2020-10-05 Meeting notes

• NHSEngland Extract is due. George will run this using the old method along side other tasks. The new submission format is required for 2019 data and it appears they will not want historic data so this will be done once Fran has finished loading the data.
• Renalware are still working on correcting their feed. This will not correct historic data as they will not be re-sending it as their feed is limited to the last two weeks. See under the UKRDC minutes how this will be addressed. Renalware have been asked to populate the external id (using the placerid) and entered at and entering organisation fields for laborders. For observations they have been asked to use the record id for the external id to ensure the record can reflect corrections made.
•  Quick load 
• Fiona is still trying to get suitable passwords from the alphastudy hosters to allow the site to be backed up and the server shut down.
• George has now managed to build the AKI validation code using the bamboo build agent after Tim resolved the ssh and licensing issues. He has given a version to Sarah for test. Sarah has also flagged that the merging of two labs has not worked as expected and needs checking.
• Andrew has started looking at the tracing  code for the UKRDC with the aim of creating some test trace xml files for loading on dev.
• PHE covid matching mark 2 is currently with PHE for matching which should be completed early this week.
• George has completed the Pedw extract and supplied it to Fran for uploading.
• Tim hasn't yet contacted NHSDigital help desk about adding more users to the MESHUI mailbox for sending the opt out data through. They did however respond to the issue raised about no file being returned saying
  "We have identified a number of common issues that cause requests to the service that filters NHS numbers using the National Data Opt out to fail. If none of these apply to you, please respond accordingly.
  If you are submitting via the MESH UI should ensure that the recipient is X26HC036 and that the Local Identifier and Subject entries are not blank
  If you are using the MESH client should ensure that the CTL file contains Data as well as settingX26HC036 and ensuring that the .. and .. entries are not blank
  If you creating a test file from the the sample data at https://digital.nhs.uk/services/national-data-opt-out/compliance-with-the-national-data-opt-out/check-for-national-data-opt-outs-service/format-of-nhs-number-data-file there should be no blank lines between entries and no addition spaces before or after the NHS numbers."
  This didn't answer the issue of files not being returned but since after sending the helpdesk request we received back the file and subsequent usage has worked as expected it is likely the service crashed. The message is included as the issues highlighted we did find when first attempting to use the opt out service.
• No news from the Star registry..
• Opt out has been covered under the MESHUI
• Not aware of any RaDaR tickets being worked on.
• No new Kings or Barts Info 

PHE bacterium will wait until 2019 loading is complete probably sometime next week.

George highlighted he is having issues sshing to the ukrdc servers and port 7999 of the Bitbucket server.

QAS software is causing issues for the Data managers. This software dates from 1997 and doesn't seem to support modern windows 10 screen scaling which seems to be part of the issue. Tim has got an installer for the new batch processing software which will need configuring and testing. Currently it flags a potential match problem when selecting the gbr dataset that it won't match unless it can identify the country. When trying to match the example file supplied it fails to match any addresses - one assumes because there is no country field. A question has been posed to experian to clarify what this means. It is possible that the configuration was not done correctly too.

We now have the higher spec desktops for the dev and stats team. George's machine is NBT26008, Andrew's is  NBT26006 they are both online to allow settings to be migrated from the old desktop and development environments configured. Once done the old machines will be removed.

AKI extract this should have been done last week and needs sorting but not at the cost of getting the HL7 messaging done.