Revert "Patch Prometheus services records"
What does this MR do?
Reverts !19956 (merged) the post-migration failed on staging https://ops.gitlab.net/gitlab-com/gl-infra/deployer/-/jobs/839358
TASK [Run migrations] **********************************************************
108 fatal: [deploy-01-sv-gstg.c.gitlab-staging-1.internal]: FAILED! => changed=true
109 cmd:
110 - /usr/bin/gitlab-rake
111 - db:migrate
112 delta: '0:02:04.549186'
113 end: '2020-01-13 21:58:11.302126'
114 msg: non-zero return code
115 rc: 1
116 start: '2020-01-13 21:56:06.752940'
117 stderr: |-
118 rake aborted!
119 StandardError: An error has occurred, all later migrations canceled:
120
121 PG::QueryCanceled: ERROR: canceling statement due to statement timeout
122 : SELECT "projects"."id" FROM "projects" LEFT JOIN services ON services.project_id = projects.id AND services.type = 'PrometheusService' WHERE (services.id IS NULL OR (services.active = FALSE AND services.properties = '{}')) AND "projects"."id" >= 154014 GROUP BY projects.id ORDER BY "projects"."id" ASC LIMIT 1 OFFSET 500
123 /opt/gitlab/embedded/service/gitlab-rails/app/models/concerns/each_batch.rb:78:in `block in each_batch'
124 /opt/gitlab/embedded/service/gitlab-rails/app/models/concerns/each_batch.rb:68:in `step'
125 /opt/gitlab/embedded/service/gitlab-rails/app/models/concerns/each_batch.rb:68:in `each_batch'
126 /opt/gitlab/embedded/service/gitlab-rails/db/post_migrate/20191220102807_patch_prometheus_services_for_shared_cluster_applications.rb:60:in `up'
127 /opt/gitlab/embedded/bin/bundle:23:in `load'
128 /opt/gitlab/embedded/bin/bundle:23:in `<main>'
129
130 Caused by:
131 ActiveRecord::QueryCanceled: PG::QueryCanceled: ERROR: canceling statement due to statement timeout
132 : SELECT "projects"."id" FROM "projects" LEFT JOIN services ON services.project_id = projects.id AND services.type = 'PrometheusService' WHERE (services.id IS NULL OR (services.active = FALSE AND services.properties = '{}')) AND "projects"."id" >= 154014 GROUP BY projects.id ORDER BY "projects"."id" ASC LIMIT 1 OFFSET 500
133 /opt/gitlab/embedded/service/gitlab-rails/app/models/concerns/each_batch.rb:78:in `block in each_batch'
134 /opt/gitlab/embedded/service/gitlab-rails/app/models/concerns/each_batch.rb:68:in `step'
135 /opt/gitlab/embedded/service/gitlab-rails/app/models/concerns/each_batch.rb:68:in `each_batch'
136 /opt/gitlab/embedded/service/gitlab-rails/db/post_migrate/20191220102807_patch_prometheus_services_for_shared_cluster_applications.rb:60:in `up'
137 /opt/gitlab/embedded/bin/bundle:23:in `load'
138 /opt/gitlab/embedded/bin/bundle:23:in `<main>'
139
140 Caused by:
141 PG::QueryCanceled: ERROR: canceling statement due to statement timeout
142 /opt/gitlab/embedded/service/gitlab-rails/app/models/concerns/each_batch.rb:78:in `block in each_batch'
143 /opt/gitlab/embedded/service/gitlab-rails/app/models/concerns/each_batch.rb:68:in `step'
144 /opt/gitlab/embedded/service/gitlab-rails/app/models/concerns/each_batch.rb:68:in `each_batch'
145 /opt/gitlab/embedded/service/gitlab-rails/db/post_migrate/20191220102807_patch_prometheus_services_for_shared_cluster_applications.rb:60:in `up'
146 /opt/gitlab/embedded/bin/bundle:23:in `load'
147 /opt/gitlab/embedded/bin/bundle:23:in `<main>'
148 Tasks: TOP => db:migrate
149 (See full trace by running task with --trace)
150 stderr_lines: <omitted>
151 stdout: |-
152 == 20191128162854 DropProjectCiCdSettingsMergeTrainsEnabled: migrating ========
153 -- remove_column(:project_ci_cd_settings, :merge_trains_enabled)
154 -> 0.0076s
155 == 20191128162854 DropProjectCiCdSettingsMergeTrainsEnabled: migrated (0.0077s)
156
157 == 20191220102807 PatchPrometheusServicesForSharedClusterApplications: migrating
158 stdout_lines: <omitted>
159 NO MORE HOSTS LEFT *************************************************************
160 PLAY RECAP *********************************************************************
161 deploy-01-sv-gstg.c.gitlab-staging-1.internal : ok=6 changed=0 unreachable=0 failed=1 skipped=1 rescued=0 ignored=0 Post-migration passed on the second retry (https://ops.gitlab.net/gitlab-com/gl-infra/deployer/-/jobs/839428), but it's worrisome that failed inside of an environment that has barely traffic, so I'm reverting to err on the safe side.
Failure can be analyzed and if it's considered a "one-time-thing" we can merge it again.
Does this MR meet the acceptance criteria?
Conformity
Edited by Mayra Cabrera