Internal service error from resource store

[Paul Brewster]

I'm getting the following error attempting to get binary resources that are pointed to in Document References.

https://fhir-ehr.sandboxcerner.com/dstu2/0b8a0111-e8e6-4c26-a91c-5069cbc6b1ca/Binary/XR-9859359

 Internal service error from resource store

[Benjamin Eichhorn (Cerner)]

Hi Paul,

We're aware of an issue with the Binary service at the moment in our public sandbox and are working on a fix to restore service.

Thank you for your patience,

Ben (Cerner)

[Paul Brewster]

Could you please get back to me when the issue is fixed.  Thanks

[Benjamin Eichhorn (Cerner)]

Hi Paul,

Yes, we will post back here once we have resolved this issue.

Thanks again for your patience,

Ben (Cerner)

[Benjamin Eichhorn (Cerner)]

Hi Paul,

Service has been restored to the Binary resource. Let us know if you have any further issues.

Thanks,

Ben (Cerner)

[Paul Brewster]

I'm getting odd responses from the Binary resource fetch

Two back to back pulls from /Binary/ for the same resource are returning different results.  See attached files.  One produces a good pdf and the other doesn't. I haven't been able to see a pattern. I have a list of Document references and I give the user a choice of which one to see and right now I can't predict which ones will return a good pdf file.

"https://fhir-ehr.sandboxcerner.com/dstu2/0b8a0111-e8e6-4c26-a91c-5069cbc6b1ca/Binary/XR-9867295"

[Benjamin Eichhorn (Cerner)]

Hi Paul,

Could you please clarify the issues that you are seeing? I was able to decode both of the base64 strings you provided into a PDF without any issues.

Thanks,

Ben (Cerner)

[Paul Brewster]

Why would they be different?  I’m not updating the Binary Resource.  Here’s what I get in the browser

 

--
You received this message because you are subscribed to the Google Groups "Cerner FHIR Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cerner-fhir-devel...@googlegroups.com.
To post to this group, send email to cerner-fhir...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cerner-fhir-developers/220a81f9-60e8-4ed6-9d62-803f21537c5b%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Paul Brewster]

Do we have any idea why I'm getting different pdf contents on successive requests for the same BinaryResource?  Do we have a configuration problem on a pair of load balanced servers?

[Benjamin Eichhorn (Cerner)]

Hi Paul,

I'm not sure on why you are unable to decode the base64 PDF. It might be an issue with the library you are using as we've successfully decoded the base64 strings you provided and the ones returned from the Binary endpoint. When calling the Binary endpoint, documents are generated on demand. The base64 encodings are different because the printed time, located at the bottom of the document, has changed. In one instance 13:01 and another 13:02. In addition it's also possible that if you were to grab the document today, and then a day later, that additional content or changes to documents were made, then the document generated would also have different content.

Thanks,

Ben (Cerner)

[Paul Brewster]

Is there a way to get back the XHTML that I posted to the Document Reference and not the pdf?

On Tuesday, February 5, 2019 at 5:33:39 PM UTC-5, Paul Brewster wrote:

[Benjamin Eichhorn (Cerner)]

Hi Paul,

No. The XHTML you post cannot be retrieved once posted. It can only be retrieved as a PDF.

Thanks,

Ben (Cerner)

[kevin maloy]

Interestingly only returning pdf might need to change if you look at proposed rule from ONC that dropped yesterday at https://www.healthit.gov/topic/laws-regulation-and-policy/notice-proposed-rulemaking-improve-interoperability-health

Page 217 ... "We also clarify that the clinical note text included in this FHIR resource [DocumentReference] would need to be represented in its “raw” text form. In other words, it would be unacceptable for the note text to be converted to another file or format (e.g., .docx, PDF) when it is provided as part of an API response."

Granted, these rule are only enforceable 24 months after final rule (if this is part of the final rule) ... so no time soon ...

--

You received this message because you are subscribed to the Google Groups "Cerner FHIR Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cerner-fhir-devel...@googlegroups.com.
To post to this group, send email to cerner-fhir...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/cerner-fhir-developers/98239e2d-0ae8-413d-9ade-ff89fe1cc0f9%40googlegroups.com.

[Andrew Torres (Cerner)]

How do you feel about it as a client?

[kevin maloy]

Still reading. 

Buts its really hard to use just pdf as a dev. 

Most interesting things you want to do with docref deals with having raw text (like the app we had at the CHC challenge). You can run through pdfjs but clunky and time consuming. 

k. 

On Tue, Feb 12, 2019 at 10:48 PM 'Andrew Torres (Cerner)' via Cerner FHIR Developers <cerner-fhir...@googlegroups.com> wrote:

How do you feel about it as a client?

--
You received this message because you are subscribed to the Google Groups "Cerner FHIR Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cerner-fhir-devel...@googlegroups.com.
To post to this group, send email to cerner-fhir...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/cerner-fhir-developers/48f2d16c-fa2d-4f68-ba68-fdab82691b20%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

mobile.

[Andrew Torres]

Even for patients? No headers, no MRNS, and nothing to indicate which org it comes from.

Sent from my iPhone

You received this message because you are subscribed to a topic in the Google Groups "Cerner FHIR Developers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cerner-fhir-developers/7dPLIXj7Gu0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cerner-fhir-devel...@googlegroups.com.

To post to this group, send email to cerner-fhir...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/cerner-fhir-developers/CAEvSfi-41T%3DKTeLneoPVJNY4Q_SF6-q963qQE7hqM%2B19m_x_%3DA%40mail.gmail.com.

[kevin maloy]

yeah - i guess we agree that there are so many people with interests in this stuff that it is kinda like a rorschah test. 

i personally would rather add the metadata to the text than have the text as a pdf. (kinda like how you cant just display a hematocrit and need to give context about it). results in observation could be pdfs as well (like the old school lab corp results i still see on nursing home patients when i’m working) ... in some cases those pdf/non atomic data are great, sometimes not so great. 

this is obviously just my opinion, but id rather get raw text and attach metadata in a ui. 

k. 

To view this discussion on the web visit https://groups.google.com/d/msgid/cerner-fhir-developers/4C16BE73-761B-4A91-93C3-16BCC0B25D8D%40gmail.com.

For more options, visit https://groups.google.com/d/optout.

--

mobile.