Automation workflow docs: Trivial good health check

runghc.py

Submit ECAL automation jobs.

usage: runghc.py [-h] [--db DBNAME] [--notify NOTIFY]
                 [--campaign CAMPAIGN [CAMPAIGN ...]] [--logurl LOGURL]
                 [--wdir WDIR] [--eosdir EOSDIR] [--pd DATASET]
                 [--dtire DATATIRE]
                 {submit,resubmit,check} ...

Named Arguments

--db

Database name, default is test db

Default: “ecal_online_test”

--notify

Mattermost incoming webhook url for notifications

--campaign

Processing campaign(s). “all” for all campaigns in the db

--logurl

Base url for the logs

Default: “https://ecallogs.web.cern.ch/”

--wdir

Working directory

--eosdir

Base path of output location on EOS

--pd

Primary dataset

Default: “ExpressCosmics”

--dtire

Data tier

Default: “FEVT”

subcommands

Select command to execute

subcommand

Possible choices: submit, resubmit, check

Sub-commands

submit

Process all runs marked as new in the automation db

runghc.py submit [-h] [--t0 | --lfn]

Named Arguments

--t0

Read input files from T0 storage

Default: False

--lfn

Use logical file names for input files

Default: False

resubmit

Check for failed jobs and resubmit them

runghc.py resubmit [-h] [--lfn]

Named Arguments

--lfn

Use logical file names for input files

Default: False

check

Check ongoing runs and mark them as done/failed if completed

runghc.py check [-h] [--max-retries MAX_RETRIES]
                [--skipped-delay SKIPPED_DELAY]

Named Arguments

--max-retries

Max number of tries, for each single job. -1 = no limit

Default: -1

--skipped-delay

Number of days after which a task is considered as stalled and is marked as skipped.

Default: 7