Pub/Sub is the global messaging fabric between Google Cloud services - the glue in event-driven architectures, streaming pipelines, and microservice fan-out. Pub/Sub incidents rarely lose messages (durability is its core promise); they show up as delivery latency, subscriber backlog growth, or publish errors in one region.
Message delivery delays - backlogs grow while publishes still succeed
Elevated publish error rates or latency in one region
Push subscriptions failing while pull subscribers keep working (or vice versa)
Downstream pipeline lag: Dataflow and Cloud Run jobs starving because delivery is delayed
Pub/Sub questions
Is Pub/Sub down right now?+
The live status above reflects our latest check. Pub/Sub degradation usually means late, not lost - watch subscription backlog (oldest unacked message age) rather than publish success to see real impact.
Will I lose messages during a Pub/Sub incident?+
Acknowledged publishes are stored durably and replicated; historical incidents have overwhelmingly been delivery delays, not data loss. Design subscribers to be idempotent so the post-incident redelivery burst is safe.
Does Pub/Sub have an SLA?+
Yes - 99.95% monthly uptime. Tiered credits apply below target, must be requested within 30 days, and cap at 50% of the month’s Pub/Sub spend.
Get alerted when Pub/Sub breaks
Instant email the moment we detect an incident - affected services and regions included.
gcp