Platform log files

VMmanager saves information about the operation of services in logs. The data from logs can be used to diagnose the platform operation. Services use various methods for storing logs:

• files inside a Docker container;
• Docker container's stdout;
• logs from the journald system service.

Logs list

Main log files

Most of the logs for the main container vm_back are stored in journald.

Some of the logs are stored in the vm_back container in the /var/log/ directory. Usually the name of the log file is the same as the service name. For example, the log file of the vmwatch service is called vmwatch.log.

Platform installation

The platform installation log is saved in the file /opt/ispsystem/vm/install.log on the platform server.

Services logged in journald

The following container logs are stored in journald:

• alert;
• auth_back4;
• back_v4;
• batch;
• customizer;
• dr;
• ipmgr;
• license;
• metric_collector;
• msgsender;
• nodewarden;
• notifier_v4;
• plugin_v4;
• san;
• taskmgr;
• updater;
• vault;
• vm_back;
• vmr.

The log identifier matches the container name. For more information on the purpose of containers, see VMmanager structure.

Various services

Log files on the cluster node

Working with logs in containers

Copy log file from a container

To copy a log file from the container to the current directory, run the command:

docker cp <container_name>:<path_to_log> ./

docker cp auth:/var/log/licupdate.log ./

View log files in a container

1. Enter the required container: 

   docker exec -it <container_name> sh

   Comments to the command

   <container_name> — container name
   Example command

   docker exec -it vm_back sh

2. View the log file using standard Linux utilities. For example, cat, tail, less, etc.

   Example command

   less /var/log/host.log

   To view the logs of a specific task:

   1. In the platform interface go to the Tasks → select the task → copy the request_id value.

   2. Run the command:

      grep <request_id> /var/log/*

      Comments to the command

      <request_id> — request_id value

3. Exit the container:

   exit

Viewing log files via stdout

Separate log files are not created for some services. Logs of these services are written to stdout. To view such logs, run the command: 

docker logs <container_name> --tail <lines> --since <period>

docker logs vm_back --tail 100 --since 60m

Working with logs in journald

To view the journald logs, enter the following command: 

journalctl

In the command options, you can specify filter criteria:

• COMPANY=isp — display the logs for all containers;
• CONTAINER_NAME=<имя_контейнера> — display the logs for a specific container;
• SYSLOG_IDENTIFIER=<идентификатор> — display the logs for a specific service;
• REQUEST_ID=<id_задачи> — display the logs for a specific task;
• --since "<дата>" — start of the period. For example, --since "2026-09-30 10:00:00", --since yesterday, --since "1 hour ago";
• --until "<дата>" — end of the period. For example, --since "2026-09-30 10:00:00", --since yesterday, --since "1 hour ago";
• -o json — display in JSON format;
• -o verbose — detailed output;
• -n — number of lines;
• -p — priority. Possible options:
  • 0 — System is down;
  • 1 — Immediate action required;
  • 2 — Critical condition;
  • 3 — Error;
  • 4 — Warning;
  • 5 — Notification;
  • 6 — Info;
  • 7 — Debug.

Read more about command parameters in the official documentation.

Example commands:

journalctl COMPANY=isp --since yesterday

journalctl CONTAINER_NAME=vm_back -p 3

journalctl SYSLOG_IDENTIFIER=vm_reader -n 100

journalctl REQUEST_ID=1234 -o json

Collecting platform logs

Collecting with the built-in service

To collect all the log files in one directory:

1. Connect to the server with the platform via SSH. For more information about connecting via SSH, see Workstation setup.

2. Run one of the following commands: 

   Collect logs from the platform and all nodes

   export SKIP_NODES_LOGS=0 && vm collect-logs

   Collect logs from the platform only

   export SKIP_NODES_LOGS=1 && vm collect-logs

   The log collection process for installations with a large number of nodes consumes a significant amount of RAM. Therefore, if you do not currently need the node logs, we recommend running the log collection only from the platform.

   You can collect logs from a specific date. To do this, specify the -s option: 

   vm collect-logs -s <date>

   Comments:

   • <date> — log start date in the format YYYY-MM-DD (for example, 2025-12-30) or ‘YYYY-MM-DD HH:MM:SS’ (for example, ‘2025-12-30 23:45:55’)

The log files will be saved in the /home/logs/ directory, the log archive will be saved in the /home/logs_YYYY_MM_DD_HH_MM_SS.tar.gz file.

The collected logs include:

• the contents of Docker logs for all running containers;
• all journal entries from the platform server that contain the COMPANY=isp attribute;
• platform installation logs /opt/ispsystem/vm/install.log.

Collecting with your own script

To collect all the log files of the platform in one directory:

1. Connect to the server with the platform via SSH.

2. Create the bash script logs_collect.sh with the following content:

   #!/bin/bash
   
   rm /home/logs -fr
   
   DOCKER_CONTAINER_NAMES=`docker ps --format '{{.Names}}'`
   
   SERVICES=($DOCKER_CONTAINER_NAMES)
   cd /home
   mkdir -p logs
   cd logs
   for service in ${SERVICES[@]}
   do
    echo -e "----\033[0;31mCopying logs from $service\033[0m----\n"
    mkdir -p $service
    docker cp $service:/var/log/. $service/.
    docker logs $service > $service/${service}_stdout.log 2>&1
   done
   
   journalctl COMPANY=isp > journal_isp.log 2>&1
   
   cp -r /opt/ispsystem/*/install.log install.log
   
   ARCHIVE_NAME="logs_$(date +'%Y_%m_%d_%H_%M_%S').tar.gz"
   
   tar -cvzf /home/$ARCHIVE_NAME -C /home/logs .

3. Run the script: 

   bash logs_collect.sh

The log files will be saved in the /home/logs/ directory, the log archive will be saved in the /home/ directory.