Question 1

How is Valkey support different from remote DBA?

Accepted Answer

Remote DBA (/databases/valkey/remote-dba) is continuous: we monitor your fleet 24/7 and act on alerts before you see them — a monthly retainer that covers ongoing operations. Support is reactive: your team owns the operation day-to-day, and you call us when something is on fire that you can't resolve in-house. Support is per-incident or hourly-retainer; nothing recurring unless an incident hits.

Question 2

What's the 15-minute SLA actually mean?

Accepted Answer

From the moment you open a P1 ticket via our emergency channel (Slack Connect, dedicated phone line, or email-to-pager), an on-call senior Valkey engineer acknowledges within 15 minutes — 24 hours a day, seven days a week, including holidays. Acknowledgment means a real engineer is on the bridge, not an auto-reply. Triage and remediation begin immediately.

Question 3

What kinds of incidents do you actually handle?

Accepted Answer

The recurring ones: OOM events under traffic spikes, replication breakdown post network event, Sentinel quorum loss, Cluster mode split-brain after AZ outage, persistence-layer corruption, sudden p99 latency cliffs, hot-key starvation, and post-failover client storms. Less common but in-scope: VM-level kernel-tunable issues, NIC saturation, jemalloc fragmentation crises. Things we won't unilaterally do: schema rewrites, application-level changes, anything outside the Valkey/Redis stack itself.

Question 4

Do I need a contract before I can call?

Accepted Answer

No, but a pre-negotiated incident retainer significantly speeds initial response. Cold-start incident (first contact with no prior agreement): typically 60-90 minutes to onboard credentials and access before we can act. Retainer incident: we already have read-only access and the runbook for your topology, so we start triaging within the 15-minute SLA. We strongly recommend a small retainer (typically a few hours per month minimum) for any production-critical Valkey deployment.

Question 5

What does Valkey support cost?

Accepted Answer

Cold-start incident: hourly billing at our senior-engineer rate, with a typical 4-hour minimum. Hourly retainer: pre-paid bank of engineer-hours at a discount (typically 20-30% off the hourly rate), drawn down on actual usage with the SLA guaranteed. Many customers start with a small retainer (8-16 hours/quarter) and escalate from there. Pricing is transparent and shared in the contract — no per-incident surge billing.

Question 6

How does this interact with my AWS / GCP support?

Accepted Answer

Cloud-provider support handles infrastructure-level issues (VM unavailable, EBS volume failure, ElastiCache control plane bugs). We handle Valkey-engine-level issues: configuration, replication health, eviction tuning, cluster behavior, persistence problems. Both are useful and they don't overlap. When an incident crosses layers (e.g., an ElastiCache failover that didn't promote cleanly), we coordinate with their support directly via ticket cross-reference.

Valkey 24/7 Support

Incident types we handle

SLA & engagement model

Support FAQ

Production on fire?