Working with Kolibri from the Command Line¶
In Windows you need to open the command prompt, for example by using the WIN + R shortcut, and then typing
On macOS open Spotlight and type
Terminal. You may also need to prefix the commands with
python -m, for example
python -m kolibri start.
If you are running Kolibri with the
.pexfile, make sure to substitute the
kolibriin below commands with the exact name of the file you downloaded preceded by
./. For example, to start Kolibri from the downloaded file
In the commands below, angle brackets and the text between them
<...> are used to denote placeholders for you to modify. Make sure to replace them with your own information.
In case you need to troubleshoot potential problems while running Kolibri, you may try to start it manually from the command line.
kolibri start --debug --foreground
Import Content Channels from Internet¶
To import content channels from Internet, run these two commands in sequence. The first downloads the channel database, and the second downloads the resources (videos, documents, etc.).
kolibri manage importchannel -- network <Channel ID> kolibri manage importcontent -- network <Channel ID>
For example (
Channel ID without angle brackets
kolibri manage importchannel -- network a9b25ac9814742c883ce1b0579448337 kolibri manage importcontent -- network a9b25ac9814742c883ce1b0579448337
When you import content channels from the command line, you still must use the 32 digit channel ID, as the command will not work with the token. Make sure to receive the correct channel ID from the person who curated the unlisted channel you need to import, or refer to Kolibri Studio user guide how to find it in Studio user interface, if you have channel editor access.
Export Content Channels¶
To export Kolibri content channels on a local drive in order to share it with another device, run these two commands in sequence. The first exports the channel database, and the second exports the resources (videos, documents, etc.).
kolibri manage exportchannel -- <Channel ID> /path/to/local/drive/KOLIBRI_DATA kolibri manage exportcontent -- <Channel ID> /path/to/local/drive/KOLIBRI_DATA
The path should be to a folder named
KOLIBRI_DATA at the root of the local drive, so it will get picked up later for importing via the Web UI.
Reorder Content Channels¶
You can set the specific order for content channels in the Learn page according to your preferences. Follow these steps.
To view the current ordered list of channels, run the command:
kolibri manage listchannels
The output will be something like:
Pos ID Name --- -- ---- 1 95a52b386f2c485cb97dd60901674a98 CK-12 Testing 2 a9b25ac9814742c883ce1b0579448337 TESSA - Teacher Resources
To set a position for a channel, run the command:
kolibri manage setchannelposition <Channel ID> <Pos>
Example with the above channels:
kolibri manage setchannelposition a9b25ac9814742c883ce1b0579 1 Pos ID Name --- -- ---- 1 a9b25ac9814742c883ce1b0579448337 TESSA - Teacher Resources 2 95a52b386f2c485cb97dd60901674a98 CK-12 Testing
Create a New Super Admin¶
In case you need to create another super admin user, either to address additional need of managing facility, or if you lost the password for the old one, run the following command.
kolibri manage createsuperuser
You will be prompted to input the Username and Password and the new super admin user account will be created.
Import Users from a CSV File¶
This is currently an experimental feature, so please forward to the development team any details about the issues you may encounter while using it.
Command works on Kolibri version 0.9 and above.
CSV File Structure¶
To import users into Kolibri with this command, you will need to provide the user data in a CSV (comma separated values) file format. You can export a CSV file from a tabular data in any spreadsheet program (Excel, Google Sheets, LibreOffice Calc, etc.).
Header row is optional, but if you do not include it, Kolibri will assume that you are providing the data in the following order:
If you do include a header row, you can provide less data, or put them a different order:
When you do not provide passwords for the imported users, Kolibri will set the default password
kolibrifor those usernames.
The facility can be either the facility name or the facility ID. If you do not provide the facility, Kolibri will import users in the default facility on the device. You can also specify the facility by adding the
--facilityargument in the command line (see below).
kolibri manage importusers your-csv-file.csv kolibri manage importusers your-csv-file.csv --facility <your-facility>
Change User’s Password¶
Run the following command to change the password for a user.
kolibri manage changepassword <username>
You will be prompted twice to input the new password for the user.
Delete Users Permanantly¶
If you need to permanently delete a Kolibri user and all the data associated with their account, for example to ensure privacy rights according to GDPR, use the following command.
kolibri manage deleteuser <username>
This will permanently erase all the user data.
kolibri language setdefault <langcode>
|Spanish (Latin America)||
Backup and Restore Kolibri Database¶
Kolibri automatically creates a backup of the database with every version upgrade. If for some reason you need to make a manual backup, use the following command.
kolibri manage dbbackup
This command will create a time-stamped
.dump file in the
./kolibri/backups folder that you can use to restore the database with the following command.
kolibri manage dbrestore --latest
To restore the DB from a specific
.dump file, use the flag
--select to see all that available sorted by date, and select the one you need.
kolibri manage dbrestore --select
This command is not intended for replication across different devices, but only for restoring on a single device from a local backup of the database.
Change the Location of Kolibri Content Files¶
Kolibri content channels may occupy a considerable amount of hard disk space over time. If you have concerns about running out of storage on your device, you can move the Kolibri content files to another drive.
If you have both SSD disk and HDD disk available on your device, it is recommended to install Kolibri on the SSD drive to allow faster access to the database, and move just the content file to the HDD drive.
To move the Kolibri content folders to another location, follow these steps.
- Stop Kolibri.
- Create a new folder that will contain all the content files and resources on the destination drive.
kolibri manage content movedirectory <destination>
For example, if you created a new folder
KolibriContenton an external drive, run this command.kolibri manage content movedirectory /mnt/my_external_drive/KolibriContent
If you are on Windows, and the new folder
KolibriContentis on the drive
F:, run this command.kolibri manage content movedirectory F:\KolibriContent
- Restart Kolibri.
This command will move the 2 subfolders
storage, from their default location inside the
.kolibri/content folder in your device’s home directory, to a new location you specified in the command.
Change the Location of ALL Kolibri Files¶
If you want to change the directory where all of Kolibri’s runtime files are located, together with the imported content channels, you need to change the environment variable called
KOLIBRI_HOME to the path of your choice.
If the variable is left unset, by default, Kolibri’s runtime files and content will be placed in your user’s home folder, under the
Adjusting this environment variable behaves differently than the
movedirectory command above:
- Adjusting the environment variable will not automatically migrate over data. You need to copy the
.kolibrifolder manually to the new location.
- If you do copy the
.kolibrifolder, the content will not be affected if it had been previously set using the
There are many ways to set an environment variable either temporarily or permanently. To start Kolibri on OSX or Linux with a different home, follow these steps.
- Stop the server.
- Move the
.kolibrifolder to the new location.
- Run the following in Terminal:
KOLIBRI_HOME=/path/to/new/home kolibri start
When you start the server again, all your files should be seamlessly detected at that location.
To change the environment variable
KOLIBRI_HOME on Windows, follow these steps.
- Stop the server.
- Move the
.kolibrifolder to the new location.
- Run the following in Command Prompt:
setx KOLIBRI_HOME "/path/to/new/home"
Restart the server, and your files should be seamlessly detected at the new location.
Alternatively, you can follow these steps in the GUI.
Go to Computer > Advanced System Settings and press the Environment Variables button.
Under User Variables for… press the New… button.
KOLIBRI_HOMEin the Variable name field, and your new path in the Variable value field, and press OK on both open windows.
Customize Kolibri Settings with the OPTIONS.INI File¶
For certain configuration settings you need to use the
options.ini file. Installing Kolibri does not generate this file by default, but you can easily add one yourself. Follow these steps.
Open the preferred text editor on your computer (eg. Notepad on Windows).
Write the required sections and keys (see details for available settings below) in the following format:
[section] key1 = a key2 = b
- Save the resulting
options.inifile in the
.kolibrifolder inside the Home folder.
options.ini file can contain several sections with one or more associated keys, depending on the requirements of your installation.
Run Kolibri from a Different Port¶
If you need Kolibri to start and run from a port different than the default
8080, add the section
[Deployment], and the key
HTTP_PORT with the value of your desired port, to the
[Deployment] HTTP_PORT = 1234 # Substitute 1234 with your desired port number
If after setting the desired port in the
options.ini file you still see Kolibri running from a different one, you probably have the environment variable
KOLIBRI_HTTP_PORT from a previous installation, which takes precedence. Check the
.bashrc file on Linux, or run the
set command in Windows command prompt, to verify and correct if necessary.
Test Kolibri Server Performance¶
You can use the following command to collect information about the device where Kolibri server is running, and details about how much of its resources it is using. This command displays a snapshot of the server state at the time the command is executed, and its output will vary depending on the current server load. In case you suspect performance problems, type this in the Terminal or Command prompt.
kolibri manage benchmark
The command will have an output similar to this:
Take a screenshot of the Terminal or Command prompt, or copy and paste the output in the community forum post.
In order to collect more than a current snapshot of Kolibri server performance, you can use the profiling command. When executed, the command will collect a series of performance indicators every 10 seconds and save them in a CSV file. Type this in the Terminal or Command prompt.
kolibri manage profile
Command collects and saves the information 60 times by default. If you want to change this value, add the
--num-samples flag with the desired number at the end.
kolibri manage profile --num-samples=100
Each log line contains this information:
- Date and time of each command execution
- Number of Kolibri active sessions (including guest sessions)
- Number of Kolibri logged users
- Number of Kolibri user interactions during the last minute
- Total percentage of CPU use
- Total memory use
- Total available memory
- Number of processes executed in the server
- Percentage of CPU used by Kolibri
- Percentage of memory used by Kolibri
To help us troubleshoot potential problems on your Kolibri server, locate and send us the
Profile Server Requests¶
If you have the
[Server] section of the OPTIONS.INI file configured with
PROFILE = 1, the above command will additionally perform a profiling of every request made by Kolibri server, and save the results in a second log file as
Each log line contains this information:
- Request path
- Time spent processing the request
- Memory (in KB) used by the Kolibri process when the request came in
- Memory (in KB) used by the Kolibri process when the response was sent
- CPU percentage used by the Kolibri process when the request came in
- CPU percentage used by the Kolibri process when the request was sent
- Flag indicating if the request is the slowest one since the analysis started
Profiling server requests can consume a lot of computer resources, and potentially slow it down. For this reason you need to explicitly allow it in the
options.ini file. Without the
PROFILE = 1 key, command will not profile server requests (but just the current server state), and it will not create the second CSV file.