Browse Source

Merge "Revert "Adapt container health check for built-in podman health checks"" into stable/train

changes/47/758447/6
Zuul 3 months ago
committed by Gerrit Code Review
parent
commit
1115e028df
2 changed files with 13 additions and 101 deletions
  1. +0
    -92
      container_config_scripts/monitoring/collectd_check_health.py
  2. +13
    -9
      deployment/metrics/collectd-container-puppet.yaml

+ 0
- 92
container_config_scripts/monitoring/collectd_check_health.py View File

@ -1,92 +0,0 @@
#!/usr/bin/env python3
#
# Copyright 2018 Red Hat Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License"); you may
# not use this file except in compliance with the License. You may obtain
# a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
# License for the specific language governing permissions and limitations
# under the License.
import datetime
import re
import sys
HCLOG = '/var/log/collectd/healthchecks.stdout'
START_RE = re.compile(
r'(?P<timestamp>\w{3} \d{2} \d{2}\:\d{2}\:\d{2}) (?P<host>[\w\-\.\:]*) systemd\[.*\]: Started /usr/bin/podman healthcheck run (?P<container_id>\w*)')
EXEC_RE = re.compile(
r'(?P<timestamp>\w{3} \d{2} \d{2}\:\d{2}\:\d{2}) (?P<host>[\w\-\.\:]*) podman\[(?P<pid>\d*)\]: (?P<trash>.*) container exec (?P<container_id>\w*) \(.*name=(?P<container_name>\w*).*\)')
RESULT_RE = re.compile(
r'(?P<timestamp>\w{3} \d{2} \d{2}\:\d{2}\:\d{2}) (?P<host>[\w\-\.\:]*) podman\[(?P<pid>\d*)\]: (?P<result>(un)?healthy)')
def process_healthcheck_output(path_to_log):
"""Process saved output of health checks and returns list of unhealthy
containers.
"""
data = {}
pid_map = {}
with open(path_to_log, "r+") as logfile:
for line in logfile:
match = START_RE.search(line)
if match:
item = data.setdefault(match.group('container_id'), {})
item['timestamp_start'] = match.group('timestamp')
item['host'] = match.group('host')
continue
match = EXEC_RE.search(line)
if match:
item = data.setdefault(match.group('container_id'), {})
item['container_name'] = match.group('container_name')
item['host'] = match.group('host')
item['pid'] = match.group('pid')
pid_map[match.group('pid')] = match.group('container_id')
continue
match = RESULT_RE.search(line)
if match:
if match.group('pid') not in pid_map:
continue
item = data[pid_map[match.group('pid')]]
item['result'] = match.group('result')
if 'timestamp_start' not in item:
continue
try:
start = datetime.datetime.strptime(item['timestamp_start'],
'%b %d %H:%M:%S')
end = datetime.datetime.strptime(match.group('timestamp'),
'%b %d %H:%M:%S')
item['duration'] = (end - start).seconds
except Exception as ex:
err = "[WARN] Failure during calculating duration: {}"
print(err.format(ex))
continue
logfile.truncate()
# truncate the file
with open(HCLOG, "w") as logfile:
pass
unhealthy = []
for container in data.values():
if 'result' not in container:
continue
if container['result'] == 'healthy':
continue
log = ('{container_name}: Container health check on host {host} '
'results as {result} after {duration}s.')
unhealthy.append(log.format(**container))
return unhealthy
if __name__ == "__main__":
unhealthy = process_healthcheck_output(HCLOG)
if unhealthy:
print(' ; '.join(unhealthy))
sys.exit(2)

+ 13
- 9
deployment/metrics/collectd-container-puppet.yaml View File

@ -330,7 +330,18 @@ parameters:
default: true
CollectdContainerHealthCheckCommand:
type: string
default: "/scripts/collectd_check_health.py"
default: |
output=""
while read line ; do
i=$(echo $line | awk '//{gsub(/:/, "", $0); print $5}')
log=$(echo $line | awk '{split($0,a,/:\s+Error:\s+/); print a[2]}')
log=${log:0:-1}
output+=" ; ${i}: ${log}"
done < <(egrep "^[a-zA-Z]{3}\s+[0-9]{2}\s+[0-9\:]{8}\s+.*\s+.*:\s+[Ee]rror\:" /var/log/collectd/healthchecks.log)
truncate -s0 /var/log/collectd/healthchecks.log
if [ ! -z "${output}" ]; then
echo ${output:3} && exit 2;
fi
CollectdContainerHealthCheckInterval:
type: number
description: The frequency in seconds the docker health check is executed.
@ -644,12 +655,6 @@ outputs:
- path: /var/log/collectd
owner: collectd:collectd
recurse: true
container_config_scripts:
map_merge:
- {get_attr: [ContainersCommon, container_config_scripts]}
- collectd_check_health.py:
mode: "0755"
content: { get_file: ../../container_config_scripts/monitoring/collectd_check_health.py }
docker_config:
step_5:
collectd:
@ -671,7 +676,6 @@ outputs:
- /var/lib/config-data/puppet-generated/collectd:/var/lib/kolla/config_files/src:ro
- /var/log/containers/collectd:/var/log/collectd:rw,z
- /var/run/:/var/run:rw
- /var/lib/container-config-scripts:/scripts:ro
- /sys/fs/cgroup:/sys/fs/cgroup:ro
environment:
KOLLA_CONFIG_STRATEGY: COPY_ALWAYS
@ -699,7 +703,7 @@ outputs:
copy:
dest: /etc/rsyslog.d/openstack-healthcheck.conf
content: |
if ($programname startswith 'podman' and ($msg contains 'container exec' or $msg contains 'healthy')) or ($programname startswith 'systemd' and $msg contains 'podman healthcheck run') then -/var/log/containers/collectd/healthchecks.stdout
if $programname startswith 'healthcheck_' then -/var/log/containers/collectd/healthchecks.log
& stop
- name: Remove healthcheck log
when:


Loading…
Cancel
Save