histRetentionDays not taking effect

[jonathonc]

It doesn't seem like the histRetentionDays property works as I expect. The logs and history files in the ~/.local/share/dagu folder don't get deleted as expected?

schedule: "*/5 * * * *"
tags: health-checks
histRetentionDays: 1
steps:
  - name: step1
    command: bash runHealthCheck

[yottahmd]

Do those DAGs have the same name? The histRetentionDays setting applies to DAG runs that share the same name. Specifically, when a DAG is executed, it removes old run histories with the same DAG name.

By default, the DAG name is the file name, but you can specify it explicitly in the DAG definition, for example:

name: daily-job

[jonathonc]

We don't explicitly name the DAG in the YAML, but it just uses the file name uptime-robot-heartbeat.yaml

[yottahmd]

I see, it should cleanup old data on each run, but let me double check using same settings. Are you running it in a docker or direct binary? I’d appreciate it if you could share your setup.

[jonathonc]

We are running the direct binaries on Ubuntu 22, running using systemd

An example of a profile

uptime-robot-heartbeat.yaml

schedule: "*/5 * * * *"
tags: health-checks
histRetentionDays: 1
steps:
  - name: step1
    command: bash runHealthCheck

It doesn't seem to respect the 1 day histRetentionDays, the 30 day histRetentionDays also doesn't seem to take affect

[yottahmd]

@jonathonc I tried but couldn't replicate the issue. Could you check one of the execution and if there's any error or warning message in the log?

Also please share the dagu's directory configuration to replicate the issue (config.yaml or environment variable).

[jonathonc]

the config.yaml

navbarTitle: DAGU - mydagu
dags: /home/ubuntu/dagu/mydagu

I can see the logs here are being removed

/home/ubuntu/.local/share/dagu/logs