Check the logs for a stack trace, perhaps it can provide a clue. Usually the only way to confuse a database setup is by not filling in the correct metadata / spatial index / spatial reference system information. GeoServer makes use of these kind of details when determining the feature type.
Problem solved - or at least avoided. Two ways potentially leads to the glorious access to MSSQL spatial based layers. Both starts with the user choosing “Add a new resource”, after that I choose the relevant data store from the drop down list. Next is one of these two:
“Configure new SQL view…”
Press “Publish”
The last option (2) opens the layer definition dialogue and shows “Reload feature type …”. The first one (1) starts with letting me write a SQL query. This one works.
I am running geoserver on a windows computer with “JVM Version: Oracle Corporation: 1.7.0_17 (Java HotSpot™ 64-Bit Server VM)”.
Check the logs for a stack trace, perhaps it can provide a clue. Usually the only way to confuse a database setup is by not filling in the correct metadata / spatial index / spatial reference system information. GeoServer makes use of these kind of details when determining the feature type.
Problem solved - or at least avoided. Two ways potentially leads to the glorious access to MSSQL spatial based layers. Both starts with the user choosing “Add a new resource”, after that I choose the relevant data store from the drop down list. Next is one of these two:
“Configure new SQL view…”
Press “Publish”
The last option (2) opens the layer definition dialogue and shows “Reload feature type …”. The first one (1) starts with letting me write a SQL query. This one works.
I am running geoserver on a windows computer with “JVM Version: Oracle Corporation: 1.7.0_17 (Java HotSpot™ 64-Bit Server VM)”.
I think there is a misunderstanding here: the option 1 and 2 are quite different in scope: the “Configure new SQL view” allows you to publish a “virtual table” from your database built using a sql query, while the second one is used to publish an existing database table “as is”, so also the configuration UI is different for the layers built using one of the two methods.
With the first one you get an “Edit SQL View” link that allows you to define the sql for the virtual table, for the second one you cannot define a sql query, you only get a “Reload feature type…” that simply reads the table fields and properties from the database and synchronizes the internal cache.
So, as it is usually said: “it’s not a bug, it’s a feature” :-).
On Tue, Feb 11, 2014 at 8:55 AM, Mauro Bartolomeoli <
mauro.bartolomeoli@anonymised.com> wrote:
With the first one you get an "Edit SQL View" link that allows you to
define the sql for the virtual table, for the second one you cannot define
a sql query, you only get a "Reload feature type..." that simply reads the
table fields and properties from the database and synchronizes the internal
cache.
So, as it is usually said: "it's not a bug, it's a feature" :-).
In other words, if you don't have tables listed, it means you either have
nothing available with the user/schema you chose,
or have permission issues (lack of grants) that prevent GeoServer from
listing the available tables
Thankyou Mauro and Andrea – your explanations of the issue made me see it clearer!
I have changed the issue in JIRA to a possible improvement in documentation or the user interface J Given time I might even submit a suggestion to update of the documentation.
Problem solved - or at least avoided. Two ways potentially leads to the glorious access to MSSQL spatial based layers. Both starts with the user choosing “Add a new resource”, after that I choose the relevant data store from the drop down list. Next is one of these two:
“Configure new SQL view…”
Press “Publish”
The last option (2) opens the layer definition dialogue and shows “Reload feature type …”. The first one (1) starts with letting me write a SQL query. This one works.
I am running geoserver on a windows computer with “JVM Version: Oracle Corporation: 1.7.0_17 (Java HotSpot™ 64-Bit Server VM)”.
I think there is a misunderstanding here: the option 1 and 2 are quite different in scope: the “Configure new SQL view” allows you to publish a “virtual table” from your database built using a sql query, while the second one is used to publish an existing database table “as is”, so also the configuration UI is different for the layers built using one of the two methods.
With the first one you get an “Edit SQL View” link that allows you to define the sql for the virtual table, for the second one you cannot define a sql query, you only get a “Reload feature type…” that simply reads the table fields and properties from the database and synchronizes the internal cache.
So, as it is usually said: “it’s not a bug, it’s a feature” :-).
On Tue, Feb 11, 2014 at 9:39 AM, Ragnvald Larsen <
ragnvald.larsen@anonymised.com> wrote:
Thankyou Mauro and Andrea - your explanations of the issue made me see
it clearer!
I have changed the issue in JIRA to a possible improvement in
documentation or the user interface J Given time I might even submit a
suggestion to update of the documentation.
Can you share screenshots of what you see? At this point I have doubts
we're talking about the same thing?