Athena

Athena is a file processor service, that consumes files stored in the files.com API and runs a series of reports over the downloaded artifacts and subsequently it talks with the Salesforce API for performing actions (currently, only comments are supported)

Basics

There are 3 software components in athena:

Athena-monitor: Monitor changes in several directories across a file.com account and if new files are found those are sent to the processor for background processing.
Nats: Nats is a light messaging daemon that allows a pubsub system to be implemented on top, its used to dispatch messages from athena-monitor to a athena-processor
Athena-processor: Subscribes to messages from monitor and routes the reports that have to be run over a given detected file, subsequently it will perform an action on salesforce (such as posting a comment, etc)

The basic flowchart of interaction is as follows

Configuration

The monitor and the processor can be configured via yaml configuration files by calling

athena-monitor --config config.yaml [--config config2.yaml [...]]
athena-processor --config config.yaml [--config config2.yaml [...]]

where configuration files are loaded in the order they are given on the command line and values in later configuration files superseed values read earlier.

The basic structure looks like

monitor:
  files-delta: 1m
  poll-every: 10s
  base-tmpdir: "/tmp/athena"
  directories:
    - "/uploads"
  processor-map:
    - type: filename
      regex: ".*sosreport.*.tar.[xz|gz]+$"
      processor: sosreports

Monitor Configuration

Processor Configuration

Hacking

In order to stand up a development environment, you will need

make
docker
docker-compose
golang >= 1.19

For running a docker based installation locally you will need a sandbox account on Salesforce and a sandbox directory on files.com. Supply

A list of the corresponding credentials in creds.yaml,

  db:
    dialect: mysql
    dsn: "athena:athena@tcp(db:3306)/athena?charset=utf8&parseTime=true"

  filescom:
    key : "***"
    endpoint: "https://..."

  salesforce:
    endpoint: "https://..."
    username: "***"
    password: "***"
    security-token: "***"

A list of directories to monitor in athena-monitor-directories.yaml,

  monitor:
    directories:
      - "/sandbox/..."
      - "/sandbox/..."

A path for where the report uploads will go in athena-processor-upload.yaml,
```
  processor:
    reports-upload-dir: "/sandbox/..."
```

And finally run

make devel

In case the docker-build step fails you can try to re-run the make command without using the cache,

NOCACHE=1 make devel

The devel deployment includes a debug container which can be used to inspect the database.

$ docker exec --interactive --tty debug bash
# mysql -h db -u athena -pathena athena
mysql> describe files;
+------------+---------------------+------+-----+---------+----------------+
| Field      | Type                | Null | Key | Default | Extra          |
+------------+---------------------+------+-----+---------+----------------+
| id         | bigint(20) unsigned | NO   | PRI | NULL    | auto_increment |
| created_at | datetime(3)         | YES  |     | NULL    |                |
| updated_at | datetime(3)         | YES  |     | NULL    |                |
| deleted_at | datetime(3)         | YES  | MUL | NULL    |                |
| created    | datetime(3)         | YES  |     | NULL    |                |
| dispatched | tinyint(1)          | YES  |     | 0       |                |
| path       | longtext            | YES  |     | NULL    |                |
+------------+---------------------+------+-----+---------+----------------+
7 rows in set (0.01 sec)