retry_fail.py
usage: retry_failed.py [-h] [--ask] [--modules MODULES] [--status STATUS]
[--fearless]
[--startrun STARTRUN | --runrange RUNRANGE RUNRANGE]
config ratv
Resets passes in a failed state to a waiting state to run again
positional arguments:
config Configuration file for database access credentials
ratv The version of rat used to select passes
optional arguments:
-h, --help show this help message and exit
--ask Prompts for each matching job instead of assuming
retry.
--modules MODULES Restrict to modules matching the (shell-style) pattern
--status STATUS Retry jobs of another status, e.g. partFail [failed]
--fearless Do not prompt the user to continue - run without fear!
--delete_attachments Remove previous attachments (.err,.out) files attached to the job documents
--startrun STARTRUN Apply to all runs after this run
--runrange RUNRANGE RUNRANGE
Restrict to a specific run range
This script may be used to rerun jobs that failed due to an intermittent issue and would now be expected to finish successfully. No new passes are created, and existing dependencies on the failed jobs will start once the retried jobs finish. This script can also operate on jobs that are not marked “failed”. For example “partFail” or “unsatisfied” jobs may be retried by passing those to the –status option. This may even be used on “running” or “submitted” jobs if the underlying site disappears, but be aware that this may result in the job running at more than one site if the site comes back online. The fearless option will not prompt a user before executing, only use if you know exactly what you are doing! The delete_attachments flag can be enable in order to remove old output and error files from previous submission trials.