diff options
Diffstat (limited to 'docs')
| -rw-r--r-- | docs/logging.md | 41 |
1 files changed, 22 insertions, 19 deletions
diff --git a/docs/logging.md b/docs/logging.md index 3d5ea61..ac2ff25 100644 --- a/docs/logging.md +++ b/docs/logging.md | |||
| @@ -18,25 +18,28 @@ This way we get complete logs also if some resource was not started from the con | |||
| 18 | * warning: Only warnings, should always be logged. | 18 | * warning: Only warnings, should always be logged. |
| 19 | * error: Critical messages that should never appear. Should always be logged. | 19 | * error: Critical messages that should never appear. Should always be logged. |
| 20 | 20 | ||
| 21 | ## Debug areas | 21 | ## Debug areas and components |
| 22 | Debug areas split the code into sections that can be enabled/disabled as one. | 22 | Debug areas and components split the code into sections that can be enabled/disabled as one. This gives finer grained control over what is logged or displayed. |
| 23 | This is supposed to give finer grained control over what is logged or displayed. | 23 | |
| 24 | 24 | Debug areas are the static part, that typically represent a class or file, but give no information about which runtime-component is executing the given code. | |
| 25 | Debug areas may align with classes, but don't have to, the should be made so that they are useful. | 25 | |
| 26 | 26 | Components are the runtime information part that can represent i.e. the resource plugin in the client process or the resource process itself. | |
| 27 | Areas could be: | 27 | |
| 28 | 28 | The full debugging area is then assembled as: "Component.Area" | |
| 29 | * resource.sync.performance | 29 | |
| 30 | * resource.sync | 30 | This can result in identifiers like: |
| 31 | * resource.listener | 31 | |
| 32 | * resource.pipeline | 32 | * $RESOURCE_IDENTIFIER.sync.performance |
| 33 | * resource.store | 33 | * $RESOURCE_IDENTIFIER.sync |
| 34 | * resource.communication | 34 | * $RESOURCE_IDENTIFIER.communication |
| 35 | * client.communication | 35 | * $RESOURCE_IDENTIFIER.pipeline |
| 36 | * client.communication.org.sink.resource.maildir.identifier1 | 36 | * kube.$RESOURCE_IDENTIFIER.communication |
| 37 | * client.queryrunner | 37 | * kube.$RESOURCE_IDENTIFIER.queryrunner |
| 38 | * client.queryrunner.performance | 38 | * kube.actions |
| 39 | * common.typeindex | 39 | |
| 40 | ## Logging guidelines | ||
| 41 | * The trace log level should be used for any information that is not continuously required. | ||
| 42 | * Messages on the Log level should scale. During a sync with 10k messages we don't want 10k messages on the log level, these should go to trace. | ||
| 40 | 43 | ||
| 41 | ## Collected information | 44 | ## Collected information |
| 42 | Additionally to the regular message we want: | 45 | Additionally to the regular message we want: |