Healthcheck causes errors in rootless mode
General
Escalation
General
Escalation
Description
How to test
None
How to document
None
Smart Checklist
Activity
Show:

C W June 23, 2022 at 8:39 AMEdited
Zero difference with 2.28.0:

Denys Kondratenko June 22, 2022 at 10:08 AM
I think this is not an issue any more, as healthcheck does `curl` now.
could you confirm and close?

Lalit Choudhary November 25, 2020 at 1:00 PM
Hi Ceri,
Thank you for the report.
Details
Details
Assignee
Unassigned
UnassignedReporter

Priority
Components
Needs QA
Yes
Needs Doc
Yes
Affects versions
Environment
Smart Checklist
Open Smart Checklist
Smart Checklist

Open Smart Checklist
Created October 23, 2020 at 10:23 AM
Updated March 6, 2024 at 3:46 AM
When trying to run a container using podman as an unprivileged user (rootless mode), the healthcheck that was added in 2.4.0 causes the container to produce error messages as the container is created.
This looks to have now been addressed in
podman
(handling healthchecks in rootless mode), but that is not present in at least the versions available to CentOS 7 and Ubuntu 18.04.To mitigate the healthcheck errors it is possible to remove it using
--health-cmd none
:This shows the next issues that would be encountered (expectation of reserved port usage inside the container), suggestions to resolve those issues are in :
Run the internal Alertmanager on a non-standard port, which allows them to run alongside each other
Run all of the internal services on non-standard, non-privileged ports and do not use wildcard bind addresses
Provide an entrypoint that either reads a definition file, or ENV variables to set addresses and ports