24. Jan. 2025 by Urs
Retrospective 6.3 - Getting to the heart of custom columns
Spirited Definition of Custom Columns
As a reminder, Retrospective performs Autofind on selected log files or container log streams in order to determine their encoding, log entry separation (delimiter), date/time format and name/value pattern fields. Detected fields may be used to define result table custom columns. So far, this was a rather tedious task that was performed in the “Profile Manager”. Fields had to be assigned or added one-by-one to new or existing custom columns..
But now, we greatly improved the definition of custom columns! Read on if you're interested in how the new version of Retrospective helps you with this process.
Assign Detected fields
In the event that a previously performed Autofind has detected potentially interesting name/value pattern fields, Retrospective now immediately displays the new “Define Custom Columns” dialog. This dialog lets you assign newly detected fields of interest to new or existing custom columns.
If any of your existing profiles already contains custom columns, the link "Columns from existing profiles" appears right to the filter field. Use it to switch to a different view that lets you adopt custom columns from existing profiles.
Adopt Existing Columns
Should any of your existing profiles already contains custom columns, this dialog lets you adopt such columns and add them to the current profile. This drastically reduces the profile configuration process in situations where profiles are based on data sources with similar content.
New Field Types
We also introduced new field types that give you an increased choice for defining specific custom columns:
| "Data After" | Such fields follow or start with a well-defined text pattern, the delimiter. This can be a single character, a word or even a phrase. When defining such fields, you decide whether the delimiter itself shall also be part of the extracted value. |
| "Data Before" | Such fields precede a well-defined text pattern, the delimiter. This can be a single character, a word or even a phrase. When defining such fields, you decide whether the delimiter itself shall also be part of the extracted value. |
| "Data Between" | Such fields are located between two identical or different delimiters. These can be single characters, words or even phrases. When defining such fields, you decide whether the left and right delimiter shall also be part of the extracted value. |
| Standalone Number | Retrospective lets you choose among the following types of Standalone Number Fields.
|
K8s Multi-Cluster Support
A profile may now contain pods from different Kubernetes clusters. The Container Browser presents all locally configured clusters and presents their internal structure. Simply browse through namespaces, deployments etc. to finally select the pods of interest.
When a profile contains pods from different clusters, searching/monitoring is performed simultaneously in the related clusters.
Other new Features
Retrospective 6.3.0 provides many other interesting features.
-
Definition of preferred locales for detecting date format in log entries.
-
New toolbar with specific functionality for the result detail view.
-
JSON data...
-
Syntax highlighting.
-
Formatting and un-escaping of nested escaped JSON.
-
Un-escaping escaped control characters.
-
-
Automatically select last opened folder in the “Add Data Source(s)” dialog.
-
Handling log entries enriched by Dynatrace.
-
Indicator that appears when search definition and result data are no longer in sync.
-
etc.
As longtime users know, most of these new features are customizable within the preferences dialog.
For further details, please take a look around the Retrospective user manual.
Ongoing Maintenance
As always, we invested much time and thought in improving the quality of our software. We updated the underlying Eclipse Target Platform together with different libraries and fixed a number of bugs. For further details, please consulted the Retrospective 6.3.0 release notes.