[GeoNetwork-users] Fed up with GeoNetwork

Folks,

we tried to use GeoNetwork for a geo-spatial infrastructure in our project
for one year now. I'm really fed up with this software. Is there anybody out
there who is using it and is satisfied with it in a productive enviroment?

I have my doubts about this and you are welcome to show me a least one
working reference installation. Googling for installations of GeoNetwork
shows up several sites with the same problems I encountered the last year.

Problems starts with the Java/Tomcat-enviroment, you need to know how to
configure memory, editing files with the wrong editor might crash the whole
system. Simple changes to the XSL-based templates require a system restart.
PostGIS-support needs that you know the right howto and user need to find
out how it works. Fancy error messages in HTML from time to time and search
results that are not really useful.

At this time I do not dare to ask about a working up- and download of
geodata in ZIP-archives and indexing metadata and thumbnails that it to be
found in the wild, i.e. created by ArcCatalog.

A fews weeks ago I deteled all the Java/Tomcat-crap and started coding a
solution on my own, open source of course and available on GitHub, a simple
LAMP-thing coded in PHP. Users can upload geodata with metadata, provide
service-URLs or website-URLs, all done very simple, feel free to have a look
on gdi.geo.hu-berlin.de to get an impression how I think a GDI should work.

Thanks for the time here,

-moenk

-----
Geomatics Lab | Geography Department | Humboldt-Universitaet zu Berlin
web: http://www.geographie.hu-berlin.de
--
View this message in context: http://osgeo-org.1560.n6.nabble.com/Fed-up-with-GeoNetwork-tp5019253.html
Sent from the GeoNetwork users mailing list archive at Nabble.com.

Hi Moenk,

I'm afraid I would have to agree. We've been using Geonetwork on the UK academic SDI for several years now and have had nothing but trouble with it. We're running in a Solaris environment. There have been numerous times when the Solaris swap has spiked up to 17GB while doing simple operations like batch deletes of records or during harvesting / re-indexing - threatening to bring the whole container down. We're also trying to run Geonetwork in a load-balanced / replicated environment and the results have been terrible - the reasons for this are numerous and have to do with strange choices when it comes to the implementation of the domain layer. Anyone heard of the Jeeves MVC framework? Thought not. The interface is slow. The permissions implementation is confusing and doing a batch change of permissions in a live environment is asking for trouble. Spatial search is poor. I continuously have to update the application settings directly in the database as trying to do this via the admin interface repeatedly causes exceptions. The relational db architecture is out of date given the existence of xquery / xml databases as the metadata workflow is xml from end to end. However it seems to be the de-facto standard for GDI but really there's room for a decent alternative...

regards,

Brian.

On 28/11/12 10:12, moenk wrote:

Folks,

we tried to use GeoNetwork for a geo-spatial infrastructure in our project
for one year now. I'm really fed up with this software. Is there anybody out
there who is using it and is satisfied with it in a productive enviroment?

I have my doubts about this and you are welcome to show me a least one
working reference installation. Googling for installations of GeoNetwork
shows up several sites with the same problems I encountered the last year.

Problems starts with the Java/Tomcat-enviroment, you need to know how to
configure memory, editing files with the wrong editor might crash the whole
system. Simple changes to the XSL-based templates require a system restart.
PostGIS-support needs that you know the right howto and user need to find
out how it works. Fancy error messages in HTML from time to time and search
results that are not really useful.

At this time I do not dare to ask about a working up- and download of
geodata in ZIP-archives and indexing metadata and thumbnails that it to be
found in the wild, i.e. created by ArcCatalog.

A fews weeks ago I deteled all the Java/Tomcat-crap and started coding a
solution on my own, open source of course and available on GitHub, a simple
LAMP-thing coded in PHP. Users can upload geodata with metadata, provide
service-URLs or website-URLs, all done very simple, feel free to have a look
on gdi.geo.hu-berlin.de to get an impression how I think a GDI should work.

Thanks for the time here,

-moenk

-----
Geomatics Lab | Geography Department | Humboldt-Universitaet zu Berlin
web: http://www.geographie.hu-berlin.de
--
View this message in context: http://osgeo-org.1560.n6.nabble.com/Fed-up-with-GeoNetwork-tp5019253.html
Sent from the GeoNetwork users mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Keep yourself connected to Go Parallel:
INSIGHTS What's next for parallel hardware, programming and related areas?
Interviews and blogs by thought leaders keep you ahead of the curve.
http://goparallel.sourceforge.net
_______________________________________________
GeoNetwork-users mailing list
GeoNetwork-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-users
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork

--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

Hi Moenk,
that is a piece of email.... Well, let me give my $0.02 contribution as
this list is sometime deadly quiet. I'm not a developer just a research
engineer facing SDI deployments needs. When you do not have in your lab the
time and skills to go from scratch, GeoNetwork is by far the most advanced
Open-Source platform. It is not perfect though and as you said there are
room for improvement. I've check your tool and for a few weeks development I
must congratulates the effort. But as most people will said I would have
think that you could contribute your time, skill and effort to improve
GeoNetwork. Maybe you did ? Why did it does not work out ?
What you report about crash did not happen to me. We have set-up a catalog
in operation four few month now and are mostly happy with it. Of course
there are bugs and some features does not work well, but this is the case
for all contributed effort development softwares. From my point of view I
would said that I'm sometimes more disappointed by the low level of
developers responses on this ML rather than by the level of maturity of the
tool.
On your tool I searched for "solar" (this is my lab's research field) and
found some data from the SWERA project.
http://gdi.geo.hu-berlin.de/details.php?uuid=EDDD25A4-D6ED-4F8F-A5A4-BC88A217E74B
Access to the distribution info links and Metadata led to a 404 error. We
also carry some of those SWERA layers (70) in our GeoNetwork portal. You can
have a look as you asked about of a living GeoNetwork instance here:
http://geocatalog.webservice-energy.org/
I would be more than happy to get your feedback.
All the best for your project.

Lionel

--
View this message in context: http://osgeo-org.1560.n6.nabble.com/Fed-up-with-GeoNetwork-tp5019253p5019316.html
Sent from the GeoNetwork users mailing list archive at Nabble.com.

Hi Moenk

For sure GeoNetwork has things to improve, for example as Brian points
migrate to a modern MVC framework and others that can be easily
listed. Improvements requires projects funding these developments and
contributions from the community either testing the application to report
bugs, providing patches, etc.

I think has been always quite easy to collaborate in the project: users and
developer mailing lists, IRC channel, a TRAC to report issues where people
can provide also patches to be integrated in the project. Few month ago we
moved to GitHub where it's even much easier to provide fixes or new
features. You can check that several users have done improvements/bug fixes
in later months and send a pull request so committers can check and commit
in the code.

Its good you're critical with the project to point things to improve, but
would be better to know why you decided to start from scratch instead of
collaborate with the project to improve it as many other
users/organizations do.

And of course, many organizations are using GeoNetwork in production
systems, you can take a look to
http://geonetwork-opensource.org/gallery/gallery.html for some examples.

Regards,
Jose García

On Wed, Nov 28, 2012 at 3:23 PM, Lionel Menard <
lionel.menard@anonymised.com> wrote:

Hi Moenk,
that is a piece of email.... Well, let me give my $0.02 contribution as
this list is sometime deadly quiet. I'm not a developer just a research
engineer facing SDI deployments needs. When you do not have in your lab the
time and skills to go from scratch, GeoNetwork is by far the most advanced
Open-Source platform. It is not perfect though and as you said there are
room for improvement. I've check your tool and for a few weeks development
I
must congratulates the effort. But as most people will said I would have
think that you could contribute your time, skill and effort to improve
GeoNetwork. Maybe you did ? Why did it does not work out ?
What you report about crash did not happen to me. We have set-up a catalog
in operation four few month now and are mostly happy with it. Of course
there are bugs and some features does not work well, but this is the case
for all contributed effort development softwares. From my point of view I
would said that I'm sometimes more disappointed by the low level of
developers responses on this ML rather than by the level of maturity of
the
tool.
On your tool I searched for "solar" (this is my lab's research field) and
found some data from the SWERA project.

http://gdi.geo.hu-berlin.de/details.php?uuid=EDDD25A4-D6ED-4F8F-A5A4-BC88A217E74B
Access to the distribution info links and Metadata led to a 404 error. We
also carry some of those SWERA layers (70) in our GeoNetwork portal. You
can
have a look as you asked about of a living GeoNetwork instance here:
http://geocatalog.webservice-energy.org/
I would be more than happy to get your feedback.
All the best for your project.

Lionel

--
View this message in context:
http://osgeo-org.1560.n6.nabble.com/Fed-up-with-GeoNetwork-tp5019253p5019316.html
Sent from the GeoNetwork users mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Keep yourself connected to Go Parallel:
INSIGHTS What's next for parallel hardware, programming and related areas?
Interviews and blogs by thought leaders keep you ahead of the curve.
http://goparallel.sourceforge.net
_______________________________________________
GeoNetwork-users mailing list
GeoNetwork-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-users
GeoNetwork OpenSource is maintained at
http://sourceforge.net/projects/geonetwork

--
*
GeoCat Bridge for ArcGIS allows instant publishing of data and metadata on
GeoServer and GeoNetwork. Visit http://geocat.net for details.
_________________________
Jose García
GeoCat bv
Veenderweg 13
6721 WD Bennekom
The Netherlands
http://GeoCat.net/&gt;

*

Hi Moenk,
I figure another user's perspective might be appropriate.
I have been successfully using Geonetwork since 2009 in a production environment.

In 2008 the institute I work for participated in an international research initiative. Participation included the requirement to provide data, etc online. I selected Geonetwork as the tool to do this. We had no problems. At around 12:00PM on the day we started to actually implement it, I started looking for a suitable host, by 4:30PM I had a host & Geonetwork admin/support agreement in place.
Within a couple of days it was running, & I prepared a "HowTo" for our researchers to populate their metadata, which enabled staff to load metadata without ever contacting me regarding the system. It was commended as perhaps the best catalogue & metadata mgmt facility of all participants in the programme.

Since 2009 we have deployed several more instances. I have demonstrated a simple, very basic install & deployment from scratch, it can take well under an hour.
All our systems are using Postgis as the underlying database.

So my results are very different from yours. I agree, it is not easy to customise, is sensitive to the setup of dependencies, & requires a good level of technical expertise to run much more than a standard installation. but have found that once installed & configured, it is a robust & effective tool.
I also agree, compared with other Open Source project I work with regularly, the level of help provided on the list is not as good, but I'm also aware of a local agency who are using ESRI's Geoportal solution, and turning this into a viable production system appears much more difficult.
Given you are looking for a tool with good integration with Arc Catalog, did you not consider Geoportal as an option?

We are extensively harvesting catalogue records, between our own catalogues, between ours & national government catalogues, between ours and international catalogues, and generally (not always) configuring Geonetwork to support this has been relatively straightforward, and the developers have provided very useful assistance (notably Simon Pigot).
In addition to harvesting, we use a third party CSW client to embed a simpler client interface in our web sites for users to search & browse the Geonetwork catalogue. This has also worked reliably & effectively (& is also Open Source).
Successive versions have consistently introduced very useful & desirable capabilities, not always as quickly as we would have liked, but at least as quickly as most commercial vendors have enhanced their apps based on users requests, at least in my 30 years experience of such things.
One point, given the large numbeer of significant catalogues running Geonetwork, which never seem to have any recourse to the mailing list, my conclusion is that many users are indeed able to simply download, install & use Geonetwork without problems, or at least without asking there for assistance.
So while you & some others may have found Geonetwork does not provide a viable solution, my experience (& that of many others) has been very different. You also suggest that a Google search found other users finding problems much like yours. I would also suggest a Google search can easily find many viable institutional, national & international Geonetwork catalogues which are running reliably & effectively.

Cheers,
Brent Wood

--- On Wed, 11/28/12, moenk <moenkemt@anonymised.com> wrote:

From: moenk <moenkemt@anonymised.com>
Subject: [GeoNetwork-users] Fed up with GeoNetwork
To: geonetwork-users@lists.sourceforge.net
Date: Wednesday, November 28, 2012, 11:12 PM

Folks,

we tried to use GeoNetwork for a geo-spatial infrastructure in our project
for one year now. I'm really fed up with this software. Is there anybody out
there who is using it and is satisfied with it in a productive enviroment?

I have my doubts about this and you are welcome to show me a least one
working reference installation. Googling for installations of GeoNetwork
shows up several sites with the same problems I encountered the last year.

Problems starts with the Java/Tomcat-enviroment, you need to know how to
configure memory, editing files with the wrong editor might crash the whole
system. Simple changes to the XSL-based templates require a system restart.
PostGIS-support needs that you know the right howto and user need to find
out how it works. Fancy error messages in HTML from time to time and search
results that are not really useful.

At this time I do not dare to ask about a working up- and download of
geodata in ZIP-archives and indexing metadata and thumbnails that it to be
found in the wild, i.e. created by ArcCatalog.

A fews weeks ago I deteled all the Java/Tomcat-crap and started coding a
solution on my own, open source of course and available on GitHub, a simple
LAMP-thing coded in PHP. Users can upload geodata with metadata, provide
service-URLs or website-URLs, all done very simple, feel free to have a look
on gdi.geo.hu-berlin.de to get an impression how I think a GDI should work.

Thanks for the time here,

-moenk

-----
Geomatics Lab | Geography Department | Humboldt-Universitaet zu Berlin
web: http://www.geographie.hu-berlin.de
--
View this message in context: http://osgeo-org.1560.n6.nabble.com/Fed-up-with-GeoNetwork-tp5019253.html
Sent from the GeoNetwork users mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Keep yourself connected to Go Parallel:
INSIGHTS What's next for parallel hardware, programming and related areas?
Interviews and blogs by thought leaders keep you ahead of the curve.
http://goparallel.sourceforge.net
_______________________________________________
GeoNetwork-users mailing list
GeoNetwork-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-users
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork

Jose,

I browsed the list of GN installations in
http://geonetwork-opensource.org/gallery/gallery.html, and not one of those
has more than 100k records, most falling in the 5k-10k range.

In contrast, I'm trying to get a 1million+ records installation to work, and
have been plagued by all sorts of stability problems: record corruption,
out-of-memory errors, the list goes on.

MEF insertion through the web interface takes a lifetime, gives no progress
feedback, and worse even, onced ended up crashing the server and corrupting
the database when I tried to load a mere 1k records. Had to reinstall
everything from scratch.

Since then (started this process back in April) we switched from MySQL to
Postgre/PostGIS, and I ended up writing a program to load MEF records into
GN. Managed to load the full ~1.060.000 records, and everything seemed fine.

But, as I write this, I'm stuggling - again - to restart GN server following
a server power outage, and I'm getting - once again - java out-of-memory
errors.

(output.log attached)

Regards,
Andre Pontes

--
View this message in context: http://osgeo-org.1560.n6.nabble.com/Fed-up-with-GeoNetwork-tp5019253p5021933.html
Sent from the GeoNetwork users mailing list archive at Nabble.com.