2026-05-29

  • PfP cert update: 26 days to complete non-proxygen testing!
    • check with Danielle next week
  • PfP CDK
    • adopt new delete stack code
    • splunk reports need updating, change requested of Ant
  • 6515:
    • PR 3126:
      • turned out to be a huge rewrite
        • update for private repos
        • reorder stateless second
        • dodgy copilot manipulation of numbers (30->29, 731->730)
        • remove sam stack name passed to stateless (copilot?)
        • add run regression tests
        • add deploy apigee

2026-05-28

  • complete investigation of cancel prescription
  • psu cdk
    • had to close 2969 due to secret
    • PR 3126: prepped empty stack pr, nearly ready, self-review tomorrow
    • PR 3127: prepped stateful only pr, possibly already broken

2026-05-27

  • investigation of cancel prescription for Jack
    • key logging seems to be “Creating Spine payload from FHIR resource” factory.ts line 87 in FHIR facade.
    • however the only thing it logs is bundle identifier, when the request is a cancellation.
    • Phil has been working on ability to log full request-response, Ant recently confirmed on a rubber duck that we don’t have it ATM.
    • next: investigate spine side logs
    • did that too, no sign, asked if they have internalID
  • pfp cdk
  • psu cdk
    • failed quality checks x2, at least one seems to be intermittent

2026-05-26

  • pfp cdk
    • copilot-written PR was ugly
    • Jim had rewritten more elegantly
    • agreed with Matt to remove skip all if active in last 24h
    • the second issue seems to have been a temp issue?
      • maybe proxygen deployment failures misled us to think first in day was considered ‘active’?
  • submit CareId app’n
  • update logging page for new PfP CDK logs
  • update PSU CDK from main

2026-05-25: bank holiday

2026-05-22: vacation

2026-05-21

  • pfp cdk
    • not deleting old stacks
    • pretty well all day investigating with Jim
    • Ant vetoed merging an adhoc clean up for main stacks (even temporarily, so Jim did manually)
    • also reworked cdk util for deletion because:
      • apparently selected wrong one as ‘active’ (first in day)
      • due to there always being an active deployment in last 24h, never deleting anything

2026-05-20

  • meeting day: show n tell, digimeds, planning
  • pfp cdk
    • cleanup workflows
    • move SnsAlarm to cdk-constructs
  • chat w Pete about zero downtime

2026-05-19

  • release cut done by Jim
  • PfP CDK passed in qa, used in release cut
  • all repo workspace and start scripts for branch and PR everywhere
  • prep for show n tell (2h on animation)

2026-05-18

  • pfp cdk
    • fix build (proxygen needs to deploy to versioned url now)
    • fix delete old PR stacks
    • hand over to Kayal for testing
  • publish rg in base images for Copilot’s benefit

2026-05-15

  • cdk
    • merged, fail on delete main stacks
    • move dependabot up so we can see if it works rather than face a backlog
    • TODO
      • delete main stacks should not need github token, but left in for now.
      • The existing CDK synth coverage only exercises the main API stack with mutualTlsTrustStoreKey set to undefined, so this mTLS-specific serviceName wiring is not asserted for the enabled path. Add a synth test that sets a trust store key and verifies the resulting mTLS/domain configuration so regressions in the new construct input are caught.
      • delete old prs stacks should tee output so see progress

2026-05-14

  • disabling notifications for single pharmacy
  • cdk
    • NOTE for PSU: after deleting SAM stack, need to disable workflows on PR as well as main
    • proxygen deployment still failing
      Error: RROR]: Read timeout on endpoint URL: "https://lambda.eu-west-2.amazonaws.com/2015-03-31/functions/lambda-resources-ProxygenPTLInstancePut/invocations"
     Error: Process completed with exit code 255.
    - Ant said "there were 2 internal dev proxies for pull requests of deployments prescriptions-for-patients-v2 that were not deployed which causes this issue. have deployed them in apigee so it should work now"
- Resolved
- Jim says retain policy was a mistake, need to understand why
  • pushed re-enablement of stacey test, await symptom of failure

2026-05-13

  • cdk
    • merged PR to add retain policy from SAM
    • issue with new delete cdk stacks
      • Jim looking at it
        • reviewed and merged
    • rebase PR to replace SAM with CDK, proxygen deploy not working
  • fixed PSU regression test
  • add newly requested pharmacy notification block

2026-05-12

  • dependabot backlog
    • need to delete cdk stacks at least in PR (both pfp and psu)
    • did manually in the end
    • merged delete cdk stacks
  • cdk
    • avoid clash on spec name when uploading
    • ready to merge first cdk PR

2026-05-11

  • meetings: duck, planning
  • dependabot tidy
  • broken ‘Stacey’ test

2026-05-07 & 08: Leave

2026-05-06

  • ceremonies
  • continue pairing with Jim

2026-05-05

  • started pairing with Jim on getting PfP CDK merged
    1. PR to release ownership from SAM

      • delete stack aws cloudformation delete-stack –stack-name pfp-pr-2627

      • check if sub-stacks still exist - yes

      • check resources still exist

    2. Comparision of CF JSON from SAM and CDK -> not much
    3. No manual aws cli required as stateless
    4. Merge CDK PR

2026-05-04: public holiday

2026-05-01

  • review for Jim that proved a lot harder than it ought, due to white space changes
  • PSU cdk
    • pitiful progress
    • migration script for secrets and params
    • ultimately a successful standalone stateful+stateless deployment on PR

2026-04-30

  • talk with Ant about yesterday’s account resources secret PR
  • 6515
    • update existing PR for time passing
    • can I
      • get both stateful and stateless stack working?
        • inc. reg tests passing
      • demonstrate small / no difference between sam and cdk
      • create pr to relinquish management of stateful from sam
      • create pr + manual script to adopt mgmt by cdk (probably manual script then pr merge)