You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Presently we have intermittent failures to send API usage notifications to users. Though theoretically a threshold could be skipped if a user has a large amount of traffic within a 12 hour window (which is how often the task runs) we have far too many people that skip threshold and an investigation of their API usage through the charts that connect to InfluxDB has shown us many examples of failed delivery.
For an incomplete list of organisations with failures consider the following query that can be run on Metabase:
SELECT"public"."organisations_organisationapiusagenotification"."id"AS"id",
"public"."organisations_organisationapiusagenotification"."percent_usage"AS"percent_usage",
"public"."organisations_organisationapiusagenotification"."notified_at"AS"notified_at",
"public"."organisations_organisationapiusagenotification"."created_at"AS"created_at",
"public"."organisations_organisationapiusagenotification"."updated_at"AS"updated_at",
"public"."organisations_organisationapiusagenotification"."organisation_id"AS"organisation_id"FROM"public"."organisations_organisationapiusagenotification"WHERE organisation_id NOT IN (SELECT organisation_id FROM"public"."organisations_organisationapiusagenotification"WHERE percent_usage =75)
ORDER BY"public"."organisations_organisationapiusagenotification"."organisation_id"DESC
Locations of the Code
See the handle_api_usage_notifications task in the organisations/tasks.py module and the handle_api_usage_notification_for_organisation function under the organisations/task_helpers.py module. Both of these modules have been carefully gone through by hand including pulling logs from AWS but the location of the offending code has yet to be found.
We have also gone through the task status and verified that the task runs are not timing out.
The Goal
The goal to closing this ticket is to figure out why the smaller thresholds are not firing correctly. Since API usage notifications run many systems which include billing customers automatically for API usage overages, it is important to get to the bottom of this.
The text was updated successfully, but these errors were encountered:
Presently we have intermittent failures to send API usage notifications to users. Though theoretically a threshold could be skipped if a user has a large amount of traffic within a 12 hour window (which is how often the task runs) we have far too many people that skip threshold and an investigation of their API usage through the charts that connect to InfluxDB has shown us many examples of failed delivery.
For an incomplete list of organisations with failures consider the following query that can be run on Metabase:
Locations of the Code
See the
handle_api_usage_notifications
task in theorganisations/tasks.py
module and thehandle_api_usage_notification_for_organisation
function under theorganisations/task_helpers.py
module. Both of these modules have been carefully gone through by hand including pulling logs from AWS but the location of the offending code has yet to be found.We have also gone through the task status and verified that the task runs are not timing out.
The Goal
The goal to closing this ticket is to figure out why the smaller thresholds are not firing correctly. Since API usage notifications run many systems which include billing customers automatically for API usage overages, it is important to get to the bottom of this.
The text was updated successfully, but these errors were encountered: