Link templates for Log Attributes

We've expanded our link templates feature to include log attributes. This allows you to create links to external services based on the attributes of a log message.

One caveat to note is that in order to use the same link template for both sample tags and log attributes, we match the attribute name without the type extension.

E.g, a log attribute user_id_string will match a link template based on user_id

Assignee filter

We've introduced search functionality to the Assignee filter, making it easier to filter, find, and assign users to incidents:

App updates

We've improved a number of smaller details in the App. Here's a list of the changes we've made:

• Selected columns on the Error/Performance sample pages are now persisted.
• Alerts in the Time Detective now link to Anomaly Incidents, instead of the edit form.
• Add log view params to the url, this allows for additional filtering on a log view, and the ability to save it as a new view.
• The chart on the log view page now handles large timeframes. It now loads data in batches, preventing timeouts, and a broken graph.
• Link templates for Error/Performance sample tags can now include dashes (-).
• Expand SQL syntax highlighting to additional Event Payload bodies, such as active_record, ecto, knex, mysql, pg, sequel, psycopg2 and sql.

Long-term storage for logging

We've added a feature that allows for long-term storage of log sources. Provide an S3 bucket and credentials; we'll transfer our logs to your bucket.

Long-term log storage is part of a suite of Business features that don't modify our core features but make administration and staying compliant more manageable.

So far, long-term log storage, enforcing two-factor authentication, and the ability to sign a Business Associate Agreement for HIPAA compliance are available. For pricing, please contact our Customer Success team.

Enforce Two-Factor Authentication

We've added a feature to enforce two-factor authentication for every member of an organization. This is part of a suite of Business features that don't modify our core features but make administration and staying compliant more manageable.

So far, enforcing two-factor authentication and the ability to sign a Business Associate Agreement for HIPAA compliance are available. For pricing, please contact our Customer Success team.

Add Ability to close Anomaly incidents

It's now possible to close Anomaly incidents manually. This can be done by clicking the "Close last occurrence" button in the actions box on the incident page.

This helps customers with an incident that isn't relevant anymore but still shows up on the list of incidents. For example, an open swap incident that has been resolved by cycling the server in a cloud environment.

Linear Integration

We've added a new integration with Linear.

Send incidents over to Linear as a new issue with a link back to the incident in AppSignal.

Visit our Documentation to learn more about this integration.

Upgrade of top monitoring features

Changed

• We've spent the past few months making our top monitoring features more consistent.

  Error, Performance, Anomalies, and Logging now all share the same navigation structure with a list of incidents, graphs, logbooks, and occurrences.

  Read more about it in our blog post.

Add ability to load older log items

• Add ability to load older log items from the "Livetail" log view. This button will take you to the "Timestamped view" with the same filters applied, allowing you to paginate older log items.

Related log lines for error samples

• Show related log lines on the error sample page if both log lines and samples are tagged with request_id.

Show starred apps on the organisation index page

• We now show starred apps on the organisation index page, for faster access to your most-used apps.

App usability improvements

• Add "breakdown" bars in the samples table for performance samples. These help you quickly figure out if it's a single event group (queries, templates, external requests) that's causing the performance issues.
• Show performance of action in the last five deploys
• Revamp the performance incidents. Bring them in line with the exception incidents. Add tabs for Logbook, Sample table, Graphs and incident Summary
• Improve N+1 detection, we'll now show multiple non-nested events repeated as a group (e.g. active record initialise and query repeated 10 times, now shows them as a group).

Date picker and filtering improvements, Golang installation flow, Markdown support for comments

• Add ability to select times in date range picker
• Allow searching hostnames with a dot (.) in the hostname filter on host metrics/usage pages
• Add install flow for Golang
• Add markdown support for logbook comments/notes
• Update styles to be consistent with homepage/docs/blog (button colours)

Various usability enhancements

• Make Duration formatter more accurate, instead of showing ~2 min, or ~2 hours, we now show 02:31 min, and 04:32 h.
• The marker selection dropdown on the incident page is now persisted in the session. You don't have to select "last 30 days" on every reload anymore.
• The exception incidents samples table now has a column select and horizontal scroll. This should help customers with lots of custom tags.
• The exception incident message/backtraces are now updated from samples, ensuring they are consistent with the latest error message/backtrace (at most once a day).
• Incident notifications now include a timestamp to the sample that generated the incident again, this helps debug issues with those specific attributes.

Alphabetical sorting of namespace filters

Filters for namespaces are now sorted alphabetically to make it easier to select the namespaces that you want to see.

Changelog update 10 January, 2023

We've renamed the "What Happened Here" feature to "Time Detective", and improved the navigation by moving the tabs from the header to a sidebar. The header is now used for added navigation. You can now navigate to the previous/next minute/hour, or pick a specific time.

Changelog update 21 December, 2022

It's now possible to use custom domains for status pages. Instead of yourapp.appsignal-status.com, you can also pick your own domain such as status.yourapp.com.

Changelog update 13 December, 2022

We've made some quality of life improvements for N+1 queries.

1. It's now possible to filter on samples that contain N+1 queries
2. We show the time with a red label for N+1 queries
3. We modified the header of the hover box on the sample timeline to show that the sample contains a N+1 query

Changelog update 23 June, 2022

Line-of-code-based error grouping and sourcemapped backtraces in notifications, see the blog post for details.

Changelog update 4 May, 2022

Add ability for organization owners to cycle the App Push API Key.