Test coverage deletion in various scenarios

[Dimitar Misev]

E.g.

• with rasdaman down but petascope up
• with rasdaman busy in a write transaction (an insert that takes a longer time)
• with the collection manually removed from rasdaman

[Bang Pham Huu]

some scripts to test delete

[Bang Pham Huu]

I've tried to test delete cases in various scenarios (by using these scripts to test ​http://rasdaman.org/attachment/ticket/1450/test_delete_scripts.tar.gz), it works fine with case 1 and 3 that Dimitar gave examples.

However, the case 2 can have a problem when ingesting a big file by wcst_import and all the rasserver are processing. One send a request to delete a collection by WCS

http://localhost:8080/rasdaman/ows?service=WCS&version=2.0.1&request=DeleteCoverage&coverageId=test_wms_3857

then the request from Petascope can delete the coverage metadata in petascopedb, but the rasql query will not be run (I think due to server are busy and somehow it forget this request). In the end, the collection in rasdaman still exists.

rasql -q 'select c from RAS_COLLECTIONNAMES as c' --out string
Result object 2: test_wms_3857

and because petascopedb was removed this coverageId, so the request to delete in WCS will have error

http://localhost:8080/rasdaman/ows?service=WCS&version=2.0.1&request=DeleteCoverage&coverageId=test_wms_3857

This XML file does not appear to have any style information associated with it. The document tree is shown below.
<ows:ExceptionReport xmlns:ows="http://www.opengis.net/ows/2.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xlink="http://www.w3.org/1999/xlink" version="2.0.1" xsi:schemaLocation="http://www.opengis.net/ows/2.0 http://schemas.opengis.net/ows/2.0/owsExceptionReport.xsd">
<ows:Exception exceptionCode="NoSuchCoverage">
<ows:ExceptionText>test_wms_3857</ows:ExceptionText>
</ows:Exception>
</ows:ExceptionReport>

[Bang Pham Huu]

move the bug for case 2 to ​http://rasdaman.org/ticket/1466 for fixing in rasserver

[Dimitar Misev]

We should add tests to the systemtest.

Also case 2, yes it's a problem in rasserver, but also in petascope. Petascope should not remove coverage metadata if rasserver is busy and fails to delete the collection.

[Bang Pham Huu]

Replying to dmisev:

We should add tests to the systemtest.

Also case 2, yes it's a problem in rasserver, but also in petascope. Petascope should not remove coverage metadata if rasserver is busy and fails to delete the collection.

I think we should try to delete the collection first, if success then delete the coverage metadata. About the tests for system test, I'll think later for suitable cases.

[Bang Pham Huu]

the patch was applied, close ticket.