Proposal that alternate path should auto create the path in the filesystem.

[tabowling]

When testing "path: /var/mytest/crash" as an alternate dump location, kdump fails to restart because the path does not exist. However, the value is written to the /etc/kdump.conf file.

I propose that when specifying an alternate path, we attempt to create that alternate with same permissions and SELinux context as /var/crash.

If we cannot properly establish the path, then we should fail and not write the value to the new kdump.config file.

Playbook output

TASK [linux-system-roles.kdump : Generate /etc/kdump.conf] ***********************************************************************************
task path: /etc/ansible/roles/linux-system-roles.kdump/tasks/main.yml:11
NOTIFIED HANDLER restart kdump
changed: [rhel7-latest] => {"backup_file": "/etc/kdump.conf.2685.2017-05-06@08:19:24~", "changed": true, "checksum": "65e4cd35d517dcb3b66171cf0b5e102dbc8dfab2", "dest": "/etc/kdump.conf", "gid": 0, "group": "root", "md5sum": "e51517147da1f4d016a19f04b93db815", "mode": "0644", "owner": "root", "size": 149, "src": "/root/.ansible/tmp/ansible-tmp-1494073163.48-101588653438446/source", "state": "file", "uid": 0}

TASK [linux-system-roles.kdump : Generate /etc/sysconfig/kdump] ******************************************************************************
task path: /etc/ansible/roles/linux-system-roles.kdump/tasks/main.yml:19
changed: [rhel7-latest] => {"backup_file": "/etc/sysconfig/kdump.2786.2017-05-06@08:19:24~", "changed": true, "checksum": "7c35057f663ca5165c0f7e266df635334d0356a3", "dest": "/etc/sysconfig/kdump", "gid": 0, "group": "root", "md5sum": "194138297db4b4029fd41ec574735520", "mode": "0644", "owner": "root", "size": 410, "src": "/root/.ansible/tmp/ansible-tmp-1494073164.11-88997200875910/source", "state": "file", "uid": 0}

RUNNING HANDLER [linux-system-roles.kdump : restart kdump] ***********************************************************************************
fatal: [rhel7-latest]: FAILED! => {"changed": false, "failed": true, "msg": "Unable to restart service kdump: Job for kdump.service failed because the control process exited with error code. See \"systemctl status kdump.service\" and \"journalctl -xe\" for details.\n"}
META: ran handlers
	to retry, use: --limit @/home/tbowling/src/virt-demo/ansible/example-linux-system-roles.kdump.retry

PLAY RECAP ***********************************************************************************************************************************
rhel7-latest               : ok=5    changed=3    unreachable=0    failed=1   

journald output

-- Unit kdump.service has begun starting up.
May 06 08:19:24 rhel7-latest kdumpctl[2856]: Detected change(s) in the following file(s):
May 06 08:19:24 rhel7-latest kdumpctl[2856]: /etc/kdump.conf
May 06 08:19:24 rhel7-latest kdumpctl[2856]: Rebuilding /boot/initramfs-3.10.0-514.el7.x86_64kdump.img
May 06 08:19:25 rhel7-latest kdumpctl[2856]: Dump path /var/mytest/crash does not exist.
May 06 08:19:25 rhel7-latest systemd[1]: kdump.service: main process exited, code=exited, status=1/FAILURE
May 06 08:19:25 rhel7-latest kdumpctl[2856]: mkdumprd: failed to make kdump initrd
May 06 08:19:25 rhel7-latest kdumpctl[2856]: Starting kdump: [FAILED]
May 06 08:19:25 rhel7-latest systemd[1]: Failed to start Crash recovery kernel arming.
-- Subject: Unit kdump.service has failed
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit kdump.service has failed.

File permissions to clone from /var/crash. Note: we may want to use our new selinux module to set the context.

[root@rhel7-latest ~]# ls -ld /var/crash/
drwxr-xr-x. 2 root root 6 Sep  5  2016 /var/crash/

[root@rhel7-latest ~]# ls -lZd /var/crash/
drwxr-xr-x. root root system_u:object_r:kdump_crash_t:s0 /var/crash/