Sprint 36 Review (2017-11-15)

Production Application releases / updates:

Product Management

Courses

• BM's agreed to 2 options
  • look into re-procurement of Intrepid CBMS module
  • TIS team to work closely with them to procure a new tool

Approvals

• BM's agreed to inputting minimal fields for MVP
• Scenarios & field validation in Confluence updated to reflect newly defined MVP

• BM's requested that re-procurement for Intrepid is completed due to complexity of handling
• option to manually input trainee entitlements and actuals (budget & time) was provided and will still be provided as part of MVP
• Study leave MVP Scenarios - Absence Management - MVP Scenarios

Case Management

• BM's agreed to using either sharepoint (they would manage), or concerns log as curretly implemented
• using concerns log as implemented would require tweaks

Questionnaire (Alistair Pringle (Unlicensed) Joanne Watson (Unlicensed))

• Discussion with Kata Varnai took place outcome of which is Questionnaire is not going to be implemented as part of the Horus E-portfolio system.

• NETS proposed as Online survey tool – West midlands/Bristol would be the pilot sites.

Admin UM - Password Management

• stories created and refined in TIS backlog

Reporting

• Discussion opened up with BM's. Proposed solutions overview run through. Outcome of which are:
  
  • Data analysts group – Process component should be defined by the business on how to drive the data output
  • Local Offices to be involved with Data analysts group to ensure their current reporting tools/standards still work when TIS go live
  • Local Offices to provide examples of reports currently used that could be created by Kais and/or JT to test the validity of the proposed plan.
• High level overview of data warehouse presented by Kais - Data marts in the data warehouse will need to have data refreshed more often if not real time.

https://hee-tis.atlassian.net/wiki/x/JgCKCQ

Revised mega menu/ navigation based on review

Menu: With current low Navigation L2 item content the mega menu is not working creating a disconnection between parent and children navigation items and excessive mouse travel. We need to either 1/ move content to sit under relevant menu item NOT justified to the left in the existing mega menu band or 2/ make individual menu drop downs. Menu can always switch back to original mega menu when content grows.

Menu Indicators:
Add Page indicator to L2 items
Subtitles for MM Admin UI/Admin

Looking at not available L2 items

Trainee Menu:
Implement page highlight
New back button on L2 mobile menu

https://hee-tis.atlassian.net/wiki/x/GIB5CQ

• Active Post with NULL ManagingDeanery values now addresses by the Data Managers, outstanding on the Trust ManagingDeanery data cleanse
• Query created to identify which Trainee is owned by which LocalOfffice How to Identify a which Local Office owns a Person from the Intrepid DR
• Query created to identify which LocalOffice a Trainee has belonged too People - Data Analysis | Trainee to Local Offices

People

• Form validation on all required fields now set to trigger on form load. This now applies to all forms currently in TCS with a 'Create'  method (.i.e Programme, Curricula).

Extras

Existing E2E tests on the Admin UI improved and issues they identify resolved, still more to work through

• Upgraded the Mirror hardware which was limiting the ingest performance

Graham O'Regan (Unlicensed)

Russell Steel (Unlicensed)

• waiting on Hicom to relase the following DRv2 fixes;
  vwCPPSPsychologicalReportOverview - currently only contains 1 column;
  
• vwCCPSCaseClose. CaseCloseReason field should be displaying a text reason, but is instead showing a system GUID.

Chris Mills (Unlicensed) 

Paul Hoang (Unlicensed)

Russell Steel (Unlicensed)

• Simplified the ingest code to make it more performant and easier to understand and optimise.
• Fayaz Abdul (Unlicensed) moved the ETLs onto their own machine which allows us to debug seperately the 'out of memory' issue and provide a better space for running such large jobs.
• ETL issue with Posts replicating on each run of the ETL resolved

• ETLs refactored and optimised, have been re-run and Production is now repopulated with data having been empty for a few days, not yet scheduled so last run Monday 13/11

Graham O'Regan (Unlicensed)

Sunil Rochani (Unlicensed)

• Download Inbound xml file, parse it and upload parsed data to TIS

Graham O'Regan (Unlicensed)

Tamer Moustafa (Deactivated)

• Sent Inbound file to ESR, they asked some questions that have been clarified and they have asked for a larger file to test against
• Raised issue around the deanery mappings that ESR use.

• Blue/green changes are ready in development, planning to release over the weekend.
  • (This will enable development to deploy code without massivly affecting the live system and roll back their changes if they see any issues on the seperate instance removing the blocker/possible implications of 'Friday Deploys')
• Separate Elastic search server and ETL server 
  • (Removes any possible issues caused by having a hefty system such as elastic search on the same system as other services. Applications are now split off anything that could affect them increasing resiliancy)
• Migration of Hosted Docker repositority onto Azure Container Registry.
  • (We now no longer need to be running another instance just for our hosted images. They're backed up automatically and managed by Azure, moving more of our infrastructure into Azure's management)
• Added helper scripts to assist creation of docker-compose files from application.yml files.
  • (Speeds up DevOps development and improves onboarding of new Ops staff)
• Added pipeline and files for Oriel ETL and ESR ETL
• Created mock SFTP server for Oriel Testing
• Refactored a number of playbooks for latest Ansible