Closed Bug 1226337 Opened 10 years ago Closed 10 years ago

EVENTLOOP_UI_{LAG,ACTIVITY}_EXP_MS needs to become an "exceptional case" in histogram.py

Categories

(Cloud Services Graveyard :: Metrics: Pipeline, defect)

defect
Not set
normal

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: chutten, Unassigned)

References

Details

With bug 1198196, we needed to add some special-casing in the server pipeline so that queries against EVENTLOOP_UI_{LAG,ACTIVITY}_EXP_MS histograms don't explode. How it's been described to me: If you ask for a histogram that isn't in the data, we shouldn't fail because not all samples will have all histograms. If you ask for a histogram that _is_ in the data but whose definition is missing, we explode because it ought to be defined someplace. bug 1198196 ran afoul of it by taking away a definition from a version where pings still contained data. Steps to reproduce: 0) Spin up a cluster 1) pings = get_pings(sc, app="Firefox", channel="nightly", version="45.0a1", fraction=0.01) subset = get_pings_properties(pings, ["clientId", "environment/system/os/name", "payload/histograms/EVENTLOOP_UI_ACTIVITY_EXP_MS", "payload/histograms/EVENTLOOP_UI_LAG_EXP_MS", "payload/histograms/GOBBLEDYGOOK", "payload/simpleMeasurements/firstPaint"]) subset.count() It'll throw a "KeyError: 'EVENTLOOP_UI_LAG_EXP_MS'"
I have landed the required changes in new-dev. The current t.m.o dashboard is based on the data produced in the old-dev IAM account though. Once 1220262 is fixed, the dashboard will switch over to the new-dev account.
Depends on: 1220262
No longer depends on: 1228308
Status: NEW → RESOLVED
Closed: 10 years ago
Resolution: --- → FIXED
Product: Cloud Services → Cloud Services Graveyard
You need to log in before you can comment on or make changes to this bug.