keystone did not start

[Masaharu Miyamoto]

## issue
I retried "./stack.sh" after fixing boot process bug (id:1695822), however it stop when trying to access keystone api by curl.
This is because "sudo systemctl restart systemd-journald" is executed before in stack.sh and after running this command, keystone access is not available (Internal Server Error).
There is no problem for first run, but it happen second or later try while running keystone.
It is resolved restarting keystone service after restarting systemd-journald, however current service running command is a "start" but not a "restart".
Should we use "restart" for the Idempotency and to run the service securely ?

## log
$ ./stack.sh
~~~~~
~~~~~
+functions:wait_for_service:401 timeout 60 bash -x
++:: curl -g -k --noproxy '*' -s -o /dev/null -w '%{http_code}' http://172.23.0.244/identity/v3/
+functions:wait_for_service:401 rval=124
+functions:wait_for_service:406 time_stop wait_for_service
+functions-common:time_stop:2503 local name
+functions-common:time_stop:2504 local end_time
+functions-common:time_stop:2505 local elapsed_time
+functions-common:time_stop:2506 local total
+functions-common:time_stop:2507 local start_time
+functions-common:time_stop:2509 name=wait_for_service
+functions-common:time_stop:2510 start_time=1497033603
+functions-common:time_stop:2512 [[ -z 1497033603 ]]
++functions-common:time_stop:2515 date +%s
+functions-common:time_stop:2515 end_time=1497033663
+functions-common:time_stop:2516 elapsed_time=60
+functions-common:time_stop:2517 total=0
+functions-common:time_stop:2519 _TIME_START[$name]=
+functions-common:time_stop:2520 _TIME_TOTAL[$name]=60
+functions:wait_for_service:407 return 124
+lib/keystone:start_keystone:559 die 559 'keystone did not start'
+functions-common:die:186 local exitcode=0
+functions-common:die:187 set +o xtrace
[Call Trace]
./stack.sh:1088:start_keystone
/opt/stack/devstack/lib/keystone:559:die
[ERROR] /opt/stack/devstack/lib/keystone:559 keystone did not start
Error on exit
World dumping... see /opt/stack/logs/worlddump-2017-06-09-184105.txt for details
$

## On CentOS7.3
$ cat /etc/redhat-release
CentOS Linux release 7.3.1611 (Core)
$ uname -r
4.11.4-1.el7.elrepo.x86_64

[Masaharu Miyamoto (masmiyam)]

I confirmed that same error happens on placement-api if restart only keystone to fix this.
[Call Trace]
./stack.sh:1307:start_placement
/opt/stack/devstack/lib/placement:179:start_placement_api
/opt/stack/devstack/lib/placement:174:die
[ERROR] /opt/stack/devstack/lib/placement:174 placement-api did not start
Error on exit
World dumping... see /opt/stack/logs/worlddump-2017-06-10-000932.txt for details

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/472877

[OpenStack Infra (hudson-openstack)]

Change abandoned by Jens Harbott (frickler) (<email address hidden>) on branch: master
Review: https://review.openstack.org/472877
Reason: The patch has been in merge conflict for a long time, feel free to restore the patch and continue working on it

[Dr. Jens Harbott (j-harbott)]

No review in progress. Given the age of the bug report, it is unclear whether this is still a valid issue. Please update with fresh logs if this issue still persists.

[Launchpad Janitor (janitor)]

[Expired for devstack because there has been no activity for 60 days.]

[Hu1buerger (hu1buerger)]

This bug seems to still exists on fca44cc375657da3d1a20f16c5c04574c6234376

# What i did:
I called `./stack.sh` like [here](https://docs.openstack.org/devstack/latest/).
After some time the error (see above) occurs.

# What i tried:
1. reboot
2. clean install (ubuntu 22.04 LTS on x86_64)
3. reboot and rerun