MAAS

  1. Bug #1354511
  2. Comment #2

Comment 2 for bug 1354511

Revision history for this message
Jeroen T. Vermeulen (jtv) wrote :

Found it! The database cluster in my trunk branch was still on UTC+1. Setting it up afresh made the test pass.

This does reveal a bug in the API code, however: it shouldn't be mixing its client-side clock with the database clock. It'd be great if it could just issue the SQL expression “now() - <interval>” instead of computing the value itself.