmail.com>
Message-ID: <cee869b0909020015g6d4ed222nd1296574a65da754@mail.gmail.com>

Stuart, pages taking 1 hour to load is an application design problem.
You should use http to show results not to trigger a one hour request
and let the user/client wait for it.

You could generate a code instantly to confirm the request is being processed.
And let the client check, using this code, whether the request is done
every minute or whatever time interval you choose.

If the request is not based on user parameters, you could use cron to
generate the result page.

Gr.
Rob Halff

2009/9/2 Rog?rio Schneider <stockrt at gmail.com>:
> Stuart, I think 60 minutes is too much time even for Varnish to wait.
> In default configuration Varnish would wait only for, if I remember, 60
> seconds and then give up on that request.
>
> This can be tuned with -p params, like "first_byte_timeout". You can read
> more here:
>
> http://varnish.projects.linpro.no/browser/trunk/varnish-cache/bin/varnishd/mgt_param.c
>
> Best Regards,
>
> On Sun, Aug 30, 2009 at 10:11 PM, stuart yeates <stuart.yeates at vuw.ac.nz>
> wrote:
>>
>> Hello
>>
>> I'm looking to use varnish in front of an XSLT/PostgreSQL/Tomcat/Java
>> site and have a question about varnish's behaviour when a client drops a
>> connection.
>>
>> The problem is that there are a relatively small number of pages which
>> are _very_ slow to generate (>60 minutes) and a large number that are
>> relatively fast. Clients tend to abort the slow pages, so (using our
>> current reverse proxy) they never make it into the cache.
>>
>> Ideally I'd like to make varnish continue the request and cache the
>> result. Is that possible?
>>
>> Is there a way to make these slow-to-generate pages have a much longer
>> cache-time?
>>
>> cheers
>> stuart
>>
>> --
>> Stuart Yeates
>> http://www.nzetc.org/ ? ? ? New Zealand Electronic Text Centre
>> http://researcharchive.vuw.ac.nz/ ? ? Institutional Repository
>> _______________________________________________
>> varnish-misc mailing list
>> varnish-misc at projects.linpro.no
>> http://projects.linpro.no/mailman/listinfo/varnish-misc
>
>
> --
> Rog?rio Schneider
>
> MSN: stockrt at hotmail.com
> GTalk: stockrt at gmail.com
> TerraVoip: stockrt
> Skype: stockrt
> http://stockrt.github.com
>
> _______________________________________________
> varnish-dev mailing list
> varnish-dev at projects.linpro.no
> http://projects.linpro.no/mailman/listinfo/varnish-dev
>
>


From kristian at redpill-linpro.com  Wed Sep  2 12:53:18 2009
From: kristian at redpill-linpro.com (Kristian Lyngstol)
Date: Wed, 2 Sep 2009 14:53:18 +0200
Subject: Saint mode
In-Reply-To: <22964b960908311958r211f8ac9g1b6eee46ed92d158@mail.gmail.com>
References: <22964b960908311958r211f8ac9g1b6eee46ed92d158@mail.gmail.com>
Message-ID: <20090902125318.GA17535@kjeks.linpro.no>

On Mon, Aug 31, 2009 at 10:58:50PM -0400, Javier Frias wrote:
> Can I achieve the same thing as "saint mode" on varnish 2.0.4? If not, when
> will there be a release with this feature?

No, you can't.

I expect it to be in 2.1 which is to be released shortly, we just need to
get persistence working and stable.

-- 
Kristian Lyngst?l
Redpill Linpro AS
Tlf: +47 21544179
Mob: +47 99014497
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 835 bytes
Desc: not available
URL: <https://www.varnish-cache.org/lists/pipermail/vinyl-dev/attachments/20090902/923a5181/attachment.bin>

From Olivier.Nicole at cs.ait.ac.th  Thu Sep  3 03:14:25 2009
From: Olivier.Nicole at cs.ait.ac.th (Olivier Nicole)
Date: Thu, 3 Sep 2009 10:14:25 +0700 (ICT)
Subject: Ban headers in the messages send to the list
Message-ID: <200909030314.n833EPKU009542@banyan.cs.ait.ac.th>

Hi,

It seems that (some of) the messages sent to varnish-dev list have a
duplicated MIME-Version: header. This causes the message to be
quarantined by Amavisd-new. It may affect other users f the list.

Below is an example of the headers.

Best regards,

Olivier

    X-Coding-System: undecided-unix
    Mail-from: From jfrias at gmail.com Thu Sep 03 09:58:15 2009
    Return-Path: <>
    Delivered-To: bad-header-quarantine
    X-Envelope-From: <varnish-dev-bounces at projects.linpro.no>
    X-Envelope-To: <on at cs.ait.ac.th>
    X-Envelope-To-Blocked: 
    X-Quarantine-ID: <DpOv5pOzCgAp>
    X-Amavis-Alert: BAD HEADER SECTION Duplicate header field: "MIME-Version"
    Received: from mail.cs.ait.ac.th ([127.0.0.1])
    	by localhost (mail.cs.ait.ac.th [127.0.0.1]) (amavisd-new, port 10024)
    	with ESMTP id DpOv5pOzCgAp for <on at cs.ait.ac.th>;
    	Tue,  1 Sep 2009 10:46:07 +0700 (ICT)
    X-Greylist: delayed 1283 seconds by postgrey-1.32 at mail.cs.ait.ac.th; Tue, 01 Sep 2009 10:46:06 ICT
    Received: from projects.linpro.no (projects.linpro.no [87.238.49.148])
    	(using TLSv1 with cipher ADH-AES256-SHA (256/256 bits))
    	(No client certificate requested)
    	by mail.cs.ait.ac.th (Postfix) with ESMTPS id C27D63A3853
    	for <on at cs.ait.ac.th>; Tue,  1 Sep 2009 10:46:06 +0700 (ICT)
    Received: from projects.linpro.no (localhost [127.0.0.1])
    	by projects.linpro.no (Postfix) with ESMTP id 050BA1F7343;
    	Tue,  1 Sep 2009 05:24:39 +0200 (CEST)
    X-Original-To: varnish-dev at projects.linpro.no
    Delivered-To: varnish-dev at projects.linpro.no
    Received: from localhost (localhost [127.0.0.1])
    	by projects.linpro.no (Postfix) with ESMTP id 393311F733F
    	for <varnish-dev at projects.linpro.no>;
    	Tue,  1 Sep 2009 05:24:37 +0200 (CEST)
    X-Virus-Scanned: Debian amavisd-new at projects.linpro.no
    Received: from projects.linpro.no ([127.0.0.1])
    	by localhost (projects.linpro.no [87.238.49.148]) (amavisd-new,
    	port 10024)
    	with ESMTP id RsE-dh1azkWg for <varnish-dev at projects.linpro.no>;
    	Tue,  1 Sep 2009 05:24:30 +0200 (CEST)
    X-Greylist: delayed 1538 seconds by postgrey-1.27 at projects;
    	Tue, 01 Sep 2009 05:24:30 CEST
    Received: from mail-gx0-f206.google.com (mail-gx0-f206.google.com
    	[209.85.217.206]) by projects.linpro.no (Postfix) with ESMTP
    	for <varnish-dev at projects.linpro.no>;
    	Tue,  1 Sep 2009 05:24:29 +0200 (CEST)
    Received: by gxk2 with SMTP id 2so6124197gxk.7
    	for <varnish-dev at projects.linpro.no>;
    	Mon, 31 Aug 2009 20:24:29 -0700 (PDT)
>>> MIME-Version: 1.0
    Received: by 10.151.86.12 with SMTP id o12mr4760716ybl.94.1251773930984; Mon, 
    	31 Aug 2009 19:58:50 -0700 (PDT)
    Date: Mon, 31 Aug 2009 22:58:50 -0400
    Message-ID: <22964b960908311958r211f8ac9g1b6eee46ed92d158 at mail.gmail.com>
    Subject: Saint mode
    From: Javier Frias <jfrias at gmail.com>
    To: varnish-dev <varnish-dev at projects.linpro.no>
    X-BeenThere: varnish-dev at projects.linpro.no
    X-Mailman-Version: 2.1.9
    Precedence: list
    List-Id: "Discussions regarding Varnish development."
    	<varnish-dev.projects.linpro.no>
    List-Unsubscribe: <http://projects.linpro.no/mailman/listinfo/varnish-dev>,
    	<mailto:varnish-dev-request at projects.linpro.no?subject=unsubscribe>
    List-Archive: <http://projects.linpro.no/pipermail/varnish-dev>
    List-Post: <mailto:varnish-dev at projects.linpro.no>
    List-Help: <mailto:varnish-dev-request at projects.linpro.no?subject=help>
    List-Subscribe: <http://projects.linpro.no/mailman/listinfo/varnish-dev>,
    	<mailto:varnish-dev-request at projects.linpro.no?subject=subscribe>
    Content-Type: multipart/mixed; boundary="===============2007422974=="
>>> Mime-version: 1.0
    Sender: varnish-dev-bounces at projects.linpro.no
    Errors-To: varnish-dev-bounces at projects.linpro.no
    


From bob.t.terminal at gmail.com  Mon Sep 14 21:00:22 2009
From: bob.t.terminal at gmail.com (Bob T. Terminal)
Date: Mon, 14 Sep 2009 16:00:22 -0500
Subject: external auth from request header values
Message-ID: <B6372FC3-7454-4D41-8A94-1C0869595756@gmail.com>

Former squid user looking at trying to transition to Varnish. Trying  
to setup an auth check for incoming requests that would look at the  
request headers cookie and pass values to a mysql server for  
authentication. Currently have a external application which can handle  
this, what is the best way to go about calling this external app  
within Varnish for the authentication check? (Squid refereed to this  
as an external acl helper)  if there is a better solution please point  
me in the right direction

Thanks
Bob


From phk at phk.freebsd.dk  Thu Sep 24 14:23:08 2009
From: phk at phk.freebsd.dk (Poul-Henning Kamp)
Date: Thu, 24 Sep 2009 14:23:08 +0000
Subject: [PATCH] Fix typo in esi diagnostic 
In-Reply-To: Your message of "Thu, 20 Aug 2009 20:05:20 +0200."
	<591B5DF8-A755-40F2-9677-BE0D3614F6F8@dwim.org> 
Message-ID: <11034.1253802188@critter.freebsd.dk>


Sorry about the delay, your email had gotten sorted into the "later"
mailbox.

It's fixed now, thanks for noticing.

Poul-Henning

In message <591B5DF8-A755-40F2-9677-BE0D3614F6F8 at dwim.org>, "Jos I. Boumans" wr
ites:
>Greetings,
>
>While browsing the source, I came a cross a small typo. Patch below sig.
>
>Cheers,
>
>--
>
>	Jos Boumans
>
>	"Whenever you find you are on the side of the majority,
>	it is time to pause and reflect." - Mark Twain
>
>Index: varnish-cache/bin/varnishd/cache_esi.c
>===================================================================
>--- varnish-cache/bin/varnishd/cache_esi.c      (revision 4207)
>+++ varnish-cache/bin/varnishd/cache_esi.c      (working copy)
>@@ -370,7 +370,7 @@
>                         continue;
>                 if (Tlen(val) == 0) {
>                         esi_error(ew, tag.b, Tlen(tag),
>-                           "ESI esi:include src attribute withou  
>value");
>+                           "ESI esi:include src attribute without  
>value");
>                         continue;
>                 }
>
>
>
>_______________________________________________
>varnish-dev mailing list
>varnish-dev at projects.linpro.no
>http://projects.linpro.no/mailman/listinfo/varnish-dev
>

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.


From yuri.linux at gmail.com  Tue Sep  8 16:12:35 2009
From: yuri.linux at gmail.com (Yuri)
Date: Tue, 08 Sep 2009 16:12:35 -0000
Subject: about vcl
Message-ID: <4aa68089.141bf30a.4f58.0849@mx.google.com>

Hi Everybody,

         I have carried out step by step in the Chinese translation of these
documents, Obviously the level of my English is not very good, so some
elements with doubt.

may you explain them by straightforward statement.

 

In VCL->Syntax overview

         "* Assignments are introduced with the set keyword. There are no
user-defined variables; values can only be assigned to variables attached to
backend, request or document objects. Most of these are typed, and the
values assigned to them must have a compatible unit suffix."

 

This paragraph I thought about it for a long time did not fully understand.

 

--Yuri

Thanks and Blessing

  _____  

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.varnish-cache.org/lists/pipermail/vinyl-dev/attachments/20090908/40744177/attachment.html>