EBMS Tickets

What were the symptoms? Error messages? What options did you use? I just imported 32558986 successfully.

On the main import page, I selected Fast Track and gave it a placement level of Passed Abstract Review. I tried uploading a PDF of the article and not and it failed both times. There isn't an error message - it's just listed under ARTICLES WITH ERRORS at the bottom of the page. I'll upload a screenshot.

I just searched for the article you imported and I'm not able to find it in the system. I'm on DEV2.

Hmm. Can you import on any other tier? I'm wondering if the problem might not be on our end.

$ curl 'https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=pubmed&rettype=medline&retmode=xml&id=32558986'{{  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current}}{{                                 Dload  Upload   Total   Spent    Left  Speed}}100   206    0   206    0     0    206      0 -::- -::- -::-   549<?xml version="1.0" ?><!DOCTYPE PubmedArticleSet PUBLIC "-//NLM//DTD PubMedArticle, 1st January 2019//EN" "https://dtd.nlm.nih.gov/ncbi/pubmed/out/pubmed_190101.dtd"><PubmedArticleSet></PubmedArticleSet>

I have reported the problem to NLM. I see this as good news. I'd rather have the NLM's site broken than have the security or Linux upgrade break the EBMS.

Well this is strange. I was also unable to import 32558986 on QA so I went and found a legit article to try to import on PROD (32534647) and this was successful. Maybe there's an issue with the test citations I was using? Something the EBMS didn't like about those articles? Or some characteristic on NLM's end?

Heard back from NLM, saying they can't reproduce the problem. Please give it another try.

I just successfully imported each of these citations on DEV2. I guess it must have been a temporary issue that resolved itself? In any case, it's good news. Thanks.

Yes, it is good news. Unfortunately, I never succeeded in conveying to the contact at NLM what was going on. She kept insisting that it was just an occasional blip and all we had to do was rewrite our software to try again when it ran into a failure, and I kept trying to explain that I was trying repeatedly and consistently getting successful retrieval for one set of articles and consistently encountering failures for another set of articles. I gave up trying to explain it.