I'm seeing a regression for a netcdf image mosaic layer when upgrading from
2.5 to 2.5.1. Starting from a given zoom level, all tiles simply go black.
There's also a variation where a set of evenly spaced tiles go black, while
others remain normal for a given zoomlevel, and then they all return black at
the next zoom level.
I've been looking for the source now for a while, but haven't been able to
figure it out. Here are some more clues I gathered:
-the changes in gt-imagemosaic do not seem to influence the behavior
-the netcdf layer has 3 bands which are merged using the MergeBehavior.Stack,
a similar single banded mosaic, built from the same source data works fine.
-changing the imageio-ext library version does not seem to affect the issue
I'll continue looking myself, but if this rings a bell somewhere then I'm
certainly interested. I also scanned the diff on github, but there's are not
that many changes to the relevant code.
I’m seeing a regression for a netcdf image mosaic layer when upgrading from
2.5 to 2.5.1. Starting from a given zoom level, all tiles simply go black.
There’s also a variation where a set of evenly spaced tiles go black, while
others remain normal for a given zoomlevel, and then they all return black at
the next zoom level.
I’ve been looking for the source now for a while, but haven’t been able to
figure it out. Here are some more clues I gathered:
-the changes in gt-imagemosaic do not seem to influence the behavior
-the netcdf layer has 3 bands which are merged using the MergeBehavior.Stack,
a similar single banded mosaic, built from the same source data works fine.
-changing the imageio-ext library version does not seem to affect the issue
I’ll continue looking myself, but if this rings a bell somewhere then I’m
certainly interested. I also scanned the diff on github, but there’s are not
that many changes to the relevant code.
thanks,
Jeroen
HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions
Find What Matters Most in Your Big Data with HPCC Systems
Open Source. Fast. Scalable. Simple. Ideal for Dirty Data.
Leverages Graph Analysis for Fast Processing & Easy Data Exploration http://p.sf.net/sfu/hpccsystems
You may wish to create a bug report for this one as it has been a couple of days without a reply. Also gives you a chance to attach screen snaps.
I think we may of upgraded ImageIO extension between 2.5 and 2.5.1.
–
Jody
HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions
Find What Matters Most in Your Big Data with HPCC Systems
Open Source. Fast. Scalable. Simple. Ideal for Dirty Data.
Leverages Graph Analysis for Fast Processing & Easy Data Exploration http://p.sf.net/sfu/hpccsystems
I’m seeing a regression for a netcdf image mosaic layer when upgrading from
2.5 to 2.5.1. Starting from a given zoom level, all tiles simply go black.
There’s also a variation where a set of evenly spaced tiles go black, while
others remain normal for a given zoomlevel, and then they all return black at
the next zoom level.
I’ve been looking for the source now for a while, but haven’t been able to
figure it out. Here are some more clues I gathered:
-the changes in gt-imagemosaic do not seem to influence the behavior
-the netcdf layer has 3 bands which are merged using the MergeBehavior.Stack,
a similar single banded mosaic, built from the same source data works fine.
-changing the imageio-ext library version does not seem to affect the issue
I’ll continue looking myself, but if this rings a bell somewhere then I’m
certainly interested. I also scanned the diff on github, but there’s are not
that many changes to the relevant code.
thanks,
Jeroen
HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions
Find What Matters Most in Your Big Data with HPCC Systems
Open Source. Fast. Scalable. Simple. Ideal for Dirty Data.
Leverages Graph Analysis for Fast Processing & Easy Data Exploration http://p.sf.net/sfu/hpccsystems
thanks, but the problem is that it's a complex setup, which basically only
runs here internally. I would have to replicate the issue in a setup without
internal dependencies to give you something.
Also screenshots are not going to be very helpful since they are simply black.
I also haven't seen anyone else complaining so I guess it's something very
specific. Perhaps I'll try debugging it a bit more later on.
thanks anyway,
Jeroen
On Tuesday 17 June 2014 15:52:11 Nicola Lagomarsini wrote:
Hi Jeroen,
sorry for the delay. Could you provide us a sample data for replicating the
issue?
Best Regards,
Nicola Lagomarsini.
2014-06-14 10:42 GMT+02:00 Jody Garnett <jody.garnett@anonymised.com>:
> You may wish to create a bug report for this one as it has been a couple
> of days without a reply. Also gives you a chance to attach screen snaps.
>
> I think we may of upgraded ImageIO extension between 2.5 and 2.5.1.
> --
> Jody
>
> Jody Garnett
>
>
> On Thu, Jun 12, 2014 at 7:59 PM, Jeroen Dries <jeroen.dries@anonymised.com>
>
> wrote:
>> Hi all,
>>
>> I'm seeing a regression for a netcdf image mosaic layer when upgrading
>> from
>> 2.5 to 2.5.1. Starting from a given zoom level, all tiles simply go
>> black.
>> There's also a variation where a set of evenly spaced tiles go black,
>> while
>> others remain normal for a given zoomlevel, and then they all return
>> black at
>> the next zoom level.
>>
>> I've been looking for the source now for a while, but haven't been able
>> to
>> figure it out. Here are some more clues I gathered:
>> -the changes in gt-imagemosaic do not seem to influence the behavior
>> -the netcdf layer has 3 bands which are merged using the
>> MergeBehavior.Stack,
>> a similar single banded mosaic, built from the same source data works
>> fine.
>> -changing the imageio-ext library version does not seem to affect the
>> issue
>>
>> I'll continue looking myself, but if this rings a bell somewhere then I'm
>> certainly interested. I also scanned the diff on github, but there's are
>> not
>> that many changes to the relevant code.
>>
>> thanks,
>> Jeroen
>>
>>
>> -------------------------------------------------------------------------
>> ----- HPCC Systems Open Source Big Data Platform from LexisNexis Risk
>> Solutions Find What Matters Most in Your Big Data with HPCC Systems
>> Open Source. Fast. Scalable. Simple. Ideal for Dirty Data.
>> Leverages Graph Analysis for Fast Processing & Easy Data Exploration
>> http://p.sf.net/sfu/hpccsystems
>> _______________________________________________
>> Geoserver-users mailing list
>> Geoserver-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/geoserver-users
>
> --------------------------------------------------------------------------
> ---- HPCC Systems Open Source Big Data Platform from LexisNexis Risk
> Solutions Find What Matters Most in Your Big Data with HPCC Systems
> Open Source. Fast. Scalable. Simple. Ideal for Dirty Data.
> Leverages Graph Analysis for Fast Processing & Easy Data Exploration
> http://p.sf.net/sfu/hpccsystems
> _______________________________________________
> Geoserver-users mailing list
> Geoserver-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/geoserver-users