workaround for file permissions issues (no read access for other)
Change-Id: Ifafa8a6db10fe7c85e192b2bd92bf4431b328ed6
Reviewed-on: https://gerrit.instructure.com/121845
Reviewed-by: Cody Cutrer <cody@instructure.com>
Tested-by: Jenkins
Product-Review: James Williams <jamesw@instructure.com>
QA-Review: James Williams <jamesw@instructure.com>
we have them in the gemfile to lock them to a specific version,
but normal behavior doesn't auto-require them
also, use 1.9 hash syntax in gemfiles (_except_ _before.rb)
Change-Id: I549c2775c65d48ff23ba1358b43713965df97813
Reviewed-on: https://gerrit.instructure.com/51636
Tested-by: Jenkins
Reviewed-by: Ethan Vizitei <evizitei@instructure.com>
Product-Review: Cody Cutrer <cody@instructure.com>
QA-Review: Cody Cutrer <cody@instructure.com>
This changed from a part of the query string in cassandra 1.1 to a
separate parameter passed to execute in cassandra 1.2 and above.
Unfortunately we need to jump through some hoops to support both, until
we've fully upgraded to 1.2.
This commit adds a placeholder to the query string %CONSISTENCY% that
will be replaced with the chosen level in 1.1, and replaced with the
empty string in 1.2. Once we've upgraded to 1.2, we can remove all this
as it'll just be another option to the method.
closes CNVS-9273
test plan:
Using each of cassandra 1.1, 1.2, and 2.0:
* Clear out the consistency level Settings
* Setting.connection.delete("DELETE FROM settings WHERE name LIKE 'event_stream.%_consistency%'")
* Verify that page views and audit logs can be fetched, verify in the
rails logs that no consistency level is given in the CQL query
lines
* Set the consistency level
* Setting.set('event_stream.read_consistency', 'ONE')
* Verify that page views and audit logs can be fetched, verify in the
rails logs that consistency level "ONE" is given in the CQL query
lines. In 1.1, this will be in the query string, in 1.2 and 2.0 it
will follow the query in an options hash
Change-Id: I3d007376d096e6ed31a40e699e77dca4cdd065a2
Reviewed-on: https://gerrit.instructure.com/35171
Tested-by: Brian Palmer <brianp@instructure.com>
Reviewed-by: Jacob Fugal <jacob@instructure.com>
QA-Review: August Thornton <august@instructure.com>
Product-Review: Brian Palmer <brianp@instructure.com>
during a previous commit
(SHA 10e7b5b003bea9aea3286b9170f5e04b8a9e3735)
we removed indentation of gems, but this indentation
had semantic significance for us: any gem which was
not required by canvas but was required by another
gem required by canvas was indented. This allowed us
to clarify that canvas didn't actually use the gems
while still locking into specific versions of each
Also added a comment explaining this to Gemfile
Change-Id: I0f476e1bed6156f2f5969e54d56d44ded5442a0f
Reviewed-on: https://gerrit.instructure.com/32588
Tested-by: Jenkins <jenkins@instructure.com>
Reviewed-by: Simon Williams <simon@instructure.com>
Product-Review: Anthus Williams <awilliams@instructure.com>
QA-Review: Anthus Williams <awilliams@instructure.com>
this is a proof-of-concept to see if partitioning
our Gemfile helps more than it hurts. It's modeled
after the way the Squash team handles its dependencies
this doesn't implement any of the particularly nice
things that can be found in the Squash set up, such
as conditionally loading gems based on the contents
of our configuration files (we can already sort of
do this with groups in bundler), but it's a start.
In particular, it allows us to add non-OSS gems to
Gemfile.d without necessarily having to release it
as part of our open-source packaging
Change-Id: If7ff1fe97409de4cd09867ad5be1c4134c5d0117
Reviewed-on: https://gerrit.instructure.com/32442
Tested-by: Jenkins <jenkins@instructure.com>
Reviewed-by: Nick Cloward <ncloward@instructure.com>
Product-Review: Anthus Williams <awilliams@instructure.com>
QA-Review: Anthus Williams <awilliams@instructure.com>