-
Notifications
You must be signed in to change notification settings - Fork 956
chore: add usage tracking package #19095
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Changes from all commits
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -3913,6 +3913,13 @@ func (q *querier) InsertTemplateVersionWorkspaceTag(ctx context.Context, arg dat | |
return q.db.InsertTemplateVersionWorkspaceTag(ctx, arg) | ||
} | ||
|
||
func (q *querier) InsertUsageEvent(ctx context.Context, arg database.InsertUsageEventParams) error { | ||
if err := q.authorizeContext(ctx, policy.ActionCreate, rbac.ResourceSystem); err != nil { | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. We should probably create a separate RBAC resource and role for these events. There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. These are currently not directly CRUDable by any users, even admins. I was going to originally add this to it's own resource, but after I realized that I decided that it could be added down the track if we add APIs for using this information There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Even so, |
||
return err | ||
} | ||
return q.db.InsertUsageEvent(ctx, arg) | ||
} | ||
|
||
func (q *querier) InsertUser(ctx context.Context, arg database.InsertUserParams) (database.User, error) { | ||
// Always check if the assigned roles can actually be assigned by this actor. | ||
impliedRoles := append([]rbac.RoleIdentifier{rbac.RoleMember()}, q.convertToDeploymentRoles(arg.RBACRoles)...) | ||
|
@@ -4260,6 +4267,14 @@ func (q *querier) RevokeDBCryptKey(ctx context.Context, activeKeyDigest string) | |
return q.db.RevokeDBCryptKey(ctx, activeKeyDigest) | ||
} | ||
|
||
func (q *querier) SelectUsageEventsForPublishing(ctx context.Context, arg time.Time) ([]database.UsageEvent, error) { | ||
// ActionUpdate because we're updating the publish_started_at column. | ||
if err := q.authorizeContext(ctx, policy.ActionUpdate, rbac.ResourceSystem); err != nil { | ||
return nil, err | ||
} | ||
return q.db.SelectUsageEventsForPublishing(ctx, arg) | ||
} | ||
|
||
func (q *querier) TryAcquireLock(ctx context.Context, id int64) (bool, error) { | ||
return q.db.TryAcquireLock(ctx, id) | ||
} | ||
|
@@ -4725,6 +4740,13 @@ func (q *querier) UpdateTemplateWorkspacesLastUsedAt(ctx context.Context, arg da | |
return fetchAndExec(q.log, q.auth, policy.ActionUpdate, fetch, q.db.UpdateTemplateWorkspacesLastUsedAt)(ctx, arg) | ||
} | ||
|
||
func (q *querier) UpdateUsageEventsPostPublish(ctx context.Context, arg database.UpdateUsageEventsPostPublishParams) error { | ||
if err := q.authorizeContext(ctx, policy.ActionUpdate, rbac.ResourceSystem); err != nil { | ||
return err | ||
} | ||
return q.db.UpdateUsageEventsPostPublish(ctx, arg) | ||
} | ||
|
||
func (q *querier) UpdateUserDeletedByID(ctx context.Context, id uuid.UUID) error { | ||
return deleteQ(q.log, q.auth, q.db.GetUserByID, q.db.UpdateUserDeletedByID)(ctx, id) | ||
} | ||
|
Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.
Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.
Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,2 @@ | ||
DROP TABLE usage_events; | ||
DROP TYPE usage_event_type; |
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Obligatory reminder to check migration number before merging! |
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,26 @@ | ||
CREATE TYPE usage_event_type AS ENUM ( | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I'd normally be in favor of an enum here, and I'm sorry I didn't think of it at RFC time, but... We are restricted from using Instead we could make the |
||
'dc_managed_agents_v1' | ||
); | ||
|
||
COMMENT ON TYPE usage_event_type IS 'The usage event type with version. "dc" means "discrete" (e.g. a single event, for counters), "hb" means "heartbeat" (e.g. a recurring event that contains a total count of usage generated from the database, for gauges).'; | ||
|
||
CREATE TABLE usage_events ( | ||
id TEXT PRIMARY KEY, | ||
event_type usage_event_type NOT NULL, | ||
event_data JSONB NOT NULL, | ||
created_at TIMESTAMP WITH TIME ZONE NOT NULL, | ||
publish_started_at TIMESTAMP WITH TIME ZONE DEFAULT NULL, | ||
published_at TIMESTAMP WITH TIME ZONE DEFAULT NULL, | ||
failure_message TEXT DEFAULT NULL | ||
); | ||
|
||
COMMENT ON TABLE usage_events IS 'usage_events contains usage data that is collected from the product and potentially shipped to the usage collector service.'; | ||
COMMENT ON COLUMN usage_events.id IS 'For "discrete" event types, this is a random UUID. For "heartbeat" event types, this is a combination of the event type and a truncated timestamp.'; | ||
COMMENT ON COLUMN usage_events.event_data IS 'Event payload. Determined by the matching usage struct for this event type.'; | ||
COMMENT ON COLUMN usage_events.publish_started_at IS 'Set to a timestamp while the event is being published by a Coder replica to the usage collector service. Used to avoid duplicate publishes by multiple replicas. Timestamps older than 1 hour are considered expired.'; | ||
COMMENT ON COLUMN usage_events.published_at IS 'Set to a timestamp when the event is successfully (or permanently unsuccessfully) published to the usage collector service. If set, the event should never be attempted to be published again.'; | ||
COMMENT ON COLUMN usage_events.failure_message IS 'Set to an error message when the event is temporarily or permanently unsuccessfully published to the usage collector service.'; | ||
|
||
CREATE INDEX idx_usage_events_created_at ON usage_events (created_at); | ||
CREATE INDEX idx_usage_events_publish_started_at ON usage_events (publish_started_at); | ||
CREATE INDEX idx_usage_events_published_at ON usage_events (published_at); | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I think you want a single index over a tuple and the order matters. Having 3 indexes is much less useful for a query that uses all 3 fields. It should be published_at first, to allow the query to ignore already published events, which will be the vast majority of the table. Next is publish_started_at, which we use to filter out in-progress events. Lastly created_at since we order by this and exclude anything older than 30 days. |
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,60 @@ | ||
INSERT INTO usage_events ( | ||
id, | ||
event_type, | ||
event_data, | ||
created_at, | ||
publish_started_at, | ||
published_at, | ||
failure_message | ||
) | ||
VALUES | ||
-- Unpublished dc_managed_agents_v1 event. | ||
( | ||
'event1', | ||
'dc_managed_agents_v1', | ||
'{"count":1}', | ||
'2023-01-01 00:00:00+00', | ||
NULL, | ||
NULL, | ||
NULL | ||
), | ||
-- Successfully published dc_managed_agents_v1 event. | ||
( | ||
'event2', | ||
'dc_managed_agents_v1', | ||
'{"count":2}', | ||
'2023-01-01 00:00:00+00', | ||
NULL, | ||
'2023-01-01 00:00:02+00', | ||
NULL | ||
), | ||
-- Publish in progress dc_managed_agents_v1 event. | ||
( | ||
'event3', | ||
'dc_managed_agents_v1', | ||
'{"count":3}', | ||
'2023-01-01 00:00:00+00', | ||
'2023-01-01 00:00:01+00', | ||
NULL, | ||
NULL | ||
), | ||
-- Temporarily failed to publish dc_managed_agents_v1 event. | ||
( | ||
'event4', | ||
'dc_managed_agents_v1', | ||
'{"count":4}', | ||
'2023-01-01 00:00:00+00', | ||
NULL, | ||
NULL, | ||
'publish failed temporarily' | ||
), | ||
-- Permanently failed to publish dc_managed_agents_v1 event. | ||
( | ||
'event5', | ||
'dc_managed_agents_v1', | ||
'{"count":5}', | ||
'2023-01-01 00:00:00+00', | ||
NULL, | ||
'2023-01-01 00:00:02+00', | ||
'publish failed permanently' | ||
) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
in general, I'd like to ensure we have 2 code owners for each thing, so that we have some ability to make progress if someone is out