Bigtable: systests leaking instances #7003
Labels
api: bigtable
Issues related to the Bigtable API.
flaky
testing
type: process
A process-related concern. May include testing, release, or the like.
Similar to #4935. We have a bunch of leaked instances, and are now seeing systest failures due to quota exhaustion. With the labels, it is possible to see that several different "kinds" of instances are leaked from systests:
dif-####
(leaked fromtest_create_instance_w_two_clusters
).g-c-p-###
w/ display name "Foo Bar Baz" (the default instance intests/system.py
, mutated bytest_update_display_name_and_labels
and not restored)g-c-p-###
w/o display name (the default instance intests/system.py
)new-###
(leaked fromtest_create_instance
systest)and from snippets:
inst-my-###
(leaked one of two snippets:test_bigtable_create_instance
ortest_bigtable_delete_instance
)snippet-###
w/ display name "My new instance" (default instance for snippets; display name set by 'test_bigtable_update_instance`)snippet-###
w/o display name (default instance for snippets)I
will deletehave deleted all but the most recent example of each kind.The text was updated successfully, but these errors were encountered: