Operational Harness status
harness.io · Official status page (Statuspage-compatible JSON)
FROM ALL SOURCES
No sign of a widespread Harness problem right now; every source we read is quiet.
confidence: medium Vendor: operational · users: normal · routing: no data.
- Vendor status page Operational · 1 open incident
- User reports 0 in 15 min vs ~0 typical
- Routing / BGP no ASN mapped, so no routing view
- News & network weather no recent mentions
- Providers it may run on (inferred) no mapped dependencies
- CISA KEV (context) no recent entries for this vendor
Regions mentioned: Custom Dashboards
| Clock | Started | Running |
|---|---|---|
| Vendor clock when the vendor's open incident started | 61 days | |
| Crowd clock when user reports began rising above normal | not seen | n/a |
| Network clock when routing anomaly data first flagged it | not seen | n/a |
| ORBIT clock when our checker first recorded the change | not seen | n/a |
Evidence
- Official page: Operational (“All Systems Operational”). evidence confidence: high
- Vendor stage: postmortem on “Dashboard aggregator experiencing timeouts across prod2 since 15:04 UTC (Sept 21)”. evidence confidence: high
- User reports are within normal range (0 in 15 min). evidence confidence: low
What we don't know: what is happening inside Harness' own systems, how many users are actually affected (reports are self-selected and unverified), and anything from routing / bgp, providers it may run on (inferred). Routing data shows reachability, not whether the app works.
Automated analysis by Cloudy_Day NetWeather; not a personal statement. Generated .
Cloudy_Day's take
Stage vs. health
The vendor's stage is “postmortem”. Stage tells you where they are in their process; health is what users experience. Right now vendor: operational · users: normal · routing: no data. Vendor status and outside signals roughly agree.
Who depends on whom
We don't map other services that depend on Harness. Dependency chains are where single points of failure hide.
Identity & trust lens
Even without an identity system involved, outages push people toward workarounds: password resets, unofficial apps and “status” links in DMs. Each one is a trust decision. Use official channels.
For users
Nothing points to a broad problem. If it's broken for you, try “Is it just me?” and check your own network first.
For security teams
Treat this as an availability event unless an official source says otherwise. Check which of your systems sit on Harness, watch for outage-themed phishing aimed at your users, and log what you see with timestamps (all four clocks).
What to watch next
Any new vendor incident, a jump in user reports above baseline, or trouble at the providers Harness may depend on.
Automated analysis by Cloudy_Day NetWeather; not a personal statement. Independent researcher; not affiliated with Harness. No blame or attack claims without an official source. Generated .
INCIDENT REPLAY
Dashboard aggregator experiencing timeouts across prod2 since 15:04 UTC (Sept 21)
minor postmortem · started · resolved
5 updates
- investigating
We are currently investigating this issue.
- identified
The issue has been identified and a fix is being implemented.
- monitoring
A fix has been implemented and we are monitoring the results.
- resolved
This incident has been resolved.
- postmortem
## Summary On Sep 21, 2026, customers in Prod 2 experienced degraded and then largely unavailable deployment dashboards for approximately three hours. The issue was caused by a set of analytics queries behind those dashboards reading far more data than they needed, which exhausted the capacity of the database serving them. A rolling restart of the affected service restored dashboards fully, and the platform has been stable since. ## Customer Impact Dashboard pages loaded slowly or timed out between 14:48 and 18:11 UTC. Pipeline execution and deployment recording run on separate infrastructure and continued to operate normally. No data loss or corruption occurred. ## Root Cause The dashboard queries were missing a time-range filter on the column used to split the largest table into weekly segments, so each query read all historical data instead of only the requested period. Correctness wa
Community comments (0)
Comments are opinions of visitors, not statements by Cloudy_Day. Every comment is held for review before it appears. Links are removed, and personal data is redacted. Community guidelines.
No approved comments yet.
How to read this: official status comes from each provider's own public status page. User reports are unverified visitor reports. Things that happen at the same time are coincidence, not causation. Cloudy_Day is independent and not an official status source for any company listed.