From 52fc380a9493e05dca1ded7c91f6b4cc52feec01 Mon Sep 17 00:00:00 2001
From: George Gritsouk <989898+gggritso@users.noreply.github.com>
Date: Wed, 6 Nov 2024 15:01:18 -0500
Subject: [PATCH 1/3] Add N sampling

---
 src/sentry/tagstore/snuba/backend.py | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/src/sentry/tagstore/snuba/backend.py b/src/sentry/tagstore/snuba/backend.py
index bf9922b42b5b94..ef35ed2915ca1a 100644
--- a/src/sentry/tagstore/snuba/backend.py
+++ b/src/sentry/tagstore/snuba/backend.py
@@ -11,10 +11,11 @@
 from sentry_relay.consts import SPAN_STATUS_CODE_TO_NAME
 from snuba_sdk import Column, Condition, Direction, Entity, Function, Op, OrderBy, Query, Request
 
-from sentry import analytics
+from sentry import analytics, features
 from sentry.api.utils import default_start_end_dates
 from sentry.issues.grouptype import GroupCategory
 from sentry.models.group import Group
+from sentry.models.organization import Organization
 from sentry.models.project import Project
 from sentry.models.release import Release
 from sentry.models.releaseenvironment import ReleaseEnvironment
@@ -433,6 +434,15 @@ def get_tag_keys_for_projects(
         max_unsampled_projects = _max_unsampled_projects
         # We want to disable FINAL in the snuba query to reduce load.
         optimize_kwargs = {"turbo": True}
+
+        # Add static sample amount to the query. Turbo will sample at 10% by
+        # default, but organizations with many events still get timeouts. A
+        # static sample creates more consistent performance.
+        organization_id = get_organization_id_from_project_ids(projects)
+        organization = Organization.objects.get_from_cache(id=organization_id)
+        if features.has("organizations:tag-key-sample-n", organization):
+            optimize_kwargs["sample"] = 1_000_000
+
         # If we are fetching less than max_unsampled_projects, then disable
         # the sampling that turbo enables so that we get more accurate results.
         # We only want sampling when we have a large number of projects, so

From 38e3e6d9ef6f09c322a233945b06ff4a7de43ef9 Mon Sep 17 00:00:00 2001
From: George Gritsouk <989898+gggritso@users.noreply.github.com>
Date: Thu, 7 Nov 2024 13:28:41 -0500
Subject: [PATCH 2/3] Register option for N value

---
 src/sentry/options/defaults.py       | 6 ++++++
 src/sentry/tagstore/snuba/backend.py | 4 ++--
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/src/sentry/options/defaults.py b/src/sentry/options/defaults.py
index 25b6c81b80c5a3..6eac8160297367 100644
--- a/src/sentry/options/defaults.py
+++ b/src/sentry/options/defaults.py
@@ -2848,3 +2848,9 @@
     default=[],
     flags=FLAG_AUTOMATOR_MODIFIABLE,
 )
+
+register(
+    "visibility.tag-key-sample-size",
+    default=1_000_000,
+    flags=FLAG_AUTOMATOR_MODIFIABLE,
+)
diff --git a/src/sentry/tagstore/snuba/backend.py b/src/sentry/tagstore/snuba/backend.py
index ef35ed2915ca1a..0f13ae3272a148 100644
--- a/src/sentry/tagstore/snuba/backend.py
+++ b/src/sentry/tagstore/snuba/backend.py
@@ -11,7 +11,7 @@
 from sentry_relay.consts import SPAN_STATUS_CODE_TO_NAME
 from snuba_sdk import Column, Condition, Direction, Entity, Function, Op, OrderBy, Query, Request
 
-from sentry import analytics, features
+from sentry import analytics, features, options
 from sentry.api.utils import default_start_end_dates
 from sentry.issues.grouptype import GroupCategory
 from sentry.models.group import Group
@@ -441,7 +441,7 @@ def get_tag_keys_for_projects(
         organization_id = get_organization_id_from_project_ids(projects)
         organization = Organization.objects.get_from_cache(id=organization_id)
         if features.has("organizations:tag-key-sample-n", organization):
-            optimize_kwargs["sample"] = 1_000_000
+            optimize_kwargs["sample"] = options.get("visibility.tag-key-sample-size")
 
         # If we are fetching less than max_unsampled_projects, then disable
         # the sampling that turbo enables so that we get more accurate results.

From 12f13f04d83c6f2f3b51c29f88d93f0313d7f378 Mon Sep 17 00:00:00 2001
From: George Gritsouk <989898+gggritso@users.noreply.github.com>
Date: Thu, 7 Nov 2024 13:54:18 -0500
Subject: [PATCH 3/3] Add comment

---
 src/sentry/options/defaults.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/sentry/options/defaults.py b/src/sentry/options/defaults.py
index dfdf542ce538e3..36e64a9a208f4f 100644
--- a/src/sentry/options/defaults.py
+++ b/src/sentry/options/defaults.py
@@ -2856,6 +2856,7 @@
     flags=FLAG_AUTOMATOR_MODIFIABLE,
 )
 
+# option for sample size when fetching project tag keys
 register(
     "visibility.tag-key-sample-size",
     default=1_000_000,