Bug 2227676

Description of problem:

In RHOSP 16.2.5, the glance_api_cron container is created even though the image cache feature of glance is not enabled[1]

In that case, tripleo_glance_api_cron_healthcheck.service will fail if the Image cache of the glance is not enabled.[2]

It is assumed that this may be due to the fact that cron jobs are not created in the glance_api_cron container if the glance image cache is not enabled.[3]


[1] 
[root@control01tbmoc ~]# podman ps|grep glance
f873fc92a78e  undercloudtb.ctlplane.localdomain:8787/rhosp-rhel8/openstack-glance-api:16.2                 kolla_start           2 weeks ago     Up 2 weeks ago             glance_api
3f9bfe1b37e0  undercloudtb.ctlplane.localdomain:8787/rhosp-rhel8/openstack-glance-api:16.2                 kolla_start           2 weeks ago     Up 2 weeks ago             glance_api_cron


/var/lib/config-data/puppet-generated/glance_api/etc/glance/glance-api.conf:
~~~
...
[paste_deploy]
...
flavor=keystone
...
~~~

[2]

[root@control01tbmoc ~]# systemctl status tripleo_glance_api_cron_healthcheck.service
● tripleo_glance_api_cron_healthcheck.service - glance_api_cron healthcheck
   Loaded: loaded (/etc/systemd/system/tripleo_glance_api_cron_healthcheck.service; disabled; vendor preset: disabled)
   Active: failed (Result: exit-code) since Wed 2023-07-26 10:32:33 +08; 31s ago
  Process: 418699 ExecStart=/usr/bin/podman exec --user root glance_api_cron /usr/share/openstack-tripleo-common/healthcheck/cron glance (code=exited, status=1/FAILURE)
 Main PID: 418699 (code=exited, status=1/FAILURE)

[3]
[root@control01tbmoc ~]# podman exec -it glance_api_cron crontab -l
no crontab for root
WARN[0000] Error resizing exec session 9ba6504bcfb042f73bb11837a07ab5ebd3fd0933b259fb5382a446f14160ec8c: could not open ctl file for terminal resize for container 3f9bfe1b37e0e1c8b51989b0151bde4382e41f14c8f841127eafa6be79ea451c: open /var/lib/containers/storage/overlay-containers/3f9bfe1b37e0e1c8b51989b0151bde4382e41f14c8f841127eafa6be79ea451c/userdata/9ba6504bcfb042f73bb11837a07ab5ebd3fd0933b259fb5382a446f14160ec8c/ctl: no such device or address

[root@control01tbmoc ~]# podman exec -it glance_api_cron cat /usr/share/openstack-tripleo-common/healthcheck/cron
#!/bin/bash


file="${1:-root}"
if [ -f /var/spool/cron/${file} ]; then
    nb_lines=$(grep -cEv '^#' /var/spool/cron/${file})
    if [ $nb_lines -ge 2 ]; then
        exit 0
    fi
fi
exit 1
WARN[0000] Error resizing exec session f40464e695091fc8a10cb04f34f5b8c5d3cad262af1c90cbec2859fd7ecd5551: could not open ctl file for terminal resize for container 3f9bfe1b37e0e1c8b51989b0151bde4382e41f14c8f841127eafa6be79ea451c: open /var/lib/containers/storage/overlay-containers/3f9bfe1b37e0e1c8b51989b0151bde4382e41f14c8f841127eafa6be79ea451c/userdata/f40464e695091fc8a10cb04f34f5b8c5d3cad262af1c90cbec2859fd7ecd5551/ctl: no such device or address



Version-Release number of selected component (if applicable):
puppet-glance-15.5.0-2.20220804175403.d54e942.el8ost.noarch

How reproducible:

This always happens when Image cache is disabled in RHOSP16.2.5

Actual results:
tripleo_glance_api_cron_healthcheck.service fail

Expected results:
tripleo_glance_api_cron_healthcheck.service does not fail



Additional info:

found this bugzilla similar to the error,

https://bugzilla.redhat.com/show_bug.cgi?id=2159566 

but the bugzilla is for RHOSP version 16.2.4 and the resolution is to upgrade to 16.2.5

The solution cannot be applied here as the customer environment is already 16.2.5