Skip to content

Commit 3346b0a

Browse files
authored
Fixed typos
1 parent 0f0fee3 commit 3346b0a

File tree

1 file changed

+4
-4
lines changed

1 file changed

+4
-4
lines changed

docs/how-does-the-new-detection-work.md

Lines changed: 4 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -5,9 +5,9 @@ description: Learn about the New Detection feature on elmah.io to help identify
55

66
# How does the new detection work
77

8-
Being able to identify when a logged error is new or a duplicate of an already logged error, is one of the most important features on elmah.io. A lot of other features are based on this mechanism to help you reduce the number of emails, Slack/Teams messages, and much more. We often get questions about how this works and how to tweak it, why this article should bring some clarity to questions about this feature.
8+
Being able to identify when a logged error is new or a duplicate of an already logged error is one of the most important features on elmah.io. Many other features are based on this mechanism to help you reduce the number of emails, Slack/Teams messages, and more. We often receive questions about how this works and how to tweak it. This article aims to bring some clarity to questions about this feature.
99

10-
When logging messages to elmah.io using either one of the integrations or through the API, we automatically set a flag named `isNew` on each log message. Calculating the value of this field is based on a rather complex algorithm. The implementation is closed-source but not a huge secret. Each message is assigned a hash value based on a range of fields like the message template, the severity, the URL, and more. Some values are normalized or modified before being sent as input to the hash function. An example of this is removing numbers from the log message which will ensure that `Error on product 1` and `Error on product 2` will be considered the same log message. When receiving a new log message, we check if an existing message with the same hash is already stored in the log. If not, the new message will be marked as *New* by setting the `isNew` flag to `true`. If we already found one or more log messages with the same hash, the new message will have its `isNew` flag set to `false`.
10+
When logging messages to elmah.io using either one of the integrations or through the API, we automatically set a flag named `isNew` on each log message. Calculating the value of this field is based on a rather complex algorithm. The implementation is closed-source but not a huge secret. Each message is assigned a hash value based on a range of fields like the message template, the severity, the URL, and more. Some values are normalized or modified before being sent as input to the hash function. An example of this is removing numbers from the log message, which will ensure that `Error on product 1` and `Error on product 2` will be considered the same log message. When receiving a new log message, we check if an existing message with the same hash is already stored in the log. If not, the new message will be marked as *New* by setting the `isNew` flag to `true`. If we already found one or more log messages with the same hash, the new message will have its `isNew` flag set to `false`.
1111

1212
## Messages and apps
1313

@@ -19,10 +19,10 @@ We sometimes get requests to modify the hash function, but unfortunately, that's
1919

2020
There are still some possibilities to force two log messages to not be unique. The obvious is to include different variables inside the log message. Remember that numbers are removed, why this must consist of letters. Another approach is to put individual values in the `source` field. This can be done in all integrations by implementing the `OnMessage` action. Some integrations also support setting the `source` field by including it as structured properties like `Error from {source}`.
2121

22-
## Setting re-occurring messages as New
22+
## Setting recurring messages as New
2323

2424
A common request is to get a notification if an error that you believed was fixed re-occur. This scenario is built into elmah.io's issue tracker. When marking a log message as fixed through the UI or API, elmah.io automatically marks all instances of the log message as fixed. If a new log message with the same hash is logged at some point, the `isNew` flag on this message will be set to `true`. This will trigger the *New Error Email* and most of the integrations again.
2525

2626
## Retention
2727

28-
Depending on your current plan, each subscription provides x days of retention for log messages. This means that log messages are automatically deleted after x days in the database. Once all instances of a log message are deleted, a new log message generating the same hash as the deleted messages will be marked as new. To increase the chance of log messages being marked as new, you can lower the retention on each log on the *Log Settings* page.
28+
Depending on your current plan, each subscription provides x days of retention for log messages. This means that log messages are automatically deleted after x days in the database. Once all instances of a log message are deleted, a new log message generating the same hash as the deleted messages will be marked as new. To increase the chance of log messages being marked as new, you can lower the retention on each log on the *Log Settings* page.

0 commit comments

Comments
 (0)