Supported OS:
The process check lets you:
The process check is included in the Datadog Agent package, so you don’t need to install anything else on your server.
Unlike many checks, the process check doesn’t monitor anything useful by default; you must tell it which processes you want to monitor, and how.
While there’s no standard default check configuration, here’s an example process.d/conf.yaml that monitors ssh/sshd processes. See the sample process.d/conf.yaml for all available configuration options:
init_config:
instances:
- name: ssh
search_string: ['ssh', 'sshd']
# To search for sshd processes using an exact cmdline
# - name: ssh
# search_string: ['/usr/sbin/sshd -D']
# exact_match: True
Our process check uses the psutil python package to check processes on your machine. By default this process check works on exact match and looks at the process names only. By setting exact_match to False in your yaml file, the agent looks at the command used to launch your process and recognizes every process that contains your keywords.
You can also configure the check to find any process by exact PID (pid) or pidfile (pid_file). If you provide more than one of search_string, pid, and pid_file, the check uses the first option it finds in that order (e.g. it uses search_string over pid_file if you configure both).
Note For agent v6.11+ on windows, the agent runs as an unprivileged ddagentuser, and does not have access to the full command line of processes running under a different users, so the exact_match: false option cannot be used in such cases. The same goes for the user option that allows you to select only processes belonging to a specific user.
To have the check search for processes in a path other than /proc, set procfs_path: <your_proc_path> in datadog.conf, NOT in process.yaml (its use has been deprecated there). Set this to /host/proc if you’re running the Agent from a Docker container (i.e. docker-dd-agent) and want to monitor processes running on the server hosting your containers. You DON’T need to set this to monitor processes running in your containers; the Docker check monitors those.
Some process metrics require either running the datadog collector as the same user as the monitored process or privileged access to be retrieved.
Where the former option is not desired, and to avoid running the datadog collector as root, the try_sudo option lets the process check try using sudo to collect this metric.
As of now, only the open_file_descriptors metric on Unix platforms is taking advantage of this setting.
Note: the appropriate sudoers rules have to be configured for this to work
dd-agent ALL=NOPASSWD: /bin/ls /proc/*/fd/
See the example configuration for more details on configuration options.
Restart the Agent to start sending process metrics and service checks to Datadog.
Run the Agent’s status subcommand and look for process under the Checks section.
|
system.processes.cpu.pct (gauge) |
The CPU utilization of a process. shown as percent |
|
system.processes.cpu.normalized_pct (gauge) |
The normalized CPU utilization of a process. shown as percent |
|
system.processes.involuntary_ctx_switches (gauge) |
The number of involuntary context switches performed by this process. shown as event |
|
system.processes.ioread_bytes (gauge) |
The number of bytes read from disk by this process. In Windows: the number of bytes read. shown as byte |
|
system.processes.ioread_count (gauge) |
The number of disk reads by this process. In Windows: the number of reads by this process. shown as read |
|
system.processes.iowrite_bytes (gauge) |
The number of bytes written to disk by this process. In Windows: the number of bytes written by this process. shown as byte |
|
system.processes.iowrite_count (gauge) |
The number of disk writes by this process. In Windows: the number of writes by this process. shown as write |
|
system.processes.mem.page_faults.minor_faults (gauge) |
The number of minor page faults per second for this process. shown as occurrence |
|
system.processes.mem.page_faults.children_minor_faults (gauge) |
The number of minor page faults per second for children of this process. shown as occurrence |
|
system.processes.mem.page_faults.major_faults (gauge) |
The number of major page faults per second for this process. shown as occurrence |
|
system.processes.mem.page_faults.children_major_faults (gauge) |
The number of major page faults per second for children of this process. shown as occurrence |
|
system.processes.mem.pct (gauge) |
The process memory consumption. shown as percent |
|
system.processes.mem.real (gauge) |
The non-swapped physical memory a process has used and cannot be shared with another process. shown as byte |
|
system.processes.mem.rss (gauge) |
The non-swapped physical memory a process has used. aka "Resident Set Size". shown as byte |
|
system.processes.mem.vms (gauge) |
The total amount of virtual memory used by the process. aka "Virtual Memory Size". shown as byte |
|
system.processes.number (gauge) |
The number of processes. shown as process |
|
system.processes.open_file_descriptors (gauge) |
The number of file descriptors used by this process (only available for processes run as the dd-agent user) |
|
system.processes.open_handles (gauge) |
The number of handles used by this process. |
|
system.processes.threads (gauge) |
The number of threads used by this process. shown as thread |
|
system.processes.voluntary_ctx_switches (gauge) |
The number of voluntary context switches performed by this process. shown as event |
|
system.processes.run_time.avg (gauge) |
The average running time of all instances of this process shown as second |
|
system.processes.run_time.max (gauge) |
The longest running time of all instances of this process shown as second |
|
system.processes.run_time.min (gauge) |
The shortest running time of all instances of this process shown as second |
All metrics are per instance configured in process.yaml, and are tagged process_name:<instance_name>.
The Process check does not include any events.
process.up:
The Agent submits this service check for each instance in process.yaml, tagging each with process:<name>.
For an instance with no thresholds specified, the service check has a status of either CRITICAL (zero processes running) or OK (at least one process running).
For an instance with thresholds specified, consider this example:
instances:
- name: my_worker_process
search_string: ['/usr/local/bin/worker']
thresholds:
critical: [1, 7]
warning: [3, 5]
The Agent submits a process.up tagged process:my_worker_process whose status is:
Need help? Contact Datadog support.
To get a better idea of how (or why) to monitor process resource consumption with Datadog, check out our series of blog posts about it.
Mistake in the docs? Feel free to contribute!