Skip to main content
Version: 0.15

Logging

Common Fate sends all logs to Amazon CloudWatch. Common Fate uses a JSON structured log format. An example of a log message is shown below.

{
  "caller": "logger/middleware.go:63",
  "level": "info",
  "method": "GET",
  "msg": "Served",
  "proto": "HTTP/1.1",
  "remote": "1.2.3.4",
  "reqId": "1e095537-6a72-4e45-9121-863e91d86c6d",
  "request": "/api/v1/users/me",
  "size": 199,
  "status": 200,
  "took": 0.65133094,
  "ts": 1657135399.0898278,
  "userId": "usr_2BWUDH5etXYQB8Eybs2RJZ7c0wK"
}

The userId field corresponds to an authenticated user in Common Fate. The reqId field is populated from the AWS Lambda request ID.

You can use gdeploy to query for logs. Under the hood, gdeploy calls CloudWatch APIs and prints the messages to your terminal.

Getting logs for all services​

To retrieve logs for all services, run:

gdeploy logs get

You should see an output similar to the below.

➜ gdeploy logs get
[i] starting to watch logs for api, log group id: /aws/lambda/Granted-APIRestAPIHandlerFunction6F0D5931-Tdrmgxjr7kZa
[i] starting to watch logs for accesshandler, log group id: /aws/lambda/Granted-AccessHandlerRestAPIHandlerFunction9FEF0B2-nbs5EVKZdQ8h
[i] starting to watch logs for idp-sync, log group id: /aws/lambda/Granted-APIIdpSyncHandlerFunctionB5E49815-YHckG94aahEX
[i] starting to watch logs for granter, log group id: /aws/lambda/Granted-AccessHandlerGranterStepHandlerFunction149-SPtI0RFiRHl6
[i] starting to watch logs for events, log group id: Granted-EventBusEventBusLogFAEB6547-Je81BRwcDgOu
[i] starting to watch logs for slack-notifer, log group id: /aws/lambda/Granted-APINotifiersSlackNotifierFunction3528BD0B-vCiYi0yfhmVX
[i] starting to watch logs for event-handler, log group id: /aws/lambda/Granted-APIEventHandlerFunctionD526CC8B-Gi3feS7zQjfn
[2022-07-06T20:25:34+01:00] (2022/07/06/[$LATEST]186ff953f07f4bc5b941b47e801e2d09) START RequestId: ce6afaae-020d-4f8c-bd7f-02443ad407da Version: $LATEST

[2022-07-06T20:25:34+01:00] (2022/07/06/[$LATEST]9474ab7678d94c73a0c1bd08c0a20e81) START RequestId: 7a57ed49-5905-4574-b28c-809e372f621b Version: $LATEST

[2022-07-06T20:25:34+01:00] (2022/07/06/[$LATEST]186ff953f07f4bc5b941b47e801e2d09) { "caller": "logger/middleware.go:63", "level": "info", "method": "GET", "msg": "Served", "proto": "HTTP/1.1", "remote": "1.2.3.4", "reqId": "ce6afaae-020d-4f8c-bd7f-02443ad407da", "request": "/api/v1/access-rules", "size": 30, "status": 200, "took": 0.123400793, "ts": 1657135534.8757534, "userId": "usr_2BWUDH5etXYQB8Eybs2RJZ7c0wK" }
[2022-07-06T20:25:34+01:00] (2022/07/06/[$LATEST]186ff953f07f4bc5b941b47e801e2d09) END RequestId: ce6afaae-020d-4f8c-bd7f-02443ad407da
...

By default logs from the last 5 minutes are loaded. To change the start and end times, provide them as arguments:

gdeploy logs get --start -15m --end -5m # retrieves logs from 15 minutes ago, until 5 minutes ago

Getting logs for a particular service​

You can filter for a particular service by providing the --service flag:

gdeploy logs get --service api

Watching logs​

You can live tail the logs by running:

gdeploy logs watch