Deadman Check
A monitoring companion for Nomad periodic jobs that alerts if periodic isn't running at the expected interval.
The deadman-check has 2 modes:
Run with the Nomad periodic job as an additional task to update a key in Consul with current EPOCH time.
Run as a separate process that will monitor the Consul key's EPOCH time value and alert if that value fails to meet a time 'freshness' threshold that is expected for that job.
Requirements
- Consul instance
- Alerting requires a SLACK_API_TOKEN environment variable to be set (use Slack Bot integration)
Example Usage
Let's say I have a Nomad periodic job that is set to run every 10 minutes. The Nomad configuration looks like this:
job "SilverBulletPeriodic" {
datacenters = ["dc1"]
type = "batch"
periodic {
cron = "*/10 * * * * *"
prohibit_overlap = true
}
group "utility" {
task "SilverBulletPeriodicProcess" {
driver = "docker"
config {
image = "silverbullet:build_1"
work_dir = "/utility/silverbullet"
command = "blaster"
}
resources {
cpu = 100
memory = 500
}
}
}
}
To monitor the SilverBulletPeriodicProcess task let's add a deadmad-check task to run post updates to a Consul endpoint (10.0.0.1 for this example)
job "SilverBulletPeriodic" {
datacenters = ["dc1"]
type = "batch"
periodic {
cron = "*/10 * * * * *"
prohibit_overlap = true
}
group "silverbullet" {
task "SilverBulletPeriodicProcess" {
driver = "docker"
config {
image = "silverbullet:build_1"
work_dir = "/utility/silverbullet"
command = "blaster"
}
resources {
cpu = 100
memory = 500
}
}
task "DeadmanSetSilverBulletPeriodicProcess" {
driver = "docker"
config {
image = "sepulworld/deadman-check"
command = "key_set"
args = [
"--host",
"10.0.0.1",
"--port",
"8500",
"--key",
"deadman/SilverBulletPeriodicProcess"]
}
resources {
cpu = 100
memory = 256
}
}
}
}
Now the key, deadman/SilverBulletPeriodicProcess, at 10.0.0.1 will be updated with the EPOCH time for each SilverBulletPeriodic job run. If the job hangs or fails to run we will know via the EPOCH time entry going stale.
Next we need a job that will run to monitor this key.
job "DeadmanMonitoring" {
datacenters = ["dc1"]
type = "service"
group "monitor" {
task "DeadmanMonitorSilverBulletPeriodicProcess" {
driver = "docker"
config {
image = "sepulworld/deadman-check"
command = "switch_monitor"
args = [
"--host",
"10.0.0.1",
"--port",
"8500",
"--key",
"deadman/SilverBulletPeriodicProcess",
"--freshness",
"800",
"--alert-to",
"slackroom",
"--daemon",
"--daemon-sleep",
"900"]
}
resources {
cpu = 100
memory = 256
}
env {
SLACK_API_TOKEN = "YourSlackApiToken"
}
}
}
}
Monitor a Consul key that contains an EPOCH time entry. Send a Slack message if EPOCH age hits given threshold
Non-Nomad Use:
Local system installation
execute:
$ bundle install
$ gem install deadman_check
Install and run deadman-check from Docker
# Optional: If you don't pull explicitly, `docker run` will do it for you
$ docker pull sepulworld/deadman-check
$ alias deadman-check='\
docker run \
-it --rm --name=deadman-check \
sepulworld/deadman-check'
(Depending on how your system is set up, you might have to add sudo in front of the above docker commands or add your user to the docker group).
If you don't do the docker pull, the first time you run deadman-check, the docker run command will automatically pull the sepulworld/deadman-check image on the Docker Hub. Subsequent runs will use a locally cached copy of the image and will not have to download anything.
Usage via Local System Install
$ deadman-check -h
NAME:
deadman-check
DESCRIPTION:
Monitor a Consul key that contains an EPOCH time entry.
Send a Slack message if EPOCH age hits given threshold
COMMANDS:
help Display global or [command] help documentation
key_set Update a given Consul key with current EPOCH
switch_monitor Target a Consul key to monitor
GLOBAL OPTIONS:
-h, --help
Display help documentation
-v, --version
Display version information
-t, --trace
Display backtrace when an error occurs
Usage for key_set command
$ deadman-check key_set -h
NAME:
key_set
SYNOPSIS:
deadman-check key_set [options]
DESCRIPTION:
EXAMPLES:
# Update a Consul key deadman/myservice, with current EPOCH time
deadman-check key_set --host 127.0.0.1 --port 8500 --key deadman/myservice
OPTIONS:
--host HOST
IP address or hostname of Consul system
--port PORT
port Consul is listening on
--key KEY
Consul key to monitor
Usage for switch_monitor command
$ deadman-check switch_monitor -h
NAME:
switch_monitor
SYNOPSIS:
deadman-check switch_monitor [options]
DESCRIPTION:
EXAMPLES:
# Target a Consul key deadman/myservice, and this key has an EPOCH
value to check looking to alert on 500 second or greater freshness
deadman-check switch_monitor \
--host 127.0.0.1 \
--port 8500 \
--key deadman/myservice \
--freshness 500 \
--alert-to slackroom
OPTIONS:
--host HOST
IP address or hostname of Consul system
--port PORT
port Consul is listening on
--key KEY
Consul key to monitor
--freshness SECONDS
The value in seconds to alert on when the recorded
EPOCH value exceeds current EPOCH
--alert-to SLACKROOM
Slackroom to alert to, don't include the # tag in name
--daemon
Run as a daemon, otherwise will run check just once
--daemon-sleep SECONDS
Set the number of seconds to sleep in between switch checks, default 300
Development
After checking out the repo, run bin/setup to install dependencies. Then, run rake test to run the tests. You can also run bin/console for an interactive prompt that will allow you to experiment.
To install this gem onto your local machine, run bundle exec rake install. To release a new version, update the version number in version.rb, and then run bundle exec rake release, which will create a git tag for the version, push git commits and tags, and push the .gem file to rubygems.org.
Contributing
Bug reports and pull requests are welcome on GitHub at https://github.com/[USERNAME]/deadman_check. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the Contributor Covenant code of conduct.
License
The gem is available as open source under the terms of the MIT License.