Concepts
The words dlogify uses and what they mean.
This page defines the terms used across the dashboard, the notifications and the API.
Organizations
An organization is your team's account: it has members, a plan, and projects. Usage and limits are counted per organization. A person can belong to several organizations, with a role in each (Members and roles).
Projects
A project holds the logs of one application or service. Error groups, tickets, channels and notification settings belong to a project.
API keys
An API key lets an application send logs to one project. Keys can only send logs; they cannot read data or change settings (Projects and API keys).
Levels
dlogify reads each event's level and maps it to INFO, WARN or ERROR; trace and debug events are dropped. ERROR and WARN events are grouped into error groups; INFO events are counted and summarized. The full table is in Level mapping.
Error groups
An error group collects the events that are the same error: the same service, level, message shape (with variable parts such as IDs and numbers set aside) and the same place in your code. A group keeps its counts, when it was first and last seen, and a status:
open: new or still happening.resolved: you marked it fixed.ignored: you do not want tickets for it; it keeps counting.regressed: it was resolved and happened again.
Tickets
A ticket is a report about an error group, written when the group is new, when it spikes (many more occurrences than usual), or when it regresses. Ongoing occurrences of a group that was already reported do not write new tickets. A ticket never changes after it is written. It contains a summary, the probable cause, a suggested fix, the facts of the group and the stack trace with the log lines around the failure.
Severity
Every ticket has a severity, which decides how it is delivered:
| Severity | Meaning | Delivery |
|---|---|---|
critical | Service down, or data loss or corruption, with no workaround | At once |
high | A core feature broken for many users | At once |
medium | Degraded or partial failure, with a workaround | Digest |
low | Cosmetic, noisy or self-recovering | Digest |
INFO summaries
INFO events are counted in time windows (60, 15 or 5 minutes depending on the plan), and each window gets a short summary: the totals, the change from the previous window, and what stood out.
Notifications
Tickets and summaries reach you through channels: email to a member, a Slack or Discord channel, or a signed webhook. critical and high tickets are sent at once; the rest are collected in a digest sent on the schedule you choose (Notifications).