You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ambari.apache.org by "Dmytro Sen (JIRA)" <ji...@apache.org> on 2018/05/07 14:05:00 UTC

[jira] [Created] (AMBARI-23780) Implement an upgrade helper in Ambari Metrics collector that copies data from the old schema to the new.

Dmytro Sen created AMBARI-23780:
-----------------------------------

             Summary: Implement an upgrade helper in Ambari Metrics collector that copies data from the old schema to the new. 
                 Key: AMBARI-23780
                 URL: https://issues.apache.org/jira/browse/AMBARI-23780
             Project: Ambari
          Issue Type: Bug
          Components: ambari-metrics
    Affects Versions: 2.7.0
            Reporter: Dmytro Sen
            Assignee: Dmytro Sen
             Fix For: 2.7.0


* The upgrade helper can be a main class that is callable through command line.
* Inputs to the upgrade helper
    # Metric list file (Default -> metrics_whitelist file in conf. If not present, copy over all the metrics)
    # Start timestamp (Default -> Last 1 month) (Max 8640 points per metric in minute table)
* The upgrade helper runs a set of SQL queries to fetch data from all aggregate tables except METRIC_AGGREGATE and uses the API to write to the corresponding new tables.
* The tool marks the completion of every metric entry by using a marker file (list of processed metric names) or a simple hbase table.
* The next time the upgrade tool is invoked, the marker is used to identify whether the metric needs to be copied.
* Let's have the upgrade helper log the status to a different file than collector log as a mechanism to monitor the status.

We have to make a decision on whether the upgrade tool helper should be *synchronous* or *asynchronous*. If it is asynchronous, we should add the ability for it to '*stop performing upgrade*'. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)