Is it down again?

Who went down, for how long, by their own admission.

Days without a Sev 1Days since the last Sev 1 anywhere was resolved.
17
Incidents, 90dEverything posted on the status page except scheduled maintenance.
642
Sev 1, 90dSev 1 incidents across every provider, 90 days.
12
It was DNSIncidents with DNS in the title or components, 90 days.
8
ComputePostgresRedisStorageCDN & DNSDeploys

Hosting

Last 90 days · Deploys · Tokyo

Worst firstBest first
1DSEV 2 · SECURITYOct 21 · 2d 14h · 48%Railway Platform OutageSEV 2Aug 17 · 7h 30m · 47%GitHub is experiencing elevated error ratesSEV 3Oct 26 · 12h 21m · 2%Degraded snapshot upload latency+2 smaller24 incidents left out · Deploys · Tokyo
Railway
PaaS
DSHIP
98.37%ADMITTED UPTIME
8INCIDENTS
2d 14hLONGEST
~$12COST/MO
2BSEV 2Oct 12 · 6h 36m · 65%Increased build and deployment failuresSEV 3Aug 26 · 1h 22m · 12%Failures logging in with Vercel CLISEV 3Sep 18 · 50m · 6%Elevated Errors Triggering Deployments+6 smaller9 incidents left out · Deploys · Tokyo
Vercel
Frontend cloud
BSHIP
99.95%ADMITTED UPTIME
14INCIDENTS
6h 36mLONGEST
~$20COST/MO
3ASEV 2Oct 20 · 3h 12m · 31%Cloudflare Zero Trust connectivitySEV 3Oct 13 · 2d 14h · 10%Audit Logs v2 ingestion is missing multiple Cloudflare products SEV 3Oct 26 · 51m · 9%Cloudflare Workers build delays+9 smaller54 incidents left out · Deploys · Tokyo
Cloudflare61 DAYSBEST VALUE
Edge platform
ASHIP
99.93%ADMITTED UPTIME
15INCIDENTS
3h 12mLONGEST
~$5COST/MO
4ASEV 2 · SECURITYOct 16 · 12h 53m · 43%Platform Wide Rolling Host RebootsSEV 3Oct 13 · 4h 25m · 41%Historical logs are currently delayedSEV 2Sep 23 · 7h 01m · 11%Partial Sprites outage+2 smaller23 incidents left out · Deploys · Tokyo
Fly.io55 DAYS
Edge compute
ASHIP
99.89%ADMITTED UPTIME
6INCIDENTS
12h 53mLONGEST
$5.69COST/MO
5ASEV 2Oct 20 · 3h 41m · 100%Deploying from GitLab may intermittently fail13 incidents left out · Deploys · Tokyo
Render
PaaS
ASHIP
99.97%ADMITTED UPTIME
1INCIDENTS
3h 41mLONGEST
$21.25COST/MO
6ASEV 3Aug 3 · 4h 02m · 60%Cloud Control Panel AccessSEV 3Oct 5 · 1h 35m · 30%VPC Networking AvailabilitySEV 2Aug 21 · 1h 53m · 10%Container Registry and Spaces Accessibility14 incidents left out · Deploys · Tokyo
DigitalOcean
Cloud
ASHIP
99.99%ADMITTED UPTIME
3INCIDENTS
1h 53mLONGEST
$6COST/MO
7A+SEV 3Oct 27 · 57m · 42%Elevated Build Errors and Agent Runner failuresSEV 3Aug 3 · 25m · 37%Elevated Build ErrorsSEV 3Oct 17 · 52m · 21%Elevated Build Errors12 incidents left out · Deploys · Tokyo
Netlify
Frontend cloud
A+SHIP
99.99%ADMITTED UPTIME
7INCIDENTS
34mLONGEST
~$20COST/MO
8A+Nothing counted, 90 days.5 incidents left out · Deploys · Tokyo
HerokuBEST VALUE
PaaS
A+SHIP
100.00%ADMITTED UPTIME
0INCIDENTS
–LONGEST
$7COST/MO
CleanSev 3–4Sev 1–2No data

PROD = your users hit. SHIP = you can't deploy or manage. How grades work

Bang for buck

One small production web app, per month512 MB app on all month, 1M requests, 100 GB out, commercial use, one seat. Cheapest plan that runs it.
A
CloudflareBEST VALUE
Workers Paid
~$5ESTIMATE
A
Fly.io
Pay as you go
$5.69LIST PRICE
Cloudflare: cheaper, same grade
A
DigitalOcean
App Platform, shared 512 MB
$6LIST PRICE
Cloudflare: cheaper, same grade
A+
HerokuBEST VALUE
Basic dyno
$7LIST PRICE
D
Railway
Hobby
~$12ESTIMATE
Heroku: cheaper, better grade
A+
Netlify
Pro
~$20ESTIMATE
Heroku: cheaper, same grade
B
Vercel
Pro, one seat
~$20ESTIMATE
Heroku: cheaper, better grade
A
Render
Hobby workspace, Starter instance
$21.25LIST PRICE
Heroku: cheaper, better grade

BEST VALUE = nothing cheaper with an equal or better grade. How we price

Also tracking

AI APIsALL UPGroqA · 99.99%CohereC · 99.90%OpenAI APIC · 99.27%All 4CloudALL UPAWSB · 99.79%Google CloudC · 99.74%All 3DatabasesALL UPUpstashA+ · 100.00%PlanetScaleA · 99.98%CockroachDB CloudB · 99.92%All 5Dev toolsALL UPDocker HubA · 99.99%npmC · 99.82%GitHubF · 99.31%All 3Dev APIs1 ISSUEMailerSendA+ · 100.00%MapboxA+ · 99.99%PaddleA+ · 100.00%All 16
Next upCDN & DNS.Suggest a provider