Skip to content

Free, Open Source, Self-Hosted WhatsApp API Gateway

docs(env): say only 0 disables the reconciler intervals

Yudhi Armyndharisae6bdbfa1880
diff --git a/.env.example b/.env.example
index f9cc5d70d..7865dd43f 100644
--- a/.env.example
+++ b/.env.example
@@ -747,7 +747,7 @@ WEBHOOK_SSRF_PROTECT=true
# delivery id — a replay is idempotent against a job that did get enqueued, and re-checks the
# instance is still eligible first. Budget exhaustion marks the event failed and files a DLQ
# row, so recovery continues through the redrive endpoint instead of looping forever.
-# INGRESS_RECONCILE_INTERVAL_MS=60000 # sweep interval; <= 0 disables the reconciler
+# INGRESS_RECONCILE_INTERVAL_MS=60000 # sweep interval; 0 disables the reconciler (a negative value falls back to the default)
# INGRESS_RECONCILE_GRACE_MS=60000 # age a 'pending' row must pass before it counts as stuck
# INGRESS_RECONCILE_BATCH_SIZE=50 # stale rows replayed per sweep, oldest first
# INGRESS_RECONCILE_MAX_ATTEMPTS=5 # replays per event before it goes terminal + DLQ (min 1)
@@ -757,7 +757,7 @@ WEBHOOK_SSRF_PROTECT=true
# nothing left behind. Each delivery is now recorded before it is attempted and retired once a
# durable owner has it (handed to the queue, or completed inline), and this sweep replays whatever
# is still stranded, reusing the STORED idempotency key so a receiver can dedup the retry.
-# WEBHOOK_RECONCILE_INTERVAL_MS=60000 # sweep interval; <= 0 disables the reconciler
+# WEBHOOK_RECONCILE_INTERVAL_MS=60000 # sweep interval; 0 disables the reconciler (a negative value falls back to the default)
# WEBHOOK_RECONCILE_GRACE_MS=60000 # age a 'pending' row must pass before it counts as stranded
# WEBHOOK_RECONCILE_BATCH_SIZE=50 # stranded rows replayed per sweep, oldest first
# WEBHOOK_RECONCILE_MAX_ATTEMPTS=5 # replays before the row goes terminal (min 1)