Stream: Clinicians on FHIR
Topic: server message error
Michael Padula (Jan 17 2019 at 16:59):
I am having trouble editing the scenarios and resources in ConMan I was working on earlier today. I get the following error:
"I was unable to retrieve the conformance resource: http://hl7.org/fhir/StructureDefinition/CarePlan from http://fhirtest.uhn.ca/baseDstu3/. Is it on-line, and accepting CORS requests?"
and other similar errors. Any idea how best to trouble shoot. Thanks, -Mike
David Hay (Jan 17 2019 at 17:29):
Looks like the hapi 3 server is down - which track?
Michael Padula (Jan 17 2019 at 17:31):
Paediatric: Infant with OTC. Thanks.
David Hay (Jan 17 2019 at 17:32):
go to the track definition, select the 'servers & IG tab' at the bottom and change the conformance & data server to the clinFHIR 3 server as shown in the attacxhed screen shot (use the 3 dots to the right to select the server). Save the track and restart the server. does that resolve it?
David Hay (Jan 17 2019 at 17:33):
Screen-Shot-2019-01-18-at-6.30.33-AM.png
David Hay (Jan 17 2019 at 17:36):
btw - I'd but a short description of the scenario in each one - effectively describing the 'state' of the scenario that the graph of resources should represent. You can also add a link to the confluence page (press the + symbol at the bottom of the scenario edit box to add it.
David Hay (Jan 17 2019 at 17:39):
It might help to remember that each scenario represents a 'single point in time' .
(Actually, more and more I'm thinking that the term 'scenario' is misleading for this type of usage -clinical. As we discussed on the call yesterday - I'm wondering if 'workflow step' might be better for this kind of track.
And even 'Track' is really not quite right...
David Hay (Jan 17 2019 at 17:42):
I'm coming to think that rather than combining the technical and clinical connectathons in a single conman app, we should have a separate conman instance for clinical - perhaps one which doesn;t change between events but rather evolves over time (remembering that it can be used outside of an actual face to face meeting). That way the work put into making the tracks/scenarios as good as they can be isn't lost...
Michael Padula (Jan 17 2019 at 17:46):
Thanks for the clarification regarding scenario. I had interpreted it more like an encounter as opposed to split into each workflow step.
I tried making those changes and got the following:
An unexpected error occurred: Uncaught TypeError: Cannot read property 'issue' of null on line 415 for http://conman.fhir.org/js/cofCtrl.js
Michael Padula (Jan 17 2019 at 17:47):
Agree it would be great to do more active work on this outside the WGMs.
David Hay (Jan 17 2019 at 17:48):
did you refresh the page?
David Hay (Jan 17 2019 at 17:50):
that error looks like it could occur when you mouse over the 'validity status' in the 'server interaction' tab - right?
David Hay (Jan 17 2019 at 17:52):
If so, it can be ignored (I've fixed it, but not uploaded the fix)
David Hay (Jan 17 2019 at 17:54):
I'm wondering if a good use of the time on friday would be to work on the detail of the track - ie each scenario - rather than focussing on actually building the graphs. The graph building can then be done (and discussed) off line...
David Hay (Jan 17 2019 at 17:57):
So: think of each 'Track' as if it actually was the scenario, and each 'Scenario' as a single step. (I can rename them later to make it easier to understand). I imagine that you could have a reasonable number of steps - eg 10 - that's fine.
Focus on the description of each step - ie what 'state' it represents - the types and other stuff can come later.
David Hay (Jan 17 2019 at 17:59):
If you want to, you can also re-order the steps by going into the track definition, then the scenarios tab and re-ordering them there (Remember in this disucssion that 'Track' = 'Scenario' and 'Scenario' - 'Step'
Michael Padula (Jan 17 2019 at 17:59):
It now works and I can edit the resources again. Thanks. I will reorganize it into descrete scenarios (steps).
David Hay (Jan 17 2019 at 18:00):
If, on the day, you have tooling issues then work on these manually & we can add them later (It will be the middle of the night for me, and while I'm dedicated.... :) )
David Hay (Jan 17 2019 at 18:00):
I look forward to seeing what you come up with!
Michael Padula (Jan 17 2019 at 18:13):
Many thanks, David. While I can see many of the attributes within the resources, I sometimes cannot edit them. Is there another recommended way to edit this when I cannot use the table editor (the blue [ '] icon)?
David Hay (Jan 17 2019 at 18:17):
Can you give me an example? It sounds like you are referring to a Backbone element (ie an element that has children) and is not intended to have a value...
Michael Padula (Jan 17 2019 at 18:26):
There are sometimes I have entered a value and cannot edit it further:
(https://chat.fhir.org/user_uploads/10155/Dff2oB0e15V1RkdUxaUgaeel/Screenshot-2019-01-17-12.21.55.png)
Other times I try to select that icon and nothing seems to happen.
David Hay (Jan 17 2019 at 18:29):
If you click on the edit icon again, you should be able to change it...
David Hay (Jan 17 2019 at 18:30):
the existing value would be replaced by the new one you select... (I assume we're discussion Condition.code)
David Hay (Jan 17 2019 at 18:32):
If you do click on the icon, and nothing seems to happen, then there could be a bug - it would be helpful if you could show the browser debug page and look for errors... (in Chrome it's View > Developer > Javascript console )
Michael Padula (Jan 17 2019 at 19:28):
You were correct, it throws an error:
Screenshot-2019-01-17-13.26.53.png
Michael Padula (Jan 17 2019 at 19:39):
For CarePlan.status I get an error:
Error from terminology server: $expand operation at the type level (no ID specified) requires an identifier or a valueSet as a part of the request.
Michael Padula (Jan 17 2019 at 19:48):
I switched back to the ontology sever and it works. Sorry for all the messages.
David Hay (Jan 17 2019 at 20:06):
Oh no - please keep the messages coming! The earlier I get them, the more likely I can fix them (and time is very short now)
Also - changed your other track away from the public hapi server... Also removed MedicationOrder (it's 'medicationRequest' in STU-3)
David Hay (Jan 17 2019 at 20:08):
so this error is due to issues creating the json representation of the resource- a bug in the generator. I'll trap & ignore it for now. Should be able to upload shortly..
Michael Padula (Jan 17 2019 at 20:22):
Great. Thanks.
David Hay (Jan 17 2019 at 20:23):
and I don't get the careplan ststus error. Can you reload the page and try again>? (it may have been the previous error getting in the way)
David Hay (Jan 17 2019 at 20:36):
Ah - I see the terminology error was resolved...
David Hay (Jan 17 2019 at 20:39):
OK. fix uploaded. If there's an error with creating the resource json, then you'll get an error message - but you should be able to continue...
David Hay (Jan 17 2019 at 20:40):
Thanks for the testing!
Michael Padula (Jan 17 2019 at 21:39):
Two other messages to relay:
Screenshot-2019-01-17-15.03.27.png Screenshot-2019-01-17-15.04.44.png
Michael Padula (Jan 17 2019 at 21:42):
I believe I was using the Medication.request, Medication, and Substance resources for these.
David Hay (Jan 17 2019 at 22:06):
anything in the console?
David Hay (Jan 17 2019 at 22:57):
So there is an issue in the json builder - and it can't handle Condition.stage.assessment . (it can handle Condition.stage.summary). This is one of those bugs that needed someone to actually try to the this element to find it. It's not going to be straightforward to fix (so it won't be done by tomorrow) but I've nixed the alert and instead just placed a message in the json for me to find later - so shouldn't interrupt your workflow too much...
But the tree view still works - which I suspect will be more useful to you - and you can continue adding resources and updating them (it's just that the json won't always be correct)
I'll be around for the next few hours (your evening) but won' be on-line tomorrow morning I'm afraid. Do pose issues here though - I will get to them when I wake (though that;'s about midday for you I'm afraid)...
I'm hopeful that the error about serializing the graph was resolved with the other fix - I can't duplicate that error (even with a copy of your graph - I used the graph import to help with debugging...)
Last updated: Apr 12 2022 at 19:14 UTC
