Wait time skipped for the 1st and 2nd steps of generator type task
Bug #1218187 reported by
Winson Chan
This bug affects 1 person
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack Heat |
Fix Released
|
Low
|
Unmesh Gurjar |
Bug Description
For a generator type task, the __call__ method of scheduler.
Changed in heat: | |
importance: | Undecided → Low |
Changed in heat: | |
status: | New → Triaged |
Changed in heat: | |
assignee: | nobody → Unmesh Gurjar (unmesh-gurjar) |
Changed in heat: | |
milestone: | none → juno-3 |
status: | Fix Committed → Fix Released |
Changed in heat: | |
milestone: | juno-3 → 2014.2 |
To post a comment you must log in.
I guess you could consider this a bug. IIRC I wrote it this way to minimise the changes to the number of _sleep() calls to mock in the unit tests. That problem has largely gone away thanks to ENABLE_SLEEP though, so it's worth reconsidering.
The way it works now means that when we e.g. delete something we check immediately to see if it is gone, rather than wait another second. So if it turned out that potentially slow operations were actually pretty quick in most cases, then this could slow things down. I suspect it's probably not a big deal though.