Simple utility that pulls counts of TP entities by process, type and state and blasts them into statsd gauges for your graphing pleasure.
Set the process or processes using --process one or more times
(doing this is recommended, as story state names overlap between
processes in TP, making the monitoring less helpful otherwise).
Set the entity types you care about using --type one or more
times. The script knows how to pluralize simple forms ("Bug" ->
"Bugs") and has an override for "UserStory" -> "Userstories", but
others you may have to update the code (and PR please!).
You need to set the API base URI, username and password, either by
--api-base, --user and --password, or the TP_API_BASE,
TP_USER and TP_PASSWORD environment variables.
Statsd can be configured using environment variables as well:
STATSD_HOST and STATSD_PORT, plus STATSD_PREFIX for prefixing the
keys and STATSD_MAXUDPSIZE if you have weird network issues to work
round. (The defaults are dependent on the statsd python package, but
should be sensible.)
MIT license; source is on github; please report bugs there.
James Aylett