Create a Support Package

This guide explains how to generate a support package for troubleshooting error scenarios.

This documentation applies to NGINX Management Suite Instance Manager 2.0.0 and later.


Use the support package script to collect information about your system for troubleshooting and debugging issues.

The script collects system and service information and then packages the data into a tar archive, which you can share with NGINX Customer Support.


The Instance Manager installer copies the script to the following location: /etc/nms/scripts/

To create a support package:

  1. Run the support package script. The script requires root privileges to run.

    sudo /etc/nms/scripts/

    The support package is saved in the same location from where the script was run (if no -t argument is passed).

    (Optional) If you use a different Instance Manager config file than the default /etc/nms/nms.conf file, run the support package script with the -c flag and specify the path to your config file:

    sudo /etc/nms/scripts/ -c /your/config.conf
  2. To extract the package, use the tar command:

    tar -xvf support-pkg-<timestamp>.tar.gz
The supported shell is bash.


The following table lists the arguments you can use with the support package script.

Short Long Description Example Default
-c --nms_config_path The path to the Instance Manager config file. -c /etc/nms/nms.conf /etc/nms/nms.conf
-h --help Prints information about the script arguments to stdout. --help N/A
-m --manager_log_path The directory where the Instance Manager log file is located. -m /var/log/nms /var/log/nms
-n --nginx_log_path The directory where the NGINX log files are located. -n /var/log/nginx /var/log/nginx
-o --ouput_dir The output directory where the tar archive is saved. -o ~/output $(pwd)
-t --target_host The Instance Manager address (host:port). -t

Package Contents

The support package includes several directories containing information about the system, service, and database state.


The access and error logs of the instances that Instance Manager monitors.

The access logs display the HTTP traffic for Instance Manager that’s routed by the NGINX instance. The error log contains NGINX errors that occurred during runtime.


The logs of the Instance Manager processes.

You can pipe the logs to grep to view entries belonging to only one of the three nms processes. For example, to view nms-core logs, run the following command:

cat nms.log | grep 'COR'

The following table shows the nms processes and pattern to grep on:

Process name Pattern
nms-core ‘COR’
nms-dpm ‘DPM’
nms-ingestion ‘ING’


Information about the Instance Manager and NGINX services running on the host. For each nms process and the nginx instance, the script collects:

  • journalctl (10000 most recent rows)
  • systemctl status


The status and state information of the host running Instance Manager, including the following:

  • System metrics (memory usage, CPU usage, etc.)
  • File permissions of the Instance Manager
  • Firewall or SELinux state
  • Network interfaces
  • Network information (hostname, iptables)
  • Environment variables
  • Disk usage of select directories
  • Operating system version
  • Installed Instance Manager version
  • Instance Manager license

dqlite snapshot

The support package script uses the -c flag ( or --nms_config_path) to get the Instance Manager configuration. If the configuration file is not specified, the script uses the default value /etc/nms/nms.conf.

If the Instance Manager configuration file does not specify addresses for the core and dpm databases, the default values are assumed: and

The support package script uses a small Go executable file called dqlite-backup (located in /etc/nms/scripts/) to connect to the databases and generate data dumps.

The collected data is saved to the directories dqlite/core, dqlite/dpm, and dqlite/acm (if you have API Connectivity Manager installed).


This folder contains status information, dumps, and statistics for the nms ClickHouse database. In particular, for metrics and events.