Final Update: Friday, October 13th 2017 11:47 UTC
We’ve confirmed that all systems are back to normal as of October 13th at 11:25 UTC. Our logs show the incident started on October 13th at 10:40 UTC and that during the 45 minute window that it took to resolve the issue. During the affected time period customers would experience slow or failed commands in the North Central US region.
Issue was mitigated by recycling the affected Web Tier.
Sorry for any inconvenience this may have caused.
- Root Cause: Current investigation is pointing to a low memory condition on one of the Web Tiers. We have collected diagnostic data and are in the process of doing further investigation.
- Chance of Re-occurrence: High as root cause is not fully understood.
- Incident Timeline: 45 minutes – October 13th at 10:40 UTC through October 13th at 11:25 UTC
Initial Update: Friday, October 13th 2017 11:15 UTC
A potentially customer impacting alert is being investigated. Triage is in progress and we will provide an update with more information.
- Next Update: Before Friday, October 13th 2017 12:30 UTC