From 11a8ae06325e9a8c9d57e3dff5fdf46a90309cd1 Mon Sep 17 00:00:00 2001
From: kegsay <kegan@matrix.org>
Date: Fri, 27 Oct 2023 18:29:20 +0100
Subject: complement: enable dirty runs (#16520)

* complement: enable dirty runs

* Add changelog

* Set a low connpool limit when running in Complement

Dirty runs can cause many containers to be running concurrently,
which seems to easily exhaust resources on the host. The increased
speedup from dirty runs also seems to use more db connections on
workers, which are misconfigured currently to have
`SUM(workers * cp_max) > max_connections`, causing
```
FATAL:  sorry, too many clients already
```
which results in tests failing.

* Try p=2 concurrency to restrict slowness of servers which causes partial state join tests to flake

* Debug logging

* Only run flakey tests

* Only adjust connection pool limits in worker mode

* Move cp vars to somewhere where they get executed in CI

* Move cp values back to where they actually work

* Debug logging

* Try p=1 to see if this makes worker mode happier

* Remove debug logging
---
 .github/workflows/tests.yml                    | 3 ++-
 changelog.d/16520.misc                         | 1 +
 docker/complement/conf/start_for_complement.sh | 5 +++++
 docker/conf/homeserver.yaml                    | 4 ++--
 scripts-dev/complement.sh                      | 6 +++++-
 5 files changed, 15 insertions(+), 4 deletions(-)
 create mode 100644 changelog.d/16520.misc

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 13746608d4..12420911b4 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -638,9 +638,10 @@ jobs:
       - name: Prepare Complement's Prerequisites
         run: synapse/.ci/scripts/setup_complement_prerequisites.sh
 
+        # use p=1 concurrency as GHA boxes are underpowered and don't like running tons of synapses at once.
       - run: |
           set -o pipefail
-          COMPLEMENT_DIR=`pwd`/complement synapse/scripts-dev/complement.sh -json 2>&1 | synapse/.ci/scripts/gotestfmt
+          COMPLEMENT_DIR=`pwd`/complement synapse/scripts-dev/complement.sh -p 1 -json 2>&1 | synapse/.ci/scripts/gotestfmt
         shell: bash
         env:
           POSTGRES: ${{ (matrix.database == 'Postgres') && 1 || '' }}
diff --git a/changelog.d/16520.misc b/changelog.d/16520.misc
new file mode 100644
index 0000000000..ea10fd4345
--- /dev/null
+++ b/changelog.d/16520.misc
@@ -0,0 +1 @@
+Enable dirty runs on Complement CI, which is significantly faster.
diff --git a/docker/complement/conf/start_for_complement.sh b/docker/complement/conf/start_for_complement.sh
index 5560ab8b95..7b012ce8ab 100755
--- a/docker/complement/conf/start_for_complement.sh
+++ b/docker/complement/conf/start_for_complement.sh
@@ -68,6 +68,11 @@ if [[ -n "$SYNAPSE_COMPLEMENT_USE_WORKERS" ]]; then
 
   fi
   log "Workers requested: $SYNAPSE_WORKER_TYPES"
+  # adjust connection pool limits on worker mode as otherwise running lots of worker synapses
+  # can make docker unhappy (in GHA)
+  export POSTGRES_CP_MIN=1
+  export POSTGRES_CP_MAX=3
+  echo "using reduced connection pool limits for worker mode"
   # Improve startup times by using a launcher based on fork()
   export SYNAPSE_USE_EXPERIMENTAL_FORKING_LAUNCHER=1
 else
diff --git a/docker/conf/homeserver.yaml b/docker/conf/homeserver.yaml
index c46b955d63..c412ba2e87 100644
--- a/docker/conf/homeserver.yaml
+++ b/docker/conf/homeserver.yaml
@@ -67,8 +67,8 @@ database:
     host: "{{ POSTGRES_HOST or "db" }}"
     port: "{{ POSTGRES_PORT or "5432" }}"
 {% endif %}
-    cp_min: 5
-    cp_max: 10
+    cp_min: {{ POSTGRES_CP_MIN or 5 }}
+    cp_max: {{ POSTGRES_CP_MAX or 10 }}
 {% else %}
 database:
   name: "sqlite3"
diff --git a/scripts-dev/complement.sh b/scripts-dev/complement.sh
index 3e0cddb527..b1a8724b7e 100755
--- a/scripts-dev/complement.sh
+++ b/scripts-dev/complement.sh
@@ -216,6 +216,10 @@ extra_test_args=()
 
 test_packages="./tests/csapi ./tests ./tests/msc3874 ./tests/msc3890 ./tests/msc3391 ./tests/msc3930 ./tests/msc3902"
 
+# Enable dirty runs, so tests will reuse the same container where possible.
+# This significantly speeds up tests, but increases the possibility of test pollution.
+export COMPLEMENT_ENABLE_DIRTY_RUNS=1
+
 # All environment variables starting with PASS_ will be shared.
 # (The prefix is stripped off before reaching the container.)
 export COMPLEMENT_SHARE_ENV_PREFIX=PASS_
@@ -274,7 +278,7 @@ fi
 export PASS_SYNAPSE_LOG_TESTING=1
 
 # Run the tests!
-echo "Images built; running complement"
+echo "Images built; running complement with ${extra_test_args[@]} $@ $test_packages"
 cd "$COMPLEMENT_DIR"
 
 go test -v -tags "synapse_blacklist" -count=1 "${extra_test_args[@]}" "$@" $test_packages
-- 
cgit 1.4.1