Skip to content
FlitzyFlitzy

Logs and health

Status

Section titled “Status”
Terminal window
flitzy status shop

shows in one view:

  • Health: healthy, degraded, crashing, starting, stopped or not_deployed, with a one-line summary.
  • The problem, if there is one: the Kubernetes state (CrashLoopBackOff, OOMKilled, ImagePullBackOff, …), the real reason behind a restart loop (for example killed for using too much memory, exit code 137), a hint what to do and the last lines the crashed container printed.
  • CPU and memory of the running version, as a share of the limits (0.5 CPU, 512 MiB).
  • Pods with restarts, recent warnings, the volume and the last 5 deploys.

degraded means the app still serves, but something is off: a new version fails while the old one keeps running, or the container restarted in the last 15 minutes.

Agents get the same from app_status. In the dashboard it’s the Health and Resources cards of an app.

Logs

Section titled “Logs”
Terminal window
flitzy logs shop --tail 200   # the last lines
flitzy logs shop -f           # follow live, Ctrl-C to stop
flitzy logs shop -f -t        # with timestamps

-f keeps following across crashes and new deploys and says when it switches to a new container. If the connection drops, it reconnects and continues without gaps or duplicates.

In the dashboard, an app’s Logs tab is a live terminal with a filter, pause and resume, timestamps and line wrapping.

Everything your app writes to stdout and stderr shows up. Logs written to files don’t.

Metrics

Section titled “Metrics”

CPU and memory are averages over about 15 seconds and appear about a minute after a container starts. Memory close to 100 % means the next spike gets the app killed with OOMKilled; see Troubleshooting.

Activity

Section titled “Activity”

The Activity tab in the dashboard and flitzy audit list every change to your apps, databases, domains and tokens: who, when, from which address and whether it worked. Denied attempts are listed too.

Terminal window
flitzy audit                    # everything of your team
flitzy audit shop               # one app
flitzy audit --action env       # deploy, env, domain, token, database, …

Reading status or logs is not recorded. Values of tokens and environment variables are never stored.