We're starting to run pioneer tests in parallel, and we get errors from update-notifier in about 50% of runs. I'm guessing that it's doing something globally.
Also, in our CI process, we don't want messages about update notification getting into the output stream when we're piping output for the build server to read. I'm not sure if update-notifier writes on stdout, so this might not be a problem, but really, we don't care about update notifications in our CI process.
To solve these two problems, we allow a --noUpdates flag from the command line to simply switch off that check.
Coverage remained the same at 89.38% when pulling a98fd843bd281b1a35a2517ca18fa5773159d041 on tomhicks:th/opt-out-update-notification into 4bda05e53397da0b0d19a7ed80cdd733815f76b0 on mojotech:master.
We're starting to run pioneer tests in parallel, and we get errors from
update-notifier
in about 50% of runs. I'm guessing that it's doing something globally.Also, in our CI process, we don't want messages about update notification getting into the output stream when we're piping output for the build server to read. I'm not sure if
update-notifier
writes onstdout
, so this might not be a problem, but really, we don't care about update notifications in our CI process.To solve these two problems, we allow a
--noUpdates
flag from the command line to simply switch off that check.