- Level
- confirmed
- Worst impact
- Degraded
- AWS says
- 2026-11-06 22:35 UTC to 2026-11-07 00:21 UTC (1h 46m)
- Components
- Amazon Elastic Compute Cloud
- Regions
- us-west-2
- Cascade
- aws-2026-11-06
Summary
The provider says EC2 instance launch failures affected all AZs in US-WEST-2 from 2:35 PM to 4:21 PM PST, caused by a subsystem supporting EBS Volumes. The issue is resolved, running instances were not impacted, and retrying failed requests is recommended.
Written by anthropic/claude-haiku-5.5 from AWS's own updates.
What to do
Check whether you use AWS Amazon Elastic Compute Cloud in us-west-2. Check your stack
Paste a package.json, lockfile, Terraform or a list of domains at down.guide/stack to see every provider you depend on.Confirm your retries, queues and scheduled jobs caught up after the 2h 21m it lasted.
Anything that failed between 2026-11-06 22:51 UTC and 2026-11-07 01:12 UTC may need a replay.Subscribe to AWS's feed to hear about the next one first. RSS feed
https://down.guide/feed?provider=aws works in Slack (/feed subscribe), Discord feed bots and any RSS reader.
Updates from AWS 4
- investigating
We are experiencing elevated launch failures in the US-WEST-2 Region.
provider
- investigating
We are experiencing elevated launch failures in the US-WEST-2 Region caused by an issue with the EBS service. Services that depend on EBS or EC2 will experience failure rates of their own. We are working to mitigate the issues. We will provide you with another update by 4:00 PM PST or sooner if new information becomes available.
provider
- investigating
We can confirm EC2 instance launch failures affecting all AZs (usw2-az1, usw2-az2, usw2-az3, usw2-az4) in the US-WEST-2 Region. We have identified the issue to be caused by a subsystem that supports EBS Volumes required during instance launch. Services that instantiate new instances, such as cross-region data copy for DataSync, within the US-WEST-2 Region as either the source or destination are also affected. We recommend that customers do not mutate running instances at this time that would result in the replacement or creation of a new EC2 instance in the Region. Running EC2 instances and already created EBS volumes continue to operate as normal. We are seeing early signs of recovery, and continue to take steps to restore launches within the Region. We will update this post by 5:30 PM PST or earlier if we have new information to share.
provider
- resolved
Between 2:35 PM and 4:21 PM PST, we experienced elevated EC2 instance launch failures affecting all AZs (usw2-az1, usw2-az2, usw2-az3, usw2-az4) in the US-WEST-2 Region. EC2 Instance Launches, AWS services that instantiated new instances, as well as services such as DataSync or AWS Backup which require new instances to support data transfers to or from the US-WEST-2 Region were affected. Our engineers were automatically engaged and immediately began investigating the root cause. We identified the issue to be caused by a subsystem that supports EBS Volumes required during instance launch. We began remediation of that issue at 3:20 PM, and completed full recovery at 4:21 PM. We recommend customers retry requests that failed during this time. Running EC2 instances and already created EBS volumes were not impacted and continued to operate as normal. The issue has been resolved and all AWS services are now operating normally.
provider
Reports count in aggregate; nothing about you is shown.
Corrections
None so far.
Something wrong on this page? Report an error or send us a note. Every fix is listed here.
Questions about DWN-89ef96e280
Is AWS still down?
No, this incident was resolved 2026-11-07 01:12 UTC. For AWS's live state, see https://down.guide/providers/aws.
How long did it last?
2h 21m, from 2026-11-06 22:51 UTC to 2026-11-07 01:12 UTC, by the times AWS posted.
What was affected?
AWS named Amazon Elastic Compute Cloud in us-west-2.
What caused it?
AWS hasn't published a root cause. If it does, it shows here.
Where does this come from?
AWS's official status page, https://health.aws.amazon.com/health/status, quoted verbatim. down.guide's own probes, reports and social counts are labeled as ours.