lizardfs-admin(8)

NAME

lizardfs-admin, lizardfs-probe - LizardFS monitoring and administration tool

SYNOPSIS

lizardfs-admin  COMMAND [OPTIONS...] [ARGUMENTS...]

Available COMMANDs

chunks-health <master ip> <master port>

Returns chunks health reports in the installation. By default (if no report is specified) all reports will be shown. In replication and deletion states, the column means the number of chunks with number of copies specified in the label to replicate/delete.

Possible command-line options:

--availability Print report about availability of chunks.
--replication Print report about about number of chunks that need replication.
--deletion Print report about about number of chunks that need deletion.
info <master ip> <master port>
Prints statistics concerning the LizardFS installation.
iolimits-status <master ip> <master port>
Prints current configuration of global I/O limiting
list-chunkservers <master ip> <master port>
Prints information about all connected chunkservers.
list-defective-files <master ip> <master port>

Prints out paths of files which have unavailable or undergoal chunks

Possible command-line options:

--unavailable Print unavailable files
--undergoal Print files with undergoal chunks
--limit Limit maximum number of printed files
list-disks <master ip> <master port>

Prints information about all connected chunkservers.

Possible command-line options:

--verbose Be a little more verbose and show operations statistics.
list-goals <master ip> <master port>

List goal definitions.

Possible command-line options:

--pretty Print nice table
list-mounts <master ip> <master port>

Prints information about all connected mounts.

Possible command-line options:

--verbose Be a little more verbose and show goal and trash time limits.
metadataserver-status <master ip> <master port>
Prints status of a master or shadow master server
list-metadataservers <master ip> <master port>
Prints status of active metadata servers.
ready-chunkservers-count <master ip> <master port>
Prints number of chunkservers ready to be written to.
promote-shadow <shadow ip> <shadow port>
Promotes metadata server. Works only if personality ‘ha-cluster-managed’ is used. Authentication with the admin password is required.
stop-master-without-saving-metadata <master ip> <master port>
Stop the master server without saving metadata in the metadata.mfs file. Used to quickly migrate a metadata server (works for all personalities). Authentication with the admin password is required.
reload-config <master ip> <master port>
Requests reloading configuration from the config file. This is synchronous (waits for reload to finish). Authentication with the admin password is required.
save-metadata <metadataserver ip> <metadataserver port>

Requests saving the current state of metadata into the metadata.mfs file. With –async fail if the process cannot be started, e.g. because the process is already in progress. Without –async, fails if either the process cannot be started or if it finishes with an error (i.e., no metadata file is created).

Authentication with the admin password is required.

Possible command-line options:

--async Don’t wait for the task to finish.
list-tasks <master ip> <master port>
Lists tasks which are currently being executed on the master. The returned id’s can be used by the stop-task command.
stop-task <master ip> <master port> <task id>
Stop execution of the task with the given id

COMMON COMMAND OPTIONS

--porcelain Make the output parsing-friendly.

REPORTING BUGS

Report bugs to <contact@lizardfs.org>.

SEE ALSO

lizardfs(7)