Use Trace Search & Analytics to filter application performance metrics and APM Events by user-defined tags. It allows deep exploration of the web requests flowing through your service.
Trace Search & Analytics can be enabled per APM service and per host. A service on which it is enabled exposes all its APM Events to Datadog.
Downstream services like databases and cache layers aren’t in the list of available services (as they don’t generate traces on their own), but their information is picked up by the top level services that call them.
In the Trace Search view you can:
An APM event is generated every time a trace is generated. It corresponds to all the tags associated with the trace, plus the top span of the trace.
APM events aren’t just traces: traces get sampled and APM events don’t, and Datadog only keeps the top span information for the APM events, not the full trace.
APM Events can be enriched with tags, like
request duration, or
product type. You can then filter and query on those tags in the Trace Search & Analytics UI.
If checked, APM Events listed in the trace stream have a trace associated with them, so you can display the full trace with all its associated spans.
All search parameters are contained in the url of the page, so it is very simple to share your view.
A query is composed of terms and operators.
There are two types of terms:
To combine multiple terms into a complex query, use any of the following boolean operators:
|Operator||**Description **||**Example **|
||Intersection: both terms are in the selected events (if nothing is added, AND is taken by default)||authentication AND failure|
||Union: either terms is contained in the selected events||authentication OR password|
||Exclusion: the following term is NOT in the event||authentication AND -password|
For instance, if your facet name is url and you want to filter on the url value www.datadoghq.com just enter:
||All traces with the tag
||All traces that contain tags
||All traces that contain
If your tags don’t follow tags best practices and don’t use the
key:value syntax, use this search query:
To perform a multi-character wildcard search, use the
* symbol as follows:
service:web*matches every trace that has a services starting with
@url:data*matches every trace that has a
>= to perform a search on numerical attributes. For instance, retrieve all traces that have a response time over 100ms with:
It is also possible to search for numerical attributes within a specific range. For instance, retrieve all your 4xx errors with:
@http.status_code:[400 TO 499]
Typing a complex query can be cumbersome. Use the search bar’s autocomplete feature to complete your query using existing values:
The following attributes are considered as special:
\ require escaping.
For instance, to search traces that contain
user=JaneDoe in their
url the following search must be entered:
The same logic must be applied to spaces within trace attributes. It is not recommended to have spaces in trace attributes but in such cases, spaces require escaping.
If an attribute is called
user.first name, perform a search on this attribute by escaping the space:
Don’t lose time building the same views everyday. Saved searches contain your search query, columns, and time horizon. They are then available in the search bar thanks to the auto-complete matching whether the search name or query.
To delete a saved search, click on the bin icon under the Trace search drop-down.
The time range allows you to display traces within a given time period. Quickly change the time range by selecting a preset range from the dropdown:
Sort the list by clicking the date column header.
Click on any trace to see more details about it:
To add more Trace details to the list, click the Columns button and select any Facets you want to see:
Choose to display one, three, or ten lines from your traces. 3 and 10 lines display are here to give you more insights on the
With one line displayed:
With three lines displayed:
With ten lines displayed:
A Facet displays all the distinct values of an attribute or a tag as well as provides some basic analytics such as the amount of traces represented. This is also a switch to easily filter your data.
Facets allow you to pivot or filter your datasets based on a given attribute. Examples Facets may include users, services, etc…
To start using an attribute as a Facet or in the search, click on it and add it as a Facet:
Use Facets to easily filters on your Traces. The search bar and url automatically reflect your selections.
Additional helpful documentation, links, and articles: