Ok, so I just posted a similar question to the mapfish users group and got
an answer (it's a client side issue, so please excuse the noise).  I'll post
it here:

 

I added:

Ext.Ajax.timeout = 120000; //2 minutes

 

To the top of my javascript now my larger maps are not timing out.

 

Steve

 

Stephen Mather
Geographic Information Systems (GIS) Manager
(216) 635-3243

[email protected]
 <http://www.clemetparks.com/> clevelandmetroparks.com

 

 

 

 

From: Stephen V. Mather [mailto:[email protected]] 
Sent: Thursday, July 21, 2011 9:24 AM
To: [email protected]; 'Andrea Aime'
Cc: [email protected]
Subject: RE: [Geoserver-users] Really BIG WMS requests

 

Hi All,

            I have GeoServer (2.1.0, build date May 11 on a Debian 64-bit
machine running in a Tomcat 6.0.32 container) configured with MapFish, with
a GeoExt client for doing print services.  If I print a sufficiently small
document (either area or dpi) it works great.  If I go over a certain size,
it fails.  Here's what I observe in the failed request:

 

I post the request (press "Print" button).  As I watch the debian machine
that GS is on, I see the CPUs go up on the server.  In a good request,
eventually that CPU rate drops on the server and a map is sent back.  On a
too big request, the request runs for 30 seconds, on the 31st second, the
GeoExt client returns without the PDF, but the request continues to run on
the server (CPUs rate high) for a longer period of time.

 

            I strongly suspect it is a timeout issue, since it stops at 30
seconds every time, but I'm not sure whether I need to worry about timeout
in the MapFish config, Tomcat, or (gulp, showing my ignorance) the browser
or all three.  Any thoughts on this?  I have modified GeoServer so that it
doesn't time out for about 2 minutes on requests.

 

Thanks,

Steve

 

 

http://www.clemetparks.com/images/esig/cmp-ms-90x122.pngStephen Mather
Geographic Information Systems (GIS) Manager
(216) 635-3243

[email protected]
 <http://www.clemetparks.com/> clevelandmetroparks.com

 

 

 

 

From: Stephen V. Mather [mailto:[email protected]] 
Sent: Friday, July 15, 2011 1:17 PM
To: [email protected]; 'Andrea Aime'
Cc: [email protected]
Subject: RE: [Geoserver-users] Really BIG WMS requests

 

I'll post over to the MapFish forum and double check the config.yaml
settings for timeouts.

 

Steve

 

http://www.clemetparks.com/images/esig/cmp-ms-90x122.pngStephen Mather
Geographic Information Systems (GIS) Manager
(216) 635-3243

[email protected]
 <http://www.clemetparks.com/> clevelandmetroparks.com

 

 

 

 

From: Stephen V. Mather [mailto:[email protected]] 
Sent: Friday, July 15, 2011 1:11 PM
To: 'Andrea Aime'
Cc: [email protected]
Subject: Re: [Geoserver-users] Really BIG WMS requests

 

Ah, you caught me not doing my homework.  It looks like it might be a
timeout on the request instead.

 

http://www.clemetparks.com/images/esig/cmp-ms-90x122.pngStephen Mather
Geographic Information Systems (GIS) Manager
(216) 635-3243

[email protected]
 <http://www.clemetparks.com/> clevelandmetroparks.com

 

 

 

 

From: [email protected] [mailto:[email protected]] On Behalf Of
Andrea Aime
Sent: Friday, July 15, 2011 12:55 PM
To: [email protected]
Cc: [email protected]
Subject: Re: [Geoserver-users] Really BIG WMS requests

 

On Fri, Jul 15, 2011 at 5:19 PM, Stephen V. Mather
<[email protected]> wrote:

Hi All,

                It's my day for questions, I guess.  Using the MapFish
plugin, we're making some pretty big WMS requests.  I've bumped up the
memory in the Tomcat container, clustered the GeoServer installation to
multiple instances, reduced the resolution of the request to an acceptable
compromise (no 1200dpi requests here. J), etc. etc..  However, with a 36 in
x48 (about and A0 for those in Europe) request, I'll still throw and error
because JAI is limited to 2GB of memory.  Is there any workaround for this,
or is this just the reality of JAI for now?

 

Are you sure the error message says 2GB and it's just not telling you that
that you went

above the _configurable_ wms limits?

Would like to see the actual error, with a full stack trace

 

Cheers

Andrea


-- 
-------------------------------------------------------
Ing. Andrea Aime
GeoSolutions S.A.S.
Tech lead

Via Poggio alle Viti 1187
55054  Massarosa (LU)
Italy

phone: +39 0584 962313
fax:      +39 0584 962313

http://www.geo-solutions.it
http://geo-solutions.blogspot.com/
http://www.youtube.com/user/GeoSolutionsIT
http://www.linkedin.com/in/andreaaime
http://twitter.com/geowolf

-------------------------------------------------------

<<image002.png>>

------------------------------------------------------------------------------
BlackBerry&reg; DevCon Americas, Oct. 18-20, San Francisco, CA
The must-attend event for mobile developers. Connect with experts. 
Get tools for creating Super Apps. See the latest technologies.
Sessions, hands-on labs, demos & much more. Register early & save!
http://p.sf.net/sfu/rim-blackberry-1
_______________________________________________
Geoserver-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geoserver-users

Reply via email to