Hi Team Fluent bit pod fails to start in case syslog output Fluent Community #fluent-bit

Hi Team, Fluent-bit pod fails to start, in case s...

Vishal Kakkar

10/09/2025, 7:16 AM

Hi Team, Fluent-bit pod fails to start, in case syslog output plugin is configured with a non resolvable host. looks like a defect, it should stop whole fluent-bit in case a user configures a invalid syslog. Specially when fluent-bit is doing much more than just a syslog output. [2025/10/09 111004] [ info] [fluent bit] version=4.0.5, commit=d9be919eab, pid=1 [2025/10/09 111004] [ info] [storage] ver=1.5.3, type=memory, sync=normal, checksum=off, max_chunks_up=128 [2025/10/09 111004] [ info] [simd ] SSE2 [2025/10/09 111004] [ info] [cmetrics] version=1.0.4 [2025/10/09 111004] [ info] [ctraces ] version=0.6.6 [2025/10/09 111004] [ info] [inputtailtail.0] initializing [2025/10/09 111004] [ info] [inputtailtail.0] storage_strategy='memory' (memory only) [2025/10/09 111004] [ info] [inputtailtail.0] multiline core started [2025/10/09 111004] [ info] [filterkuberneteskubernetes.0] https=1 host=kubernetes.default.svc port=443 [2025/10/09 111004] [ info] [filterkuberneteskubernetes.0] token updated [2025/10/09 111004] [ info] [filterkuberneteskubernetes.0] local POD info OK [2025/10/09 111004] [ info] [filterkuberneteskubernetes.0] testing connectivity with API server... [2025/10/09 111004] [ info] [filterkuberneteskubernetes.0] connectivity OK [2025/10/09 111004] [ info] [inputemitterk8s_tag_rewriter] initializing [2025/10/09 111004] [ info] [inputemitterk8s_tag_rewriter] storage_strategy='memory' (memory only) [2025/10/09 111004] [ info] [outputstdoutstdout.0] worker #0 started [2025/10/09 111004] [ info] [outputfilefile.1] worker #0 started [2025/10/09 111004] [ warn] net]: getaddrinfo(host='2.2.2.coma'): No address associated with hostname [2025/10/09 111004] [error] [output] failed to initialize 'syslog' plugin [2025/10/09 111004] [error] [engine] output initialization failed [2025/10/09 111005] [ info] [input] pausing tail.0 [2025/10/09 111005] [ info] [input] pausing k8s_tag_rewriter [2025/10/09 111005] [ info] [outputstdoutstdout.0] thread worker #0 stopping... [2025/10/09 111005] [ info] [outputstdoutstdout.0] thread worker #0 stopped [2025/10/09 111005] [ info] [outputfilefile.1] thread worker #0 stopping... [2025/10/09 111005] [ info] [outputfilefile.1] thread worker #0 stopped

Vishal Kakkar

10/23/2025, 2:59 PM

Any help!!

Pat

10/23/2025, 3:07 PM

yeah it fails with a simple example - I've no idea what you're configuring above so tried to reproduce:

Copy code

$ docker run --pull=always --rm -it fluent/fluent-bit -i dummy -o syslog -p match='*' -p host='idontexit'
latest: Pulling from fluent/fluent-bit
Digest: sha256:2c2d19b881b17ad3ee2e5dd75302b799151d2106f191e2652a479018a743d6e1
Status: Image is up to date for fluent/fluent-bit:latest
Fluent Bit v4.0.13
* Copyright (C) 2015-2025 The Fluent Bit Authors
* Fluent Bit is a CNCF sub-project under the umbrella of Fluentd
* <https://fluentbit.io>

______ _                  _    ______ _ _             ___  _____ 
|  ___| |                | |   | ___ (_) |           /   ||  _  |
| |_  | |_   _  ___ _ __ | |_  | |_/ /_| |_  __   __/ /| || |/' |
|  _| | | | | |/ _ \ '_ \| __| | ___ \ | __| \ \ / / /_| ||  /| |
| |   | | |_| |  __/ | | | |_  | |_/ / | |_   \ V /\___  |\ |_/ /
\_|   |_|\__,_|\___|_| |_|\__| \____/|_|\__|   \_/     |_(_)___/ 


[2025/10/23 15:06:05] [ info] [fluent bit] version=4.0.13, commit=2a1d3c5d7f, pid=1
[2025/10/23 15:06:05] [ info] [storage] ver=1.5.3, type=memory, sync=normal, checksum=off, max_chunks_up=128
[2025/10/23 15:06:05] [ info] [simd    ] SSE2
[2025/10/23 15:06:05] [ info] [cmetrics] version=1.0.5
[2025/10/23 15:06:05] [ info] [ctraces ] version=0.6.6
[2025/10/23 15:06:05] [ info] [input:dummy:dummy.0] initializing
[2025/10/23 15:06:05] [ info] [input:dummy:dummy.0] storage_strategy='memory' (memory only)
[2025/10/23 15:06:05] [ warn] net]: getaddrinfo(host='idontexit'): No address associated with hostname
[2025/10/23 15:06:05] [error] [output] failed to initialize 'syslog' plugin
[2025/10/23 15:06:05] [error] [engine] output initialization failed
[2025/10/23 15:06:06] [ info] [input] pausing dummy.0

I'm not sure what folks would want when output never resolves though, you don't want it necessarily to carry on running but doing nothing

Pat

10/23/2025, 3:08 PM

Ah TCP does work and keeps retrying so I would raise an issue on that with the reproducers to demonstrate:

Copy code

$ docker run --pull=always --rm -it fluent/fluent-bit -i dummy -o tcp -p match='*' -p host='idontexit'
latest: Pulling from fluent/fluent-bit
Digest: sha256:2c2d19b881b17ad3ee2e5dd75302b799151d2106f191e2652a479018a743d6e1
Status: Image is up to date for fluent/fluent-bit:latest
Fluent Bit v4.0.13
* Copyright (C) 2015-2025 The Fluent Bit Authors
* Fluent Bit is a CNCF sub-project under the umbrella of Fluentd
* <https://fluentbit.io>

______ _                  _    ______ _ _             ___  _____ 
|  ___| |                | |   | ___ (_) |           /   ||  _  |
| |_  | |_   _  ___ _ __ | |_  | |_/ /_| |_  __   __/ /| || |/' |
|  _| | | | | |/ _ \ '_ \| __| | ___ \ | __| \ \ / / /_| ||  /| |
| |   | | |_| |  __/ | | | |_  | |_/ / | |_   \ V /\___  |\ |_/ /
\_|   |_|\__,_|\___|_| |_|\__| \____/|_|\__|   \_/     |_(_)___/ 


[2025/10/23 15:07:53] [ info] [fluent bit] version=4.0.13, commit=2a1d3c5d7f, pid=1
[2025/10/23 15:07:53] [ info] [storage] ver=1.5.3, type=memory, sync=normal, checksum=off, max_chunks_up=128
[2025/10/23 15:07:53] [ info] [simd    ] SSE2
[2025/10/23 15:07:53] [ info] [cmetrics] version=1.0.5
[2025/10/23 15:07:53] [ info] [ctraces ] version=0.6.6
[2025/10/23 15:07:53] [ info] [input:dummy:dummy.0] initializing
[2025/10/23 15:07:53] [ info] [input:dummy:dummy.0] storage_strategy='memory' (memory only)
[2025/10/23 15:07:53] [ info] [sp] stream processor started
[2025/10/23 15:07:53] [ info] [engine] Shutdown Grace Period=5, Shutdown Input Grace Period=2
[2025/10/23 15:07:53] [ info] [output:tcp:tcp.0] worker #0 started
[2025/10/23 15:07:53] [ info] [output:tcp:tcp.0] worker #1 started
[2025/10/23 15:07:54] [ warn] [net] getaddrinfo(host='idontexit', err=1): DNS server returned answer with no data
[2025/10/23 15:07:54] [error] [output:tcp:tcp.0] no upstream connections available to idontexit:5170
[2025/10/23 15:07:54] [ warn] [engine] failed to flush chunk '1-1761232073.452548542.flb', retry in 7 seconds: task_id=0, input=dummy.0 > output=tcp.0 (out_id=0)
[2025/10/23 15:07:55] [ warn] [net] getaddrinfo(host='idontexit', err=1): DNS server returned answer with no data
[2025/10/23 15:07:55] [error] [output:tcp:tcp.0] no upstream connections available to idontexit:5170
[2025/10/23 15:07:55] [ warn] [engine] failed to flush chunk '1-1761232074.452458997.flb', retry in 8 seconds: task_id=1, input=dummy.0 > output=tcp.0 (out_id=0)
[2025/10/23 15:07:56] [ warn] [net] getaddrinfo(host='idontexit', err=1): DNS server returned answer with no data
[2025/10/23 15:07:56] [error] [output:tcp:tcp.0] no upstream connections available to idontexit:5170

Pat

10/23/2025, 3:08 PM

Syslog does not retry, TCP does

Vishal Kakkar

10/24/2025, 5:30 AM

@Pat our fluent-bit is configured with multiple output pipelines like one for centralized logging to folder, toher for syslog streaming to external syslog servers. Now giving an invalid FQDN (non routable) is causing fluentbit pod failure at startup. Expectation would be either to ignore or just fail syslog part not impacting the whole fluent-bit micro-service.

Pat

10/24/2025, 7:59 AM

Sure, as I say raise an issue on it as it's not consistent either

Open in Slack

Previous Next