Hi all,
After upgrading from GeoServer 2.4.8 to 2.26.1 (yes, a big leap!), I have noticed in my testing phase that the total number of layers listed on the Layer Preview screen erroneously includes the layers that I have Enabled but notAdvertised. For example, the part near the top of the Layer Preview screen shows:
Results 1 to 25 (out of 739 items)
But I only have 703 items that are Advertised, so this number is misleading. In version 2.4.8, this number correctly shows 703 items, so something got scrambled in the interim. In contrast, my Welcome screen in 2.26.1 correctly shows 703 layers:
GeoServer Web Service, pacioos access to 2 workspaces, with 703 layers.
Is the Layer Preview total something that can be rectified in a future release? Many thanks!
Aloha,
John Maurer
Data System Engineer
Pacific Islands Ocean Observing System (PacIOOS)
University of Hawaii at Manoa
Thank you, @aaime-geosolutions for the info since I’m new to this process. So, it sounds like I first report a bug to the user forum (i.e., here), then wait for feedback. Depending on what is discussed in the forum, I then submit the bug using the issue tracker that @jive pointed me to (for a different topic). However, I should have no expectations of it being fixed unless:
I fix it myself, or
I help pay for somebody else to fix it.
And since I am not a Java developer (my skill set is JavaScript and Python), that limits me to the second option. Otherwise, it sounds like it may be fruitless to report bugs. Thanks again for the information.
Hi John,
It seems that you’re right that the layers with Advertised unchecked should have been left out of the Layer Preview. Here is what the User Manual says:
Advertised—A layer is advertised by default. A non-advertised layer will be available in all data access requests (for example, WMS GetMap, WMS GetFeature) but won’t appear in any capabilities document or in the layer preview.
You might be right about it being fruitless to report a bug, but it wouldn’t hurt to try it.
You might be right about it being fruitless to report a bug, but it wouldn’t hurt to try it.
Opening a bug report without funding basically amounts to hoping that someone will decide to use a couple of hours
(fix the bug, write a test) for that specific issue, on a sunday, rather than doing something else.
Can it happen? Yes, it does. Is it likely? No.
The other possibility is that someone is bothered by the same issue enough to spend support contract hours on it,
or that someone selling GeoServer hosting decides to do the same.
That document is not trying to be harsh, just to be real, there is a relatively small set of developers
caring for GeoServer, and they are all around their 50s, normally squashed between work, growing kids,
and aging elders… spare time is so very limited. In order to care for random user reports, we need more
developers, and more importantly, ones that have a ton of spare time.
Either that, or adopt a GDAL like model, where annually a small set of gold and silver donors drop well over 200k on the project so that everyone else can have their question, bug fixes and feature
request dealt with for the year to come, timely and for free (actually not free… but with someone else’s money).
QGIS is following the same approach, and also getting over 200k annually, by the looks of their sponsorship page.
GeoServer tried to get some ongoing, yearly funding going, with no success so far.
Going back to reporting the issue… yep, I’d recommend doing it anyways.
By opening a bug report, if nothing else, one gets notification when the issue is finally fixed.
I hear you, @aaime-geosolutions. Thank you for the honest reply. I sincerely hope GeoServer can find funding to help sustain itself and grow. Many thanks to all in the GeoServer developer community for providing and supporting a powerful OSGIS server.