/
2021-01-25 Meeting notes

2021-01-25 Meeting notes

Owned by Tim Whitlock (Deactivated)
Last updated: Mar 02, 2021 by George Swinnerton

Date

Attendees

Goals

  • Review out standing tasks progress

Discussion items

ItemWhoNotes
Outstanding TasksTim Whitlock (Deactivated)
  • No progress on the xml tracing code
  • George hasn't chased renalware this week but will today. It is especially important if we want to get the schema changes for dialysis done now Nick is progressing code.
  • Fran has started reviewing the latest validation release and has confirmed a couple of changes are working but has just hit an issue with svn when validating a file. The SVN code had to be updated in the move to python3 codebase which probably explains the issues. George will look at this after the meeting. There is a validation meeting this week where this can be discussed further.
  • Tim took a look at the Atlassian VM which was only using a quarter of available ram suggesting the memory issues may be due to the jvm config rather than the available ram.
  • George has been talking directly with the support people about the alphastudy database dump required and is hoping they now know what he would like.
  • Andre has started looking at the reloading of the AKI files using a local instance. The dump of the site is 100GB.
  • Andre has completed the first set of changes for transplantfirst and put the code live. She has also documented the way the graphing routines work on the site so their plotting can be better understood. This documentation will be put on confluence
  • Manuella has now got some Welsh patients that need adding to the Nephwork site which she will load when she receives the file.
  • Andrew has done a number of tickets this week however has found that several he thought were completed still need some work doing on them.
  • Andrew has received the latest NHSBT file to work on so will be progressing it this week. Our contact has now changed at NHSBT to Rebecca Curtis.
  • Andrew also needs to make some minor changes to the covid data importer. Currently the process requires a valid date for a positive test before a file will load. This has meant artificial dates have been added to the files to allow them to load unfortunately these dates are recent valid dates. The code will be changed to accept 1/1/1900 as the unknown date value to prevent any confusion. Also currently the validation allows any date from the beginning of 2020 for the positive test date meaning mistyped 2021 dates don't get flagged so it is going to change to a rolling year date range. 
 AOB
  •  Joel reminded everyone that pip no longer supports python 2 and you will need to pin pip to be able to use it with python2. This is likely to be an increasing problem and we need to ensure we phase out our python2 codebases.
  • Tim updated people on the fact that the web client used for managing our AIMES vms is base don flash which reached EOL in December meaning it will no longer work. The current work around is to use the Client App which doesn't support all the new features but will work. To move to the HTML5 version of the client the vmware needs to be upgraded. Why AIMEs has not done this is not known. 
  • Tim made people aware that NBT have indicated there decreasing enthusiasm for keeping us part of the hospital network and that they would prefer us to separate. This then opens up the question of what sort of infrastructure we should look to implement. What issues in the current structure could be addressed and we should ensure we allow for etc. Joel will circulate some of his thoughts later in the week.  
  • Joel raised the issue of not following the development guidelines like ensuring all code is formatted using black. He will circulate some revised advice.
  • Over the next couple of weeks the Internet and N3 servers will be penetration tested. When we get the report some action may be required to correct any issues found.

Action items