Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

ERROR: pacemaker-controld:metadata: got no meta-data, does this RA exist? #982

Open
samuelfusato opened this issue Jun 10, 2022 · 5 comments

Comments

@samuelfusato
Copy link

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.

@liangxin1300
Copy link
Collaborator

Hi @samuelfusato

I guess you've set core.user in crm.conf? see

if config.core.user:

In above line, run su -E -u <user> pacemaker-controld, got the above error
sudo: pacemaker-controld command not found

@samuelfusato
Copy link
Author

Dear @liangxin1300,

Thanks for your answer. This is the content of /etc/crm/crm.conf:

 [core]
debug = yes
[path]
crm_daemon_user = hacluster
ocf_root = /usr/lib/ocf

@liangxin1300
Copy link
Collaborator

Could you please remove add_sudo in this line?

rc, s = get_stdout(add_sudo(cmd), stderr_on=stderr_on, shell=shell)

Like rc, s = get_stdout(cmd, stderr_on=stderr_on, shell=shell)

If it works, please print config.core.user in this function to see what value it is

@samuelfusato
Copy link
Author

Hi @liangxin1300,

Your suggestion fixed the issue. However, I could not obtain the print from my customer (I tried once but I preferred not to push them on this) and I hope you understand.

Thank you so much for your kind assistance.

@liangxin1300
Copy link
Collaborator

Yeah, that is only a workaround, not fix the root cause yet

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants