Tile seeding stops on timeout request

Description

I'm trying to seed my tiles. Unfortunately for whatever reason some of the tile renderings are timing out.

> 2014-01-14 15:41:16,243 ERROR -
> org.geoserver.platform.ServiceException: This requested used more time than allowed and has been forcefully stopped. Max rendering time is 60.0s

The problem however, is that once the seeder receives this, it stops the entire thread so no further tiles are seeded, even though only one failed. This means it then becomes impossible to seed the layer.

Environment

None

Activity

Show:
codehaus
April 10, 2015, 4:17 PM

CodeHaus Comment From: jratike80 - Time: Wed, 26 Mar 2014 12:55:28 -0500
---------------------
<p>This is really a GeoWebCache issue, but because GWC is integrated with Geoserver it is natural that users are reporting issues here. Right thing to do would be to forward the issue into GWC bug tracker in Github. But who is responsible for doing so and what should be done for the GS issue? Close it or keep open until the bug is resolved in GWC? The latter means probably that the GS issue stays open for a very long time because nobody checks what has happened in the GWC side.</p>

<p>Perhaps we should try to guide Geoserver users to file GWC tickets directly into Github. I know that last week I said the opposite.</p>

codehaus
April 10, 2015, 4:17 PM

CodeHaus Comment From: jonathan_wcc - Time: Fri, 28 Mar 2014 06:47:26 -0500
---------------------
<p>Personally it's a bug-bear of mine when projects have tons of separate trackers (QGIS is especially bad at this). Relying on the user to know which of the innumerable trackers they're supposed to be filing stuff into is a losing proposition - they are unlikely to know about the difference. And that's assuming they can find the other tracker and want to go through the effort of creating an account for it (often required).

I'm not sure what the solution is, but I don't think making users do the splitting is the right thing to do unless they're being told to report the bug (in which case point they can be pointed to the correct list).</p>

<p>(Of course, the irony here is that in this case as someone who's fairly active in the community and follows the dev list, I'm probably better placed to try and make the GWC/GS distinction than a regular user.)</p>

<p>((Sorry, offtopic.))</p>

Jukka Rahkonen
January 14, 2020, 5:10 PM

Good news, Andrea is working on this!

Assignee

Unassigned

Reporter

codehaus

Triage

None

Fix versions

None

Affects versions

Priority

Medium
Configure