Skip to content

Latest commit

 

History

History
287 lines (193 loc) · 7.9 KB

README.md

File metadata and controls

287 lines (193 loc) · 7.9 KB
Raw

manage_touchstone_tools

manage_touchstone_tools role is for managing the use of the touchstone system performance tools. For more information on the touchstone-tools, visit the project homepage: https://gitlab.com/touchstone/touchstone-tools.

There are eight tasks that this role can perform:

  1. Start collecting system stats
  2. Stop collecting system stats
  3. Start collecting database stats
  4. Stop collecting database stats
  5. Process pidstat data
  6. Plot sar data
  7. Plot pidstat data
  8. Plot database data

Requirements

Following are the requirements of this role.

  1. Ansible
  2. edb_devops.edb_postgres -> setup_repo role for setting the repository on the systems.
  3. edb_devops.edb_postgres -> setup_touchstone_tools role for installing the touchstone appimage.

Role Variables

The variables that can be configured and are available in the:

Each task within this role has a different set of required variables. Below is the documentation for each of those tasks and their variables'.

start_system_stats

This task starts the collection of system statistics.

ts_output_dir

The output directory for the touchstone-tools data collection. This must be a directory that does not already exist.

sec_bw_sample

The number of seconds between samples, default 60.

How to include start_system_stats in your playbook

Below is an example of how to put start_system_stats in your playbook.

- name: Include start_system_stats tasks to start sysstat data collection
  ansible.builtin.include_role:
    name: manage_touchstone_tools
    tasks_from: start_system_stats
  vars:
    ts_output_dir: "/var/log/touchstone/system_stats"
    sec_bw_sample: 30

stop_system_stats

This task stops the collection of system statistics.

ts_output_dir

The output directory for the touchstone-tools data collection. This must be a directory that does not already exist.

How to include stop_system_stats in your playbook

Below is an example of how to put stop_system_stats in your playbook.

- name: Include stop_system_stats tasks to stop sysstat data collection
  ansible.builtin.include_role:
    name: manage_touchstone_tools
    tasks_from: stop_system_stats
  vars:
    ts_output_dir: "/var/log/touchstone/system_stats"

start_db_stats

This task starts the collection of database statistics.

ts_output_dir

The output directory for the touchstone-tools database data collection.

sec_bw_sample

The number of seconds between samples, default 60.

pg_database

The database name to connect to.

pg_server_hostname

Database server host or docker directory. Same as -h in psql commands.

pg_port

The database server port. Same as -p in psql commands.

pg_owner

Database connections username. Same as -U in psql commands.

How to include start_db_stats in your playbook

Below is an example of how to put start_db_stats in your playbook.

# if vars pg_database, pg_port, pg_owner are already set by ansible, they are not required to be passed in
- name: Include start_db_stats tasks to start pgsql stats data collection
  ansible.builtin.include_role:
    name: manage_touchstone_tools
    tasks_from: start_db_stats
  vars:
    ts_output_dir: "/var/log/touchstone/db_stats"
    sec_bw_sample: 15
    pg_database: "postgres"
    pg_server_hostname: "primary1"
    pg_port: 5432
    pg_owner: "postgres"

stop_db_stats

This task stops the collection of database statistics.

ts_output_dir

The output directory for the touchstone-tools database data collection.

How to include stop_db_stats in your playbook

Below is an example of how to put stop_db_stats in your playbook.

- name: Include stop_db_stats tasks to stop pgsql stats data collection
  ansible.builtin.include_role:
    name: manage_touchstone_tools
    tasks_from: stop_db_stats
  vars:
    ts_output_dir: "/var/log/touchstone/db_stats"

process_pidstat_data

This task processes the pidstat.txt file generated during system stat collection. This task generates the pidstat.csv file and places in the same directory as the pidstat.txt.

pidstat_txt_file

The file location of the pidstat.txt. This file is generated during sysstat collection. Usually placed within the ts_output_file directory used during stop_system_stats. If not specified, it will use the ts_output_file directory to generate, but this may not be an accurate location.

How to include process_pidstat_data in your playbook

Below is an example of how to put process_pidstat_data in your playbook.

- name: Include process_pidstat_data tasks to start sysstat data collection
  ansible.builtin.include_role:
    name: manage_touchstone_tools
    tasks_from: process_pidstat_data
  vars:
    pidstat_txt_file: "/var/log/touchstone/system_stats/pidstat.txt"

plot_sar_data

This task plots the generated sar data.

ts_output_dir

The directory of the ts-sysstat processed sar data. This location is the output directory (ts_output_dir) from the stop_system_stats tasks.

ts_plot_output_dir

The directory for the generated plots to be placed. Plotting the sar data will generate new directories within this directory. Defaults to ts_output_dir if not value specified.

plot_size

The dimensions of the generated plots. Default 1600,1000.

How to include plot_sar_data in your playbook

Below is an example of how to put plot_sar_data in your playbook.

- name: Include plot_sar_data tasks to plot sar data
  ansible.builtin.include_role:
    name: manage_touchstone_tools
    tasks_from: plot_sar_data
  vars:
    ts_output_dir: "/var/log/touchstone/system_stats"
    ts_plot_output_dir: "/var/log/touchstone/system_stats"
    plot_size: "800,500"

plot_pidstat_data

This task plots the generated and processed pidstat data in pidstat.csv.

pidstat_csv_file

The location of the processed pidstat.csv file.

ts_plot_output_dir

The directory for the generated plots to be placed.

plot_size

The dimensions of the generated plots. Default 1600,1000.

How to include plot_pidstat_data in your playbook

Below is an example of how to put plot_pidstat_data in your playbook.

- name: Include plot_pidstat_data tasks to plot pidstat data
  ansible.builtin.include_role:
    name: manage_touchstone_tools
    tasks_from: plot_pidstat_data
  vars:
    pidstat_csv_file: "/var/log/touchstone/system_stats/pidstat.csv"
    ts_plot_output_dir: "/var/log/touchstone/system_stats/pidstat_plots"
    plot_size: "1600,1000"

plot_db_data

This task plots the generated database stats.

pg_databse

The database name to plot the stats from.

ts_output_dir

The directory of the ts-pgsql-stat collected data. This location is the output directory (ts_output_dir) from the stop_db_stats tasks.

ts_plot_output_dir

The directory for the generated plots to be placed. Defaults to ts_output_dir if not value specified.

plot_size

The dimensions of the generated plots. Default 1600,1000.

How to include plot_db_data in your playbook

Below is an example of how to put plot_db_data in your playbook.

- name: Include plot_db_data tasks to plot pgsql data
  ansible.builtin.include_role:
    name: manage_touchstone_tools
    tasks_from: plot_db_data
  vars:
    pg_database: "postgres"
    ts_output_dir: "/var/log/touchstone/db_stats"
    ts_plot_output_dir: "/var/log/touchstone/db_stats/pgsql_plots"
    plot_size: "800,500"

License

BSD

Author information

Author: