ElasticSearch-based sink for Microsoft Application Insights.
For now, you have to build the .NET Core project. Next thing you do, is add the following contents to the appsettings.json file:
"ElasticSearchUrls": ["http://<address of elasticsearch node 1>:9200/"],
"OutputIndexNameFormat": "twla-{tenant-id}-{datetime}-{message-type}", /* The format of indexes to use in elasticsearch */
"OutputIndexNameDateFormat": "yyyy-MM",
"TenantListIndexName": "twla-config" /* the elasticsearch index to use to store the tenants */
Multiple node addresses can be specified.
Now you can start the server process. This will start listening on all IP addresses at port 5000
.
Now you need to add known instrumentation keys. Do so by POST
ing the following json document to http://twla:5000/ where twla is the address of the machine:
{
"identifier": "<instrumentation key>",
"name": "<descriptive name>"
}
The identifier is the exact instrumentation key to use in the Client SDK configuration. The name is a descriptive name, which is not yet used anywhere.
Now you have to restart the service, to make the changes effect. (This is a known limitation, to be resolved at some point. See #1)
Now the service is usable, and is able to process Application Insight data.
Copyright (c) 2018, ter Woord Computers, Germany.
For details, see license.md