On 20/08/2013 14:53, Sergey Beryozkin wrote:
Hi Francesco

I did a minor fix to the code exposed after the refactoring I did there and I see Syncope tests passing (mvn clean verify in the core module) with the locally rebuilt CXF 2.7.x,

Give it a try please when you get a chance

Hi Sergey,
I have rebuilt CXF 2.7.x from source (as per your instructions below), then rebuilt Syncope and I can confirm that everything is back working. I'll be waiting for next SHAPSHOT artifacts deployment for closing this issue on Syncope.

As usual, thanks for your support.

...BTW, watching Maven building CXF gives a rough idea of its wideness and complexity, so I can only figure out how difficult could be finding and fixing bugs there: nice works guys!

Regards.

On 19/08/13 17:08, Sergey Beryozkin wrote:
I've rebuilt CXF 2.7.x & Syncope with Java 7, and now it is consistently
failing.
OK then, no way I can skip fixing the issue :-)

I'll update the thread when I get more info

Cheers, Sergey

On 19/08/13 16:20, Francesco Chicchiriccò wrote:
On 19/08/2013 15:00, Sergey Beryozkin wrote:
Hi,

Can you please try something similar to what I did:

svn co http://svn.apache.org/repos/asf/cxf/branches/2.7.x-fixes/

and then 'mvn install -Pfastinstall'

and then rerun Syncope tests ?

Nope, everything the same, I have also tried with OpenJDK 6 / OpenJDK 7
/ Oracle JDK 7, nothing changes.

Regards.

On 19/08/13 15:37, Sergey Beryozkin wrote:
Hi,
On 19/08/13 14:51, Francesco Chicchiriccò wrote:
On 19/08/2013 12:57, Sergey Beryozkin wrote:
Hi Francesco

I reproduced it. Looking into it - can't guarantee a fix today - will
have to sign off in few hours, but it's the only issue I'm dealing
with next.

Don't worry Sergey, take it easy and thank you very much for caring:
the
QoS is extremely high in these places...

Thanks Francesco, sadly I seem to be spending a lot of time recently on
the regressions caused by my fixes :-).

Actually, strangely enough, I saw Syncope tests failing, then I rebuilt
2.7.x locally (before doing the debugging) and then all tests passed.

I wonder what am I missing, is it possible the failures are caused by
the stale snapshot ?

Cheers, Sergey

Regards.

On 19/08/13 12:41, Francesco Chicchiriccò wrote:
On 16/08/2013 21:24, Sergey Beryozkin wrote:
Hi
On 16/08/13 10:56, ilgrosso wrote:
Sergey Beryozkin wrote
Hi Robert

I'm hoping I've fixed it as part of
https://issues.apache.org/jira/browse/CXF-5204

Please try the snapshots a bit later on

Hi Sergey,
after a while, this morning I have re-built Apache Syncope trunk
(featuring
CXF 2.7.7-SNAPSHOT) and started getting internal server errors
(not
much
more info than this, actually...) when accessing generic methods.

Could it be that changes for CXF-5204 interfere with fixes you
have
provided
during our last discussion [1] in CXF-5140?


We have all the tests I added as part of our earlier work
passing, so
I'm not sure at the moment. I'll look into it in Monday: if you
happen
to have an example of (the class/method signature) where you see
failures then please paste them here, otherwise I can build Syncope
trunk myself,

Hi Sergey,
for example, any request to the Schema service [2] ends up with
(Content-Type: application/json)

11:27:50.017 INFO org.apache.syncope.core.rest.AbstractTest -
Performing IT with content type application/json
11:27:50.086 ERROR
org.apache.syncope.core.rest.utils.RestServiceExceptionMapper -
Exception thrown by REST method: Can not construct instance of
org.apache.syncope.core.services.AbstractServiceImpl, problem:
abstract
types either need to be mapped to concrete types, have custom
deserializer, or be instantiated with additional type information
  at [Source:
org.apache.cxf.transport.http.AbstractHTTPDestination$1@1c3d5104;
line:
1, column: 1]
com.fasterxml.jackson.databind.JsonMappingException: Can not
construct
instance of org.apache.syncope.core.services.AbstractServiceImpl,
problem: abstract types either need to be mapped to concrete types,
have
custom deserializer, or be instantiated with additional type
information
  at [Source:
org.apache.cxf.transport.http.AbstractHTTPDestination$1@1c3d5104;
line:
1, column: 1]
     at
com.fasterxml.jackson.databind.JsonMappingException.from(JsonMappingException.java:164)




~[jackson-databind-2.2.2.jar:2.2.2]
[...]

while with (Content-type: application/xml) everything is working
fine
but list() that ends up with

Response-Code: 500
Encoding: UTF-8
Content-Type: text/plain;charset=UTF-8
Headers: {connection=[close], Content-Length=[67],
content-type=[text/plain;charset=UTF-8], Date=[Mon, 19 Aug 2013
09:37:47
GMT], Server=[Apache-Coyote/1.1]}
Payload: No message body writer has been found for response class
ArrayList.

Any clue? Thanks.

Regards.

Thanks for your support.
Regards.

[1]
http://cxf.547215.n5.nabble.com/Cannot-deserialize-JSON-via-JacksonJaxbJsonProvider-that-works-via-Jackson-s-ObjectMapperm-anyway-td5731029.html
[2]
https://svn.apache.org/repos/asf/syncope/trunk/common/src/main/java/org/apache/syncope/common/services/SchemaService.java

--
Francesco Chicchiriccò

ASF Member, Apache Syncope PMC chair, Apache Cocoon PMC Member
http://people.apache.org/~ilgrosso/

Reply via email to