Consumption Analytics Documentation

Home > Documentation for older versions > Cloud Cruiser 3 > Setting Up Collection > Native collectors > Microsoft Windows Azure Pack > Recovering after a period with no collection

Recovering after a period with no collection

Table of contents
No headers

Because the load job maintains a list of active plans and subscriptions by adding and removing items based on each day’s collected events, collection of Windows Azure Pack must run continuously to be successful. When this job doesn’t run successfully for a day or more, subscriptions and plans created during that period are never added to the list. Simply resuming scheduled execution produces warnings because some usage can’t be associated with a subscription or plan, and charges for this usage are not loaded.

Before you begin

Fix any outage or error that was causing collection to fail.

To recover after a gap in collection

  1. Go to Data > Job Loads.
  2. Find the most recent load group for the azure_pack process and note its Accounting Date.
    This is the last date for which your jobs ran successfully. For information about using the Job Loads window, see Tracking job data loads.
  3. Go to Data > Job Maintenance and select the azure_pack_collect job.
  4. Set the Start Date to yesterday’s date and click Run batch job.
    The collect job collects all usage and events since the last time it successfully ran, storing them in files based on the date they ended.
  5. Select the azure_pack_daily_load job, set the Start Date to the day after the most recent load group you noted, and leave the End Date set to yesterday.
  6. Click Run batch job.
    Because the End Date is still set to yesterday, the job runs against a range of dates. It does this with separate runs for each date from earliest to latest, allowing the state file to roll over one day at a time.

Charges for the gap period are now viewable, and you can resume scheduled nightly execution of both jobs.

Last modified


This page has no custom tags.


This page has no classifications.

 (c) Copyright 2017-2020 Hewlett Packard Enterprise Development LP