Cloud Providers Health Report - February 2023

Check our February 2023 health report on the top most popular cloud providers. We analyze the health of the cloud providers based on the number of outages and problems during the month. The source of the data is made available by the cloud providers themselves via their status page. We normalize it and use it to generate the report.

Please analyze the information with a grain of salt as each provider has their own framework for reporting incidents, and in no way one provider having more outages, means that they are worse than any other. Also we are comparing providers with very different sizes and market share.

In this report, we compare AWS, Azure, DigitalOcean, Fly.io, Heroku, Linode, Netlify, Vercel according to their reported outages.

AWS

Metrics

Total Incidents: 6
0 MAJOR 6 MINOR
Total Outage Time: ~15.2 hours
Average Outage Time Per Incident: 130 minutes
Weekday with most problems: Thursday with 3 outages

List of Outages

02/02MINORElevated CloudFront latencies and error rates for IPv6 traffic
02/02MINORIncreased error rates when accessing Cost Explorer, Budgets, and Cost and Usage Reports
07/02MINORIncreased Publish API Latencies and Errors
10/02MINORConnectivity issues affecting some instances
23/02MINORNetwork Connectivity
26/02MINORIncreased Error Rates for AWS Snowball

Azure

Metrics

Total Incidents: 3
0 MAJOR 3 MINOR
Total Outage Time: ~36.4 hours
Average Outage Time Per Incident: 364 minutes
Weekday with most problems: Wednesday with 3 outages

List of Outages

02/02MINORAzure Synapse Analytics - Errors when attempting to create clusters - Investigating
07/02MINORDatacenter Cooling Event - Southeast Asia
24/02MINORSign-In Issues - Conditional Access Policies Based on GeoIP Location

DigitalOcean

Metrics

Total Incidents: 7
0 MAJOR 7 MINOR
Total Outage Time: ~39.9 hours
Average Outage Time Per Incident: 342 minutes
Weekday with most problems: Thursday and Monday with 3 outages

List of Outages

02/02MINORLON1 Networking
03/02MINOREvent Processing in FRA1
06/02MINOREvents Failures in BLR1
06/02MINORSpaces Availability in FRA1
09/02MINORApp Platform CDN
13/02MINORLoad Balancer creation in NYC1
16/02MINORApp Platform and Container Registry

Fly.io

Metrics

Total Incidents: 11
0 MAJOR 11 MINOR
Total Outage Time: ~63.8 hours
Average Outage Time Per Incident: 348 minutes
Weekday with most problems: Saturday with 3 outages

List of Outages

03/02MINOREdge routing issue
04/02MINORRelease command issue
04/02MINORUnable the schedule remote builder causing errors
07/02MINORFlycast connectivity broken from certain (new) hosts
09/02MINORDelayed state causing proxying / deployment issues
11/02MINORCapacity needed in YUL
15/02MINORNetwork connectivity issues in SCL region.
16/02MINORDNS Resolution for fly.dev occasionally failing
24/02MINORMachine API Unavailable on some hosts/regions.
26/02MINORCapacity Issues in AMS region
27/02MINORStatics are broken

Heroku

Metrics

Total Incidents: 2
2 MAJOR 0 MINOR
Total Outage Time: ~5.5 hours
Average Outage Time Per Incident: 164 minutes
Weekday with most problems: Wednesday and Saturday with 1 outage

List of Outages

Linode

Metrics

Total Incidents: 4
1 MAJOR 3 MINOR
Total Outage Time: ~34.5 hours
Average Outage Time Per Incident: 517 minutes
Weekday with most problems: Thursday and Sunday and Tuesday and Friday with 1 outage