Skip to main content

SRE tabs in Context

What the SRE tabs show from Datadog, what each needs, and how to set them up.

The SRE group of Context shows how production behaves, with the data your team already records in Datadog. It needs the Datadog integration; each tab reads what its Datadog permissions allow and, when one is missing, says which.


​

The tabs


​

  • Transactions — your transactions metric in Datadog: the period's summary (total, approval and the peaks per second, minute, hour and day), each month, each product, the product families and the products with the most problems.

  • Availability — whether your services meet their SLOs (compliance and error budget), and Observability coverage: how many of your services send logs, traces and metrics to Datadog.

  • Incidents and tickets — Datadog incidents and cases opened, closed and still open, by severity or priority, and who in your team took part.

  • Deployments — Production deployments, Change Failure Rate, Mean Time to Recover and Incidents, per service. This tab always counts Datadog deployments, whatever DORA counts, so its figures can differ from the DORA tab.

  • One tab per Datadog dashboard — the widgets of each dashboard you choose to show, each saying when it was refreshed.

  • Datadog usage — what your Datadog account uses and costs, as Datadog's Plan & Usage page shows it: the month's cost and its month-end projection, the cost of each month, each product's usage this month against the previous one (measured the way Datadog bills it), and the custom metrics that add the most. The cost needs the billing_read permission.


​

Adding the tabs


​

Pick the SRE idea in New view to create a view with the tabs, or add any of them from the + of a view's tabs, where your Datadog dashboards are offered one by one. Whoever can see Performance Delivery can open them.


​

Settings


​

Whoever manages the organization, or can connect Datadog, sets the tabs up in Settings → Organization Settings → Tabs → SRE:


​

  • the transactions metric, its filter and status, and the tags for product, country, channel, family and service;

  • the latency metrics;

  • the dashboards to show: all of them, or the ones you pick;

  • the services that should be observed;

  • which Leanmote team each Datadog team is, and the Datadog people matched to Leanmote people.


​

Changing the transactions, the latency or the dashboards reads the last 180 days again. A last block lists what each part can read and the permission it is missing.


​

People and teams


​

Datadog people are matched to Leanmote people by email; one without a match can be linked to a Leanmote person right where they show, or created there if they aren't in Leanmote yet (Create «name» in the search, without sending an invitation). A Datadog team can be linked to a Leanmote team the same way. The view's team and person filters then narrow incidents, cases, SLOs, coverage and deployments — never transactions, dashboard widgets or the account's usage.


​

Show details, filters and time units


​

  • Every View … link opens the rows behind its figure, with a filter on each column and Export to CSV.

  • Filtering Production deployments, Change Failure Rate, Mean Time to Recover, Incidents or the opened incidents and cases from their Show details measures the figure with that filter for everyone who opens the view; Remove filter goes back.

  • Recovery and resolution times take the unit set from the card's ⋮ › Change time unit.

  • Country (Datadog) shows in Filters while the view shows Datadog transactions from two or more countries.


​

Duplicate a card by service


​

Whoever edits the view can compare services side by side: on Production deployments, Change Failure Rate, Mean Time to Recover, Incidents or Observability coverage, open ⋮ › Duplicate, tick the services and click Make N copies. The copies go to a tab named «<card> by service» together with the card (or to the tab where the card is already compared by service), each one measured over its service.


​

Related articles


​


​

Did this answer your question?