gnt-job¶
Name¶
gnt-job - Job commands
Synopsis¶
gnt-job {command} [arguments...]
DESCRIPTION¶
The gnt-job is used for examining and manipulating the job queue.
COMMANDS¶
ARCHIVE¶
archive {job-id...}
This command can be used to archive job by their IDs. Only jobs that have finished execution (i.e either success, error or canceled jobs).
AUTOARCHIVE¶
autoarchive {age | all}
Archive jobs by their age. This command can archive jobs older than age seconds, or alternatively all finished jobs can be archived if the string all is passed.
CANCEL¶
Cancel the job(s) identified by the given job id. Only jobs that have
not yet started to run can be canceled; that is, jobs in either the
queued or waiting state. To skip a confirmation, pass --force.
--queued and waiting can be used to cancel all jobs in the
respective state, --pending includes both.
If the --kill option is given, jobs will be killed, even if in running
state, using SIGKILL in the latter case. This is dangerous, as the job will
not have the chance to do any clean up; so it will most likely leave any
objects it touched in an inconsistent state.
CHANGE-PRIORITY¶
Changes the priority of one or multiple pending jobs. Jobs currently
running have only the priority of remaining opcodes changed.
--priority must be specified. --queued and waiting can be
used to re-prioritize all jobs in the respective state, --pending
includes both. To skip a confirmation, pass --force.
INFO¶
info {job-id...}
Show detailed information about the given job id(s). If no job id is given, all jobs are examined (warning, this is a lot of information).
LIST¶
Lists the jobs and their status. By default, the job id, job status, and a small job description is listed, but additional parameters can be selected.
The --no-headers option will skip the initial header line. The
--separator option takes an argument which denotes what will be
used between the output fields. Both these options are to help
scripting.
The -o option takes a comma-separated list of output fields.
The available fields and their meaning are:
archived- Whether job is archived
end_ts- Timestamp of job end (tuple containing seconds and microseconds)
id- Job ID
opend- List of opcode execution end timestamps
opexec- List of opcode execution start timestamps (after acquiring locks)
oplog- List of opcode output logs
oppriority- List of opcode priorities
opresult- List of opcodes results
ops- List of all opcodes
opstart- List of opcode start timestamps (before acquiring locks)
opstatus- List of opcodes status
priority- Current job priority (19 to -20)
received_ts- Timestamp of when job was received (tuple containing seconds and microseconds)
start_ts- Timestamp of job start (tuple containing seconds and microseconds)
status- Job status
summary- List of per-opcode summaries
If the value of the option starts with the character +, the new
fields will be added to the default list. This allows one to quickly
see the default list plus a few other fields, instead of retyping
the entire list of fields.
To include archived jobs in the list the --archived option can be
used.
The following options can be used to show only specific jobs:
--pending- Show only jobs pending execution.
--running- Show jobs currently running only.
--error- Show failed jobs only.
--finished- Show finished jobs only.
If exactly one argument is given and it appears to be a query filter
(see ganeti(7)), the query result is filtered accordingly. For
ambiguous cases (e.g. a single field name as a filter) the --filter
(-F) option forces the argument to be treated as a filter.