ClusterLabs / ClusterLabs/crmsh
ERROR: pacemaker-controld:metadata: got no meta-data, does this RA exist?
- Dominant language
- Python
- Stars
- 143
- Forks
- 96
- Avg merge
- 2d 6h
- Merged PRs (30d)
- 16
Description
Hello,
TL;DR - SLES15-SP3, two-nodes cluster, same /etc/crm/crm.conf file on both nodes, when user invokes the "crm configure" shell, then type "property" the following error appears on the console:
```
crm(live/hostname) configure# property sudo: pacemaker-controld command not found
DEBUG: pacemaker-controld metadata exited with code 1
ERROR: pacemaker-controld:metadata: got no meta-data, does this RA exist?
```
The debug does not go too far, as per the above messages. The sudoers files are okay and user is logging into the system as root directly. The `env` variables are exactly the same on both nodes (the node `01` has this `crm` issue while the `02` does not).
Package version:
```
crmsh-4.3.1+20220505.cf4ab649-150200.5.80.1.noarch
crmsh-scripts-4.3.1+20220505.cf4ab649-150200.5.80.1.noarch
---
kmod-bash-completion-29-4.15.1.noarch
libvirt-bash-completion-7.1.0-150300.6.29.1.noarch
bash-completion-devel-2.7-4.6.1.noarch
bash-completion-2.7-4.6.1.noarch
bash-lang-4.4-19.6.1.noarch
bash-doc-4.4-19.6.1.noarch
bash-4.4-19.6.1.x86_64
bash-completion-doc-2.7-4.6.1.noarch
```
Could you kindly help on this? Thanks for any help you can provide.
Cheers.
Contributor guide
Research direction
Start with the crm configure shell completion path for property and the pacemaker-controld metadata command described in the report. Reproduce the failure on SLES15-SP3, compare the working and failing nodes, and verify that completion no longer reports missing metadata when the underlying command is available.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- bash, linux, python
- Domain
- cli, operating-systems
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100