Unable to have optional inputs in config

[cwchristerw]

Relevant telegraf.conf

[[inputs.docker]]
    endpoint = "tcp://server.devices.example.com:2376"
    container_name_exclude = ["*_*"]
    timeout = "30s"
    total_include = ["cpu", "blkio", "network"]
    insecure_skip_verify = true
    startup_error_behavior = "retry"

[[inputs.docker]]
    endpoint = "tcp://laptop.devices.example.com:2376"
    container_name_exclude = ["*_*"]
    timeout = "30s"
    total_include = ["cpu", "blkio", "network"]
    insecure_skip_verify = true
    startup_error_behavior = "retry"

Logs from Telegraf

-

System info

Telegraf 1.40.0

Docker

No response

Steps to reproduce

1. Create config with multiple docker inputs and defining startup_error_behaviour = "retry".
2. Start Telegraf in Docker container

...

Expected behavior

I would expect it skip plugins that fail and retry periodically. If laptop is offline and server is online, telegraf would connect to server and get data from it. When laptop is online connection is established with it also. Using ignore doesn't also work because it will not retry again when laptop is available after startup has disabled it.

Actual behavior

Telegraf will not connect to any docker inputs if one input unavailable.

Additional info

No response

[cwchristerw]

I want legacy way of getting to ignore status of inputs during startup. I want to just ignore if doesn't work but not disable inputs because it will prevent me from getting data when laptop is available.

[cwchristerw]

One way I could think that I could use probe but it will also disable plugin during startup.

[srebhan]

@cwchristerw this works for me. For a machine that does not exist I see

2026-09-22T13:20:05Z I! Loading config: test_configs/docker_issue_19749.conf
2026-09-22T13:20:05Z I! Starting Telegraf 1.41.0-75ba6fc9 brought to you by InfluxData the makers of InfluxDB
2026-09-22T13:20:05Z I! Available plugins: 244 inputs, 9 aggregators, 35 processors, 26 parsers, 68 outputs, 8 secret stores
2026-09-22T13:20:05Z I! Loaded inputs: docker (2x)
2026-09-22T13:20:05Z I! Loaded aggregators:
2026-09-22T13:20:05Z I! Loaded processors:
2026-09-22T13:20:05Z I! Loaded secretstores:
2026-09-22T13:20:05Z I! Loaded outputs: file
2026-09-22T13:20:05Z I! Tags enabled: host=Hugin
2026-09-22T13:20:05Z I! [agent] Config: Interval:1s, Quiet:false, Hostname:"Hugin", Flush Interval:10s
2026-09-22T13:20:05Z W! [agent] The default value of 'skip_processors_after_aggregators' will change to 'true' with Telegraf v1.40.0! If you need the current default behavior, please explicitly set the option to 'false'!
2026-09-22T13:20:05Z I! [inputs.docker::nonexistent] Startup failed: failed to ping daemon: failed to connect to the docker API at tcp://laptop.devices.example.com:2376: lookup laptop.devices.example.com on 192.168.119.17:53: no such host; retrying...
2026-09-22T13:20:06Z E! [inputs.docker::nonexistent] Error in plugin: not connected
2026-09-22T13:20:07Z E! [inputs.docker::nonexistent] Error in plugin: not connected
2026-09-22T13:20:08Z E! [inputs.docker::nonexistent] Error in plugin: not connected
2026-09-22T13:20:09Z E! [inputs.docker::nonexistent] Error in plugin: not connected
2026-09-22T13:20:10Z E! [inputs.docker::nonexistent] Error in plugin: not connected
2026-09-22T13:20:11Z E! [inputs.docker::nonexistent] Error in plugin: not connected
2026-09-22T13:20:12Z E! [inputs.docker::nonexistent] Error in plugin: not connected
2026-09-22T13:20:13Z E! [inputs.docker::nonexistent] Error in plugin: not connected
2026-09-22T13:20:14Z E! [inputs.docker::nonexistent] Error in plugin: not connected
2026-09-22T13:20:15Z E! [inputs.docker::nonexistent] Error in plugin: not connected
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 n_used_file_descriptors=23i,n_containers_running=0i,n_images=37i,n_cpus=16i,n_containers=2i,n_containers_stopped=2i,n_containers_paused=0i,n_goroutines=44i,n_listener_events=0i 1790083206000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 memory_total=67338375168i 1790083206000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 n_listener_events=0i,n_containers_running=0i,n_containers_stopped=2i,n_goroutines=44i,n_cpus=16i,n_used_file_descriptors=23i,n_containers=2i,n_containers_paused=0i,n_images=37i 1790083207000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 memory_total=67338375168i 1790083207000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 n_goroutines=44i,n_used_file_descriptors=23i,n_containers=2i,n_containers_running=0i,n_containers_stopped=2i,n_images=37i,n_listener_events=0i,n_cpus=16i,n_containers_paused=0i 1790083208000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 memory_total=67338375168i 1790083208000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 n_containers_paused=0i,n_goroutines=44i,n_listener_events=0i,n_used_file_descriptors=23i,n_containers=2i,n_images=37i,n_cpus=16i,n_containers_running=0i,n_containers_stopped=2i 1790083209000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 memory_total=67338375168i 1790083209000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 n_cpus=16i,n_used_file_descriptors=23i,n_containers=2i,n_containers_paused=0i,n_containers_running=0i,n_containers_stopped=2i,n_images=37i,n_goroutines=44i,n_listener_events=0i 1790083210000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 memory_total=67338375168i 1790083210000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 n_listener_events=0i,n_cpus=16i,n_used_file_descriptors=23i,n_containers=2i,n_containers_running=0i,n_containers_stopped=2i,n_containers_paused=0i,n_images=37i,n_goroutines=44i 1790083211000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 memory_total=67338375168i 1790083211000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 n_goroutines=44i,n_cpus=16i,n_containers_paused=0i,n_listener_events=0i,n_used_file_descriptors=23i,n_containers=2i,n_containers_running=0i,n_containers_stopped=2i,n_images=37i 1790083212000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 memory_total=67338375168i 1790083212000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 n_cpus=16i,n_containers_running=0i,n_containers_paused=0i,n_images=37i,n_listener_events=0i,n_used_file_descriptors=23i,n_containers=2i,n_containers_stopped=2i,n_goroutines=44i 1790083213000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 memory_total=67338375168i 1790083213000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 n_cpus=16i,n_containers_running=0i,n_used_file_descriptors=23i,n_containers=2i,n_containers_stopped=2i,n_containers_paused=0i,n_images=37i,n_goroutines=44i,n_listener_events=0i 1790083214000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 memory_total=67338375168i 1790083214000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 n_used_file_descriptors=23i,n_containers=2i,n_containers_stopped=2i,n_images=37i,n_cpus=16i,n_containers_running=0i,n_containers_paused=0i,n_goroutines=44i,n_listener_events=0i 1790083215000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 memory_total=67338375168i 1790083215000000000
2026-09-22T13:20:16Z E! [inputs.docker::nonexistent] Error in plugin: not connected
2026-09-22T13:20:17Z E! [inputs.docker::nonexistent] Error in plugin: not connected
^C2026-09-22T13:20:17Z I! [agent] Hang on, flushing any cached metrics before shutdown
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 n_cpus=16i,n_containers_running=0i,n_containers_stopped=2i,n_containers_paused=0i,n_images=37i,n_used_file_descriptors=23i,n_containers=2i,n_goroutines=44i,n_listener_events=0i 1790083216000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 memory_total=67338375168i 1790083216000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 n_cpus=16i,n_used_file_descriptors=23i,n_listener_events=0i,n_containers=2i,n_containers_running=0i,n_containers_stopped=2i,n_containers_paused=0i,n_images=37i,n_goroutines=44i 1790083217000000000
docker,engine_host=Hugin,host=Hugin,server_version=29.8.1 memory_total=67338375168i 1790083217000000000
20```

with your config above. As you can see the plugin with the working machine (alias `local`) produces metrics while the failing plugin (alias `nonexistent`) retries to start with every gather cycle...

What exactly is the issue here?