Uptimeby GlyphPinecone Database (pod-based and serverless) SLA
Pinecone · Databases · promises up to 99.95% · SLA of Aug 29, 2024 · checked 47d ago
1 major outage in the last 365 days. Worst region eastus2: 30 min counted, 99.994% observed against the 99.95% Enterprise Advance Commitment, which applies per region.
1 of 12 months fell below it (credits are claimed per month).
Last 90 days: no major outage.
Incidents, last 365 days
[Metrics Endpoint] 5xx Errors on */metrics/discovery
A minor incident on Console, Index Management for 93 min, not counted as downtime.
6 days ago
[Pod-Based Indexes] Stalled control plane operations.A minor incident on Index Management for 3.3 h, not counted as downtime.
7 days ago
[Inference] 5xx errors on some `llama-text-embed-v2` requestsA minor incident for 0 min, not counted as downtime.
3 weeks ago
[SERVERLESS][GCP][us-central1] Newly created indexes failing to initializeA partial outage or slowdown on Index Management for 70 min, not counted as downtime.
4 weeks ago
[Pods][AWS][us-east-1] Write failures and high latencyA minor incident on us-east-1 for 14 min, not counted as downtime.
6 weeks ago
[Serverless][AWS][eu-west-1] 5xx on read requestsA partial outage or slowdown on eu-west-1 for 58 min, not counted as downtime.
6 weeks ago
[Serverless][AWS][us-west-2] 5xx errors on the readpathA partial outage or slowdown on us-west-2 for 11 h, not counted as downtime.
7 weeks ago
5xx errors on some control plane operationsA partial outage or slowdown on Console, Index Management, Assistant for 4.6 h, not counted as downtime.
2 months ago
[SERVERLESS][GCP][us-central1] 5xx errors on some read and write requestsA partial outage or slowdown on us-central1 for 21 min, not counted as downtime.
2 months ago
[SERVERLESS][Azure][eastus2] 5xx errors for some namespacesA major outage on Azure eastus2 for 30 min, counted as downtime.
3 months ago
[SERVERLESS][AWS][us-east-1] Increased latency and 5xx errors for some indexesA minor incident on us-east-1 for 75 min, not counted as downtime.
3 months ago
[Serverless][AWS][us-east-1] Increase in freshness lag for some namespacesA minor incident on us-east-1 for 4.9 h, not counted as downtime.
4 months ago
[Serverless][Azure][eastus2] 5xx errors on some requests for some indexesA partial outage or slowdown on Azure eastus2 for 4.8 h, not counted as downtime.
4 months ago
Some initial queries to infrequently read namespaces incorrectly returning empty resultsA minor incident on us-west-2, eu-west-1, us-central1 and 1 more for 8.4 h, not counted as downtime.
4 months ago
[Serverless][AWS][us-west-2] 5xx errors for some requests on some indexesA partial outage or slowdown on us-west-2 for 3.6 h, not counted as downtime.
5 months ago
[Serverless][AWS][us-west-2] Read path operations for some indexes may experience elevated latency and timeoutsA minor incident on us-west-2 for 2.8 h, not counted as downtime.
5 months ago
[AWS][us-east-1] 5xx errors for some requests on some indexesA minor incident on us-east-1 for 27 min, not counted as downtime.
5 months ago
[Serverless][AWS][us-west-2] 5xx errors and longer response times on upsert requests for some indexes.A partial outage or slowdown on us-west-2 for 79 min, not counted as downtime.
5 months ago
[AWS][us-east-1] 5xx errors on read and write operations for some indexesA partial outage or slowdown on us-east-1 for 61 min, not counted as downtime.
6 months ago
[AWS][us-east-1] 5xx errors for some requests on some indexesA partial outage or slowdown on us-east-1 for 27 min, not counted as downtime.
6 months ago
Pinecone Console loading slowly for some usersA minor incident on Console for 20 min, not counted as downtime.
6 months ago
[GCP][europe-west4] Errors for upsert operations on some indexesA partial outage or slowdown on europe-west4 for 63 min, not counted as downtime.
8 months ago
[AWS][us-east-1] Prometheus metrics are unavailable through the Pinecone console or the metric proxyA minor incident on us-east-1 for 65 min, not counted as downtime.
8 months ago
[Serverless][GCP][us-central1] 5xx errors on some read operations for some indexesA partial outage or slowdown on us-central1 for 70 min, not counted as downtime.
8 months ago
[GCP][us-central1][europe-west4] Errors on some read operationsA partial outage or slowdown on us-central1, europe-west4 for 2.6 h, not counted as downtime.
8 months ago
Latest 25 of 35.
Major outage time per month
Allowed per month: 22 min
Commitments and credits
Enterprise Advance Commitment
Database Services under an Enterprise plan with a 12+ month prepaid Advance Commitment; pod-based indexes must keep at least two replicas all month
Credit: below 99.95%: 10% · below 99%: 25% · below 95%: 50%
SLA changes
Pinecone Database (pod-based and serverless)
Support Policy: the top tier was renamed from 'Enterprise' to 'Premium'. The policy says 'No other changes were made' (support tier naming only, not uptime).
1 year agoSource
More in databases
Amazon DynamoDBAWS promises 99.999% uptime, with a 10% credit when it misses; its status page showed 3 d of major outage in the last year.
Observed 99.178% over 365 days
Amazon RDS (Multi-AZ)AWS promises 99.95% uptime, with a 10% credit when it misses; its status page showed 3 d of major outage in the last year.
Observed 99.178% over 365 days
Databricks promises 99.99% uptime, with a 5% credit when it misses; its status page has no history this site can read.
DigitalOcean promises 99.95% uptime, with a 10% credit when it misses; its status page showed no major outage in the last year.
Observed 100% over 365 days
Elastic promises 99.95% uptime, with a 10% credit when it misses; its status page showed 3.1 d of major outage in the last year.
Observed 99.157% over 365 days
Elastic promises 99.95% uptime, with a 10% credit when it misses; its status page showed 3.1 d of major outage in the last year.
Observed 99.157% over 365 days
Google Cloud promises 99.99% uptime, with a 10% credit when it misses; its status page showed 3.7 h of major outage in the 252 days on file.
Observed 99.939% over 365 days
Microsoft Azure promises 99.995% uptime, with a 10% credit when it misses; its status page has no history this site can read.
Lakebase PostgresDatabricks promises 99.99% uptime, with a 5% credit when it misses; its status page has no history this site can read.99.99%First credit5%Observed 365dNo feed
Managed DatabasesDigitalOcean promises 99.95% uptime, with a 10% credit when it misses; its status page showed no major outage in the last year.99.95%First credit10%Observed 365d100%All incidents 99.989%
Elastic Cloud Hosted (formerly Elasticsearch Service)Elastic promises 99.95% uptime, with a 10% credit when it misses; its status page showed 3.1 d of major outage in the last year.99.95%First credit10%Observed 365d99.157%All incidents 99.896%
Cloud SQLGoogle Cloud promises 99.99% uptime, with a 10% credit when it misses; its status page showed 3.7 h of major outage in the 252 days on file.99.99%First credit10%Observed 365d99.939%All incidents 99.467%
Azure SQL DatabaseMicrosoft Azure promises 99.995% uptime, with a 10% credit when it misses; its status page has no history this site can read.99.995%First credit10%Observed 365dNo feed
Terms, exclusions and sources
- Measured
- Monthly Uptime Percentage per index; a full minute is Downtime if continuous connection attempts fail (per minute-by-minute logs)
- Credit cap
- 50% of Database Monthly Fees (lowest band)
- Claim window
- Support Ticket within 60 minutes of the start of Downtime; credit request by the end of the following calendar month
- How to claim
- Separate Support Ticket with subject 'Request for SLA Credit' plus logs, affected users, dates/duration and remediation attempts
- Not covered
- Customer breach of an Order or the Agreement; Configuration or use not per Documentation; Customer Systems or Connections failures; Cloud Provider failures; Force Majeure Events; Noticed maintenance (via console or status page)
- Status history
- Status history since Aug 5, 2025
- Weighted uptime
- All incidents, weighted: 99.921% · Downtime
Pinecone will use commercially reasonable efforts to make Database Services available at least 99.95% of the time during each calendar month
The pricing page lists '99.95% Uptime SLA' only on the Enterprise plan ($500/month minimum usage). Preview Offerings are excluded. The status page's uptime figure is informational only and 'does not equate to the Monthly Uptime Percentage'.