chore(scaletest/templates/scaletest-runner): fix dashboard command invocation, autoscale provisioners (#10177)

johnstcn · mafredri · web-flow · commit 2506415defd7 · 2023-10-10T15:33:55.000+01:00
add --retries on kubectl cp
remove --count parameter to scaletest dashboard
scale provisioners up and down

Co-authored-by: Mathias Fredriksson &lt;mafredri@gmail.com&gt;
diff --git a/scaletest/templates/scaletest-runner/scripts/cleanup.sh b/scaletest/templates/scaletest-runner/scripts/cleanup.sh
@@ -28,6 +28,12 @@ coder exp scaletest cleanup \
 	tee "${SCALETEST_RESULTS_DIR}/cleanup-${event}.txt"
 end_phase
 
+if [[ $event != prepare ]]; then
+	start_phase "Scaling down provisioners..."
+	maybedryrun "$DRY_RUN" kubectl scale deployment/coder-provisioner --replicas 1
+	maybedryrun "$DRY_RUN" kubectl rollout status deployment/coder-provisioner
+fi
+
 if [[ $event = manual ]]; then
 	echo 'Press any key to continue...'
 	read -s -r -n 1
diff --git a/scaletest/templates/scaletest-runner/scripts/lib.sh b/scaletest/templates/scaletest-runner/scripts/lib.sh
@@ -271,10 +271,12 @@ fetch_coder_full() {
 		exit 1
 	fi
 	log "Fetching full Coder binary from ${pod}"
+	# We need --retries due to https://github.com/kubernetes/kubernetes/issues/60140 :(
 	maybedryrun "${DRY_RUN}" kubectl \
 		--namespace "${namespace}" \
 		cp \
 		--container coder \
+		--retries 10 \
 		"${pod}:/opt/coder" "${SCALETEST_CODER_BINARY}"
 	maybedryrun "${DRY_RUN}" chmod +x "${SCALETEST_CODER_BINARY}"
 	log "Full Coder binary downloaded to ${SCALETEST_CODER_BINARY}"
diff --git a/scaletest/templates/scaletest-runner/scripts/prepare.sh b/scaletest/templates/scaletest-runner/scripts/prepare.sh
@@ -51,3 +51,15 @@ log "Cleaning up from previous runs (if applicable)..."
 "${SCRIPTS_DIR}/cleanup.sh" "prepare"
 
 log "Preparation complete!"
+
+PROVISIONER_REPLICA_COUNT="${SCALETEST_PARAM_CREATE_CONCURRENCY:-0}"
+if [[ "${PROVISIONER_REPLICA_COUNT}" -eq 0 ]]; then
+	# TODO(Cian): what is a good default value here?
+	echo "Setting PROVISIONER_REPLICA_COUNT to 10 since SCALETEST_PARAM_CREATE_CONCURRENCY is 0"
+	PROVISIONER_REPLICA_COUNT=10
+fi
+log "Scaling up provisioners to ${PROVISIONER_REPLICA_COUNT}..."
+maybedryrun "$DRY_RUN" kubectl scale deployment/coder-provisioner \
+	--replicas "${PROVISIONER_REPLICA_COUNT}"
+log "Waiting for provisioners to scale up..."
+maybedryrun "$DRY_RUN" kubectl rollout status deployment/coder-provisioner
diff --git a/scaletest/templates/scaletest-runner/scripts/run.sh b/scaletest/templates/scaletest-runner/scripts/run.sh
@@ -49,7 +49,6 @@ for scenario in "${SCALETEST_PARAM_LOAD_SCENARIOS[@]}"; do
 		;;
 	"Dashboard Traffic")
 		coder exp scaletest dashboard \
-			--count "${SCALETEST_PARAM_NUM_WORKSPACES}" \
 			--timeout "${SCALETEST_PARAM_LOAD_SCENARIO_DASHBOARD_TRAFFIC_DURATION}m" \
 			--job-timeout "${SCALETEST_PARAM_LOAD_SCENARIO_DASHBOARD_TRAFFIC_DURATION}m30s" \
 			--output json:"${SCALETEST_RESULTS_DIR}/traffic-dashboard.json" \
