2020-04-20 Meeting notes

• George Swinnerton has fixed the final NHS number issue raised by Shaun.
• The OPTEPro fix to prevent empty survey files being sent back to PV  is par tof a set of fixes including datetime issue and medication unknown date for medication start date. These are now working but deployment is delayed to allow incorporation of covid 19 schema changes and code to prevent covid 19 results being sent to PV.
• The NHS Extract is now going to be returned using a separate study id and not the RR number as originally done. George Swinnerton will adapt his code this week to use a study Id instead.
• James hasn't yet managed to make much progress on the automatic tracing of UKRDC patients. Now the covid-19 code is nearing completion some progress can be made on this.
• The default tomcat servers on PV live and staging have now been disabled so will not auto start. This change will remain in place until the EPEL version of tomcat7 is updated to version 103 which fixes the issue that causes PV not to start with the current version. Currently if the servers are rebooted the temporary tomcat versions will need to be manually restarted.
• George is waiting on Tim at renalware to sort the ODS 0000 issue and NHS numbers with spaces in. They are now ready to send through covid19 data but this will not be enabled until the current deployment of updates has been done to ensure that the results don't get sent to PV.  The issue of the spaces getting through is because the validation ignores spaces but doesn't modify the value so a valid NHS number with spaces is passed through. Ideally we should reject files with NHS numbers with spaces in however it would be preferable that initially the files are processed with the spaces removed but a warning is sent to the unit giving them a finite time to resolve the issue. If they don't then the files get rejected until they do. It also looks like some files have been stored with invalid (>10 character) NHSnumbers. George will investigate and resolve. The UI is unable to display the data for these records so this also needs investigating.
• Andrew has had a look at the bleach library as flagged by github as having a vulnerability. He wasn't able to see how it would impact RaDaR. He wasn't aware of the other library warnings. Tim has forwarded the most recent warnings through for further investigation. In addition the results of the Penetration Tests need to be reviewed and raised vulnerabilities checked and mitigated if required. The latest results are in   R:\0 Systems\Projects\Pen Tests\2020 N3 Testing  and R:\0 Systems\Projects\Pen Tests\2020 Internet 
• George has updated the schema for PV to include covid 19 reporting (covid19pcr, covid19antibody) this will need deploying as part of the latest release. Documentation has been added Result / Observation Codes (UKRDC Extract) and /wiki/spaces/ST/pages/1426226510.
• George has looked at the issue raised by Shaun with UKT matching and raised a ticket to fix the issue. He is confident that no adverse issues will be caused by the problem in the mean time and it will be properly flagged.
• The NHSBT file processing hasn't been completed due to a field size issue on CAUSE_OF_FAILURE_TEXT which is too small for one of the latest entries and needs increasing to accommodate the field size. Once that is done the file will be completed.
• Andrew has some fixes for the syncing issues with ukrdc outgoing data ie data to RaDaR and RDR-624 however Fioan has now requested some extra changes for this ticket so this will be done before release. Alongside Andrew will check the incoming syncing issue of RWP50 which should map to RRK02 in the code_map table this should fix the weekly errors for radar imports to ukrdc. Hopefully this should all be resolved this week. This should then allow work on the Demo update to progress. Once the issue around creating a pg_dump of the demo database has been resolved.
• James has continued to refine the covid 19 weekly reporting code making the automated validation and creation of the tracing file. There are still a few issues with date variations and one site that is sending a password protected spreadsheet which ideally would be saved unlocked to the latest folder. The latest folder is scanned every 15 minutes but currently doesn't use pyinotify type libraries. There is still work to be done on the the scheduled creation of the tracing file and the automatic detection of the returned traced file and the creation of the final output file. In addition the sqllite master index needs to be completed This should allow quick lookups of the complete submitted data set and if augmented with tracing success/failure and radar memberships could provide lists for checking/ reporting for radar. 

Thwate is being fully absorbed by DigiCert and migrating all Thwate users to the DigiCert console. This needs to be done by the 30 Apr 2020 Currently the emails for this only go to George. When we transition we need to ensure the notifications are sent to rrsystems so all are aware.