FHIR Chat · Grahame's test servers

I am getting a 500 error on the /r3 version of the server. First, I am not sure what that covers. The /r4 has the Device-DeviceComponent merge. Before this merge there were versions of FHIR that had DeviceComponent with a property element. I am not sure if this /r3 version supports that or the one I am looking for is just an outdated version of r4

In any case, on the r3 version I am getting this error
FHIR Response indicates: 500
Found a property of type array looking for an object at $.entry[1].resource.identifier

With /r4, I get a similar 500 error but the message is that the DeviceComponent does not exist; which is true for the latest r4 version. But I do not understand the above error. Anyone have any ideas?

Grahame Grieve (Sep 27 2018 at 11:18):

does your resource conform to r3?

Brian Reinhold (Sep 27 2018 at 11:19):

does your resource conform to r3?

@Grahame Grieve I am not sure. Its 3.2.0. Where does version 3 stop and version 4 begin?

Grahame Grieve (Sep 27 2018 at 11:20):

r3 = 3.0.1. 3.2.0 is no longer supported anywhere I know

Brian Reinhold (Sep 27 2018 at 11:22):

r3 = 3.0.1. 3.2.0 is no longer supported anywhere I know

Okay. The UHN server supports 3.2.0; I just tried it. The resources I just uploaded are prior to the Device-DeviceComponent merge. I have been using HAPI FHIR and it doesnt support 3.5.0 yet. It Says it supports 3.5.0 but NOT the structures. So the DeviceComponent, for example, is still present and the new Device elements are not present.

Grahame Grieve (Sep 27 2018 at 11:23):

@James Agnew I thought a HAPI release was planned?

James Agnew (Sep 27 2018 at 12:14):

Unfortunately we didn't manage to bump the R4 snapshot as a part of the last release. We tried, but there have been quite a few significant changes recently in R4 that completely broke HAPI's build when we went to sync it (e.g. the new mandatory narrative rule broke just about every unit test we have, the resolve() that has been added to many search parameter paths completely breaks the way we do indexing, the paths for compartment definitions have become fhirpaths which means our security module needs a bunch of work, etc.) I'm desperately trying to get a snapshot release out in time for Baltimore right now.

Grahame Grieve (Sep 27 2018 at 12:24):

thanks

Lloyd McKenzie (Sep 27 2018 at 15:04):

What mandatory narrative rule??

Brian Reinhold (Sep 28 2018 at 08:48):

Unfortunately we didn't manage to bump the R4 snapshot as a part of the last release. We tried, but there have been quite a few significant changes recently in R4 that completely broke HAPI's build when we went to sync it (e.g. the new mandatory narrative rule broke just about every unit test we have, the resolve() that has been added to many search parameter paths completely breaks the way we do indexing, the paths for compartment definitions have become fhirpaths which means our security module needs a bunch of work, etc.) I'm desperately trying to get a snapshot release out in time for Baltimore right now.

@James Agnew Anything yet? Slim to no chance to implement the new PHD uploader in a day anyways but it would be cool to try it against Grahame's server (the only one in town that appears to support R4). Is branch 'sync_r4' available as a snapshot? 3.6.0-SNAPSHOT does not seem to do it.

Jay Gustafson (Apr 30 2020 at 14:53):

@Grahame Grieve How compliant is the test.fhir.org/R4 sandbox data with the USCDI model?

Grahame Grieve (Apr 30 2020 at 22:58):

pretty much not. Sometimes USCDI data is posted to my server, since anyone can post anything, but there's no reason to expect that the data that is there is US based, let alone USCDI compliant. On the other hand, I think my server supports all of US-Core search parameters etc

Robert Scanlon (May 01 2020 at 13:32):

If you are looking for complete USCDI v1 compliant data sets @Jay Gustafson check out https://github.com/inferno-community/uscore-data-script and https://github.com/inferno-community/uscore-data-sets

John Moehrke (May 01 2020 at 15:36):

Seems Inferno also has a validating server for USCDI https://github.com/inferno-community/fhir-validator-wrapper

René Spronk (Jun 11 2020 at 10:29):

.. test.fhir.org/r4 has been unreachable for at least 24 hours.

Grahame Grieve (Jun 11 2020 at 10:41):

thanks. server has run out of disk space :-(

René Spronk (Jun 16 2020 at 06:21):

..birefly online yesterday, now it's gone again.

Grahame Grieve (Jun 16 2020 at 07:55):

coming again.

Tim Harsch (Jun 24 2020 at 23:47):

I was trying to access the server at http://test.fhir.org/r4 , it appears to be down. fwiw: I originally looked on the site for a link for reporting issues before coming here. image.png