cryosparcm reference

How to use the cryosparcm utility for starting and stopping the CryoSPARC instance, checking status or logs, managing users and using CryoSPARC's command-line interface.

Access the command line utility, cryosparcm

The CryoSPARC master node hosts the web server and manages job resource allocation.

Workstations or master nodes with a cryosparc_master installation have access to cryosparcm, CryoSPARC's built-in utility for all administrative, management and advanced usage tasks.

To use it, log into the machine onto which . Open a Terminal running a shell (such as bash) and enter any of the commands described below.

cryosparcuser@emserver1:~$ cryosparcm status
----------------------------------------------------------------------------
CryoSPARC System master node installed at
/home/cryosparcuser/sw/cryosparc_master
Current cryoSPARC version: v4.3.1
----------------------------------------------------------------------------

CryoSPARC process status:

app                              RUNNING   pid 1599310, uptime 0:30:11
app_api                          RUNNING   pid 1599643, uptime 0:30:10
app_api_dev                      STOPPED   Not started
app_legacy                       STOPPED   Not started
app_legacy_dev                   STOPPED   Not started
command_core                     RUNNING   pid 1593955, uptime 0:30:26
command_rtp                      RUNNING   pid 1595620, uptime 0:30:16
command_vis                      RUNNING   pid 1594923, uptime 0:30:17
database                         RUNNING   pid 1593553, uptime 0:30:29

----------------------------------------------------------------------------
License is valid
----------------------------------------------------------------------------

global config variables:
export CRYOSPARC_LICENSE_ID="xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
export CRYOSPARC_MASTER_HOSTNAME=emserver1.internal
export CRYOSPARC_DB_PATH="/home/cryosparcuser/sw/cryosparc_db_61561"
export CRYOSPARC_BASE_PORT=61560
export CRYOSPARC_DB_CONNECTION_TIMEOUT_MS=20000
export CRYOSPARC_INSECURE=false
export CRYOSPARC_DB_ENABLE_AUTH=true
export CRYOSPARC_CLUSTER_JOB_MONITOR_INTERVAL=10
export CRYOSPARC_CLUSTER_JOB_MONITOR_MAX_RETRIES=1000000
export CRYOSPARC_PROJECT_DIR_PREFIX='CS-'
export CRYOSPARC_DEVELOP=false
export CRYOSPARC_CLICK_WRAP=true

cryosparcuser@emserver1:~$

For systems that do not support this, navigate to the cryosparc_master installation directory and run ./bin/cryosparcm instead of cryosparcm

Instance Setup

`cryosparcm update`

This command can only be run by the UNIX user that owns the CryoSPARC installation directory.

`cryosparcm patch`

Apply the latest patches available for your installed version of CryoSPARC without doing a full version update.

Specify the --help flag to see full usage

$ cryosparcm patch --help
usage: cryosparcm patch [-h] [-f] [-y] [--check] [--download] [--install]

Download and apply cryoSPARC patches. Run cryosparcm patch to automatically
install the latest patches on master and worker nodes

optional arguments:
  -h, --help   show this help message and exit
  -f, --force  install latest patch again even if already installed
  -y, --yes    confirm patch installation without prompt
  --check      check to see if a patch is available
  --download   download master and worker patches for manual installation
  --install    manually install a downloaded patch file

Frequently used commands:

cryosparcm patch: Automatically installs the latest patches on workstations or master node and connected workers. Not recommended for clusters: Use the --download and --install flags instead.
cryosparcm patch --force: Reinstalls the latest patches in case something went wrong with a previous attempt
cryosparcm patch --check: Shows information about the latest patches without installing
cryosparcm patch --download: Downloads the latest patches without installing them. Follow the resulting instructions to install the master and worker patches
cryosparcm patch --install: Run this command immediately after a --download to install the patch on the master node.

This command can only be run by the UNIX user that owns the CryoSPARC installation directory.

`cryosparcm cluster`

Use with one of the following sub-commands.

`cryosparcm cluster example <cluster_type>`

Writes config and script template files to current working directory, cluster_info.json and cluster_script.sh respectively.

`cryosparcm cluster dump <name>`

Outputs the existing config and script to current working directory for the cluster with the given name.

`cryosparcm cluster connect`

Reads cluster_info.json and cluster_script.sh from the current directory. Connects a new or updates an existing cluster configuration using the name from cluster_info.json.

`cryosparcm cluster remove <name>`

Removes a cluster configuration from the scheduler.

`cryosparcm test`

Verifies the instance has been correctly installed by running several tests. Provides a report upon completion. For more information, see Guide: Installation Testing with cryosparcm test

Specify the --help flag to see full usage

cryosparcm test install

Tests the core components of CryoSPARC (HTTP connections, licensing, workers, etc.) that are required to start running jobs and provides information on the status of the CryoSPARC instance (e.g., which version is running, whether a patch is available, etc.).

cryosparcm test workers <project_uid>

Tests workers connected to CryoSPARC to ensure they can correctly run CryoSPARC jobs by testing if the worker can launch jobs, cache particles to an SSD (if an SSD is configured), and utilize the GPU correctly.

Instance Status and Management

The following commands are only allowed to be executed by <cryosparcuser> (the user that installed CryoSPARC), and can only be executed on the master node. If these conditions are not met, you may see the following error:

You can mitigate errors temporarily by specifying the CRYOSPARC_FORCE_HOSTNAME variable just before calling the command:

cryosparcuser@wrong.server:~$ CRYOSPARC_FORCE_HOSTNAME="true"
cryosparcuser@wrong.server:~$ cryosparcm status
----------------------------------------------------------------------------
CryoSPARC System master node installed at
/home/cryosparcuser/cryosparc/cryosparc_master
----------------------------------------------------------------------------

cryosparcm process status:
...

If this error message is incorrect (the hostname specified in the error message is actually the same host, just a different identifier), you can set the hostname that the management script will use to compare by adding the variable CRYOSPARC_HOSTNAME_CHECK to cryosparc_master/config.sh. You can also set CRYOSPARC_FORCE_HOSTNAME in this file to permanently suppress this error.

File: /home/cryosparcuser/cryosparc/cryosparc_master/config.sh                          

export CRYOSPARC_LICENSE_ID="xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
export CRYOSPARC_MASTER_HOSTNAME="uoft"
export CRYOSPARC_DB_PATH="/home/cryosparcuser/cryosparc/cryosparc_master"
export CRYOSPARC_BASE_PORT=61000
export CRYOSPARC_HOSTNAME_CHECK="uoft"

Available CryoSPARC Services

app
app_api
app_legacy
command_core
command_rtp
command_vis
database

`cryosparcm status`

Prints out current status of the CryoSPARC master system, including the status of all individual processes (database, app, command_core, etc). Also prints out configuration environment variables.

This command can only be run by the UNIX user that owns the CryoSPARC installation directory.

`cryosparcm start [<service name>]`

Starts the CryoSPARC instance if stopped, including the database, the command server, the web interface, etc.

All processes start in the background, including all jobs and the web interface; the command line may be closed after running start.

In v4.0.0, app_legacy (the original web application, formerly "webapp") is not started by default, but can be at any time.

This command can only be run by the UNIX user that owns the CryoSPARC installation directory.

`cryosparcm stop [<service name>]`

Stops the CryoSPARC instance if running. This will gracefully kill all the master processes, and will cause any running jobs (potentially on other nodes) to fail.

This command can only be run by the UNIX user that owns the CryoSPARC installation directory.

`cryosparcm restart [<service name>]`

Equivalent to running cryosparcm stop && cryosparcm start

Stops the CryoSPARC instance if running. This will gracefully kill all the master processes, and will cause any running jobs (potentially on other nodes) to fail.

`cryosparcm jobstatus`

Show a summary of how many jobs are active.

`cryosparcm backup`

Backs up the CryoSPARC MongoDB database using mongodump. By default, the function creates a folder named backup inside the database path specified by the CRYOSPARC_DB_PATH environment variable in cryosparc_master/config.sh (e.g., /u/cryosparcuser/cryosparc_db/backup) and saves the backup as an .archive file with the current date and time in its path (e.g., cryosparc_backup_2021_06_14_11h27.archive) inside this folder.

Do not allow the backup to fill up the filesystem on which the database is stored. If needed, specify a custom alternative path where the backup will be written.

To change the directory the backup will be written to, specify the --dir flag. To change the name of the backup file, specify the --file flag.

Moreover, and perhaps more importantly, once CryoSPARC projects or jobs are created, deleted, or otherwise modified during or after the backup, a database restored from the resulting backup file will no longer be compatible with the modified project directories.

cryosparcm backup \
    --dir=/path/to/backups \
    --file=custombackupfile

`cryosparcm restore`

A database backup becomes outdated and incompatible with project directories as soon as CryoSPARC projects or jobs are created, deleted or modified following the the database backup. Do not restore an outdated database backup. Restoration of an outdated database backup and subsequent use with CryoSPARC is likely to corrupt CryoSPARC projects.

CryoSPARC must be turned off before running this command.

The database directory (CRYOSPARC_DB_PATH found in cryosparc_master/config.sh) must exist and be empty before running this command.

cryosparcm restore --file=/path/to/backups/backupfile

This command can only be run by the UNIX user that owns the CryoSPARC installation directory.

`cryosparcm changeport <port>`

Change CryoSPARC's base port to something else. Use the --yes or -y flag to proceed without confirmation.

cryosparcm changeport 40000

This command can only be run by the UNIX user that owns the CryoSPARC installation directory.

`cryosparcm checkdb`

Ensure the database is running with the correct host configuration

`cryosparcm fixdbport`

Run this command after manually changing CryoSPARC's base port number to ensure the Mongo database registers the change.

`cryosparcm licensestatus`

Check configured CryoSPARC license ID to see if there are any issues with verifying it.

`cryosparcm maintenancemode [on|off|status]`

Stops queued jobs from running while allowing running jobs to finish. Can be used to facilitate a better user experience while CryoSPARC is undergoing maintenance, for example during restart, patch, or update. For more information, see Guide: Maintenance Mode and Configurable User Facing Messages

`cryosparcm test`

Run cryosparcm test --help for usage information.

Test all components of a CryoSPARC instance to confirm it is working properly. For more information, see Guide: Installation Testing with cryosparcm test

Logs

The CryoSPARC system maintains several log files for its various processes that help with debugging any components of the CryoSPARC system that are not working.

`cryosparcm log <process>`

Tails the output log of the master node process denoted by <process> which can be one of the following:

command_core
command_rtp
command_vis
database
app
app_legacy
app_api

For example

cryosparcm log command_core

The log is live and automatically updates while the command-line remains open and new data is added to the log. To stop live updates and return to the shell, press control C on your keyboard, and then q.

To save the full log, redirect the output to a file

cryosparcm log command_core > command_core.log

To show only the last $x$ lines of the log, use tail. For example, to see the last 1000 lines of the log:

cryosparcm log command_core | tail -n 1000

`cryosparcm filterlog <process>`

Shows the output log of the master node process denoted by <process> which can be one of the following:

command_core
command_rtp
command_vis
database
app
app_legacy
app_api

For example

cryosparcm filterlog command_core

The log is live and automatically updates while the command-line remains open and new data is added to the log. To stop live updates and return to the shell, press control C on your keyboard.

Additional Arguments

To see cryosparcm filterlog usage, run cryosparcm filterlog -h:

Usage:
    cryosparcm filterlog SERVICE
    cryosparcm filterlog [--days|-d N] [--date|-D YYYY-MM-DD] [--max_lines|-m MAX_LINES] [--name|-n NAME] [--func|-f FUNCTION] [--level|-l LEVEL] [--tail|-t] SERVICE
Where SERVICE is one of:
    app
    app_legacy
    app_api
    command_core
    command_rtp
    command_vis
    database
Some flags not available for all services.

Please note that only command_core, command_rtp and command_vis support the following arguments with the exception of database, which also supports the date filter.

--days|-d N

Show last N days worth of logs

--date|-D YYYY-MM-DD

Show only logs from the given date

--max_lines|-m MAX_LINES

the maximum number of lines to return

--name|-n NAME

Show only logs with the given log name as a prefix (before dots). For example: cryosparcm filterlog command_core -n COMMAND.SCHEDULER

--func|-f FUNCTION

Show only logs from the given log function. For example cryosparcm filterlog command_core -f get_gpu_info_run

--level|-l LEVEL

Show only logs with the given level. For example: cryosparcm filterlog command_core -l ERROR

--tail|-t

Will tail the log with the filters applied. To stop live updates and return to the shell, press control C.

`cryosparcm joblog PX JXX`

Shows a live output log for job JXX in project PX. Includes the standard input and error from the python process for the job.

For example, to show the output of Job 123 in Project 3, run the following

cryosparcm joblog P3 J123

joblog shows the full stdout for the job, which is more comprehensive than the job log in the web interface and is more helpful for debugging.

`cryosparcm eventlog PX JXX`

Prints the text of the processing log for job JXX in project PX to stdout.

For example, to write text components of the processing log for CryoSPARC job J123 in project P3 to a file in the current working directory, run:

cryosparcm eventlog P3 J123 > P3_J123_events.log

`cryosparcm snaplogs`

Compresses all .log files in the cryosparc_master/run folder into a .tgz file inside cryosparc_master/run.

`cryosparcm errorreport`

Run cryosparcm errorreport --help for usage information.

Create a CryoSPARC error report which includes diagnostic information and CryoSPARC instance logs. For more information, see Guide: Download Error Reports

User Management

`cryosparcm listusers`

Outputs a table of users registered with the system, including their names, email addresses and admin status.

`cryosparcm createuser`

Creates a new user account for log in through the web interface. Full use:

cryosparcm createuser \
    --email "<email address>" \
    --username "<login username>" \
    --firstname "<given name>"\
    --lastname "<surname>" \
    [--password "<new password>"]

If the--password argument is not specified, a silent input prompt is provided.

This command can only be run by the UNIX user that owns the CryoSPARC installation directory.

`cryosparcm resetpassword`

Resets the password for indicated user with the new <password> provided. Full use:

cryosparcm resetpassword \
    --email "<email address>" \
    [--password "<new password>"]

If the --password argument is not specified, a silent input prompt is provided.

This command can only be run by the UNIX user that owns the CryoSPARC installation directory.

`cryosparcm updateuser`

Updates a user profile to change the user name or set admin privileges. Full use:

cryosparcm updateuser \
  --email "<email associated with cryoSPARC account>" \
  --username "<new username for that user>" \
  --firstname "<new first name(s) for that user>" \
  --lastname "<new last name(s) for that user>" \
  --admin "<true|false>" \
  [--password "<corresponding cryoSPARC password>"]

If the --password argument is not specified, a silent input prompt is provided.

--email is required and can not be changed with this command.

--username, --firstname, --lastname are used for display and do not otherwise affect CryoSPARC function.

Other than for the first-created user account, new users do not have administrative privileges applied by default. Following creation of the first-created user account, other accounts can also be created through the user interface if preferred:

This command can only be run by the UNIX user that owns the CryoSPARC installation directory.

Command-Line Interface

`cryosparcm env`

Prints a list of environment variables for use in a command-line shell to replicate the exact environment used to when running CryoSPARC processes.

Run this command with eval to define the variables output by the env command.

eval $(cryosparcm env)

`cryosparcm cli`

Runs a command using with cryoSPARC's command-line interface

cryosparcm cli <command>

See the full command reference

`cryosparcm icli`

Runs an an interactive cryosparc shell that connects to the master processes. Use this to interactively run Python commands in the same environment as CryoSPARC.