Bug 1486062

Created attachment 1325942 [details]
File contains contents ansible-playbook log

Hi Sebastien,

I faced similar issue while configuring mgrs. 


TASK [ceph-defaults : set_fact] 

 [WARNING]: when statements should not include jinja2 templating delimiters such as {{ }} or {% %}. Found: ceph_release_num.{{ ceph_stable_release }} > ceph_release_num.jewel


We are setting ceph_stable_release to 'luminous' by default to avoid these issues. Is that okay ? Please let us know if these errors will be handled from ceph-ansible's end or do we need this to be added in Doc.


Redgards,
Vasishta

Appropriate log snippet (obsoletes the one mentioned in Comment 5)

fatal: [magna012]: FAILED! => {"failed": true, "msg": "The conditional check 'ceph_release_num.{{ ceph_stable_release }} > ceph_release_num.jewel' failed. The error was: error while evaluating conditional (ceph_release_num.{{ ceph_stable_release }} > ceph_release_num.jewel): 'dict object' has no attribute 'dummy'\n\nThe error appears to have been in '/usr/share/ceph-ansible/roles/ceph-defaults/tasks/facts.yml': line 2, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n---\n- set_fact:\n  ^ here\n"}

Hi, 

Based on Comment 5 and Comment 6, moving back to ASSIGNED state

Regards,
Vasishta

Vasishta,

could you provide the value you set for the variable 'ceph_repository' while encountering the issue you mentioned in c6 ?

Hi Guillaume,

I had kept it unchanged,

$ cat group_vars/all.yml | egrep -v ^# | grep -v ^$ |grep ceph_repository
ceph_repository: rhcs


Regards,
Vasishta

Do you mind testing with the last rc version available? We are not able to reproduce this. Once you can reproduce again, please send the complete output of the play.
Thanks.

Created attachment 1331342 [details]
ceph-ansible playbook log

Hi Leseb,

Seeing the issue with latest rc version on Ubuntu node

ceph-ansible version: 3.0.0~rc11-2redhat1 
Ubuntu : 16.04

all.yml file:
ubuntu@magna003:/usr/share/ceph-ansible$ cat group_vars/all.yml | egrep -v ^# | grep -v ^$
---
dummy:
fetch_directory: ~/ceph-ansible-keys
ceph_origin: repository
ceph_repository: rhcs
generate_fsid: true
monitor_interface: eth0
public_network: 10.8.128.0/21


playbook snippet:
TASK [ceph-common : install red hat storage ceph-fuse client for debian] *******
skipping: [magna003.ceph.redhat.com]

TASK [ceph-common : install red hat storage ceph-common for debian] ************
skipping: [magna003.ceph.redhat.com]

TASK [ceph-common : install red hat storage nfs gateway for debian] ************
skipping: [magna003.ceph.redhat.com]

TASK [ceph-common : install red hat storage nfs file gateway] ******************
skipping: [magna003.ceph.redhat.com]

TASK [ceph-common : install red hat storage nfs obj gateway] *******************
skipping: [magna003.ceph.redhat.com]

TASK [ceph-common : install ceph mgr for debian] *******************************
 [WARNING]: when statements should not include jinja2 templating delimiters
such as {{ }} or {% %}. Found: ceph_release_num.{{ ceph_release }} >
ceph_release_num.jewel

fatal: [magna003.ceph.redhat.com]: FAILED! => {"failed": true, "msg": "The conditional check 'ceph_release_num.{{ ceph_release }} > ceph_release_num.jewel' failed. The error was: error while evaluating conditional (ceph_release_num.{{ ceph_release }} > ceph_release_num.jewel): 'dict object' has no attribute 'dummy'\n\nThe error appears to have been in '/usr/share/ceph-ansible/roles/ceph-common/tasks/installs/install_debian_rhcs_packages.yml': line 74, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n\n- name: install ceph mgr for debian\n  ^ here\n"}

PLAY RECAP *********************************************************************
magna003.ceph.redhat.com   : ok=27   changed=1    unreachable=0    failed=1   
magna030.ceph.redhat.com   : ok=2    changed=0    unreachable=0    failed=0   
magna056.ceph.redhat.com   : ok=2    changed=0    unreachable=0    failed=0   
magna084.ceph.redhat.com   : ok=2    changed=0    unreachable=0    failed=0   
magna092.ceph.redhat.com   : ok=2    changed=0    unreachable=0    failed=0   

ubuntu@magna003:/usr/share/ceph-ansible$

Created attachment 1331938 [details]
File contains ansible-playbook log and contents of all.yml

Hi Sebastien, 

Today I faced similar issue in containerized scenario. 
Task - ceph-defaults : set_fact monitor_name ansible_hostname

Regards,
Vasishta

Fixed, will be in rc13.

Created attachment 1333611 [details]
File contains contents ansible-playbook log

Hi Guillaume,

Fix is working fine for configuration of mgrs, Thanks.

I faced similar issue while configuring nfs, in task ceph-defaults : set_fact monitor_name ansible_hostname. Out of curiosity, I tried the fix in PR 1961 and changed the contents in site-docker.yml, but I faced same issue in task ceph-defaults : generate cluster fsid.

Moving BZ back to ASSIGNED state, please let me know if there are any concerns.

Regards,
Vasishta

Could you please test with https://github.com/ceph/ceph-ansible/pull/1975 ?

Thanks

Hi Sebastien, 

I tried using nfs-site branch, working fine for me.

Regards,
Vasishta

Thanks! This will be in rc15.

Created attachment 1334751 [details]
rolling update log

Please don't send fragments of the play but the entire one.
By looking at your log I don't know at which stage you're at.

Can you provide more info please?
Thanks.

fixed upstream here : https://github.com/ceph/ceph-ansible/commit/8fb68297a2b456f626655919ff7ba347e6bc4a36

will be in rc17

Working fine for me.
Checked with Madhavi (mkasturi) and Parikshith and Got to know that this is working fine.

Moving to VERIFIED state.

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2017:3387