Sudden spike in demand causes issues in Azure East US region
(2025/08/08)
- Reference: 1754661609
- News link: https://www.theregister.co.uk/2025/08/08/sudden_spike_in_demand_azure_issues/
- Source link:
A problem with resources for virtual machines is still affecting users in Azure's East US region after more than a week, frustrated admins have told us, despite Microsoft saying the incident is now resolved.
Customers using this Microsoft Azure cloud region experienced allocation failures when attempting to create or update virtual machines. The troubles began on July 29, and were supposedly rectified by August 5.
According to details seen by The Register , the issue was caused by a sudden spike in demand for compute resources in the East US region, leading to service management operations failing due to insufficient capacity being available.
[1]
The resource pool for General Compute instances became highly constrained, affecting all associated sets in the region. This surge in utilization actually pushed hardware beyond safe operational thresholds, according to Microsoft, which prevented successful virtual machine allocations.
[2]
[3]
Such problems only appeared to affect a handful of instance types, however, and Microsoft recommended using alternate types or switching to the East US 2 region as workarounds.
A hint to what may have precipitated this came from one of our readers:
[4]
"Capacity issues with several generations of Intel and AMD VMs have resulted in an epidemic of the dreaded ZonalAllocationFailed error for AKS clusters when trying to upgrade them, just during the week when many had to upgrade to k8s 1.31 from 1.30 (which went EOL at the end of the month)," our source said.
Microsoft said in a Mitigation Statement that after a period of monitoring telemetry, it could confirm that the service had now been restored, and customer impact was mitigated. The [5]Azure status page online was also showing no currently active events at the time of publication.
Not according to Reg readers. "Although the incident has been marked resolved, in practice it lingers," one told us.
[6]
"There was also a wave of incidents and cancelled maintenances around the same time in the same region, hinting that the capacity issues may have hit Microsoft's own services as well," they added.
[7]Google agrees to pause AI workloads to protect the grid when power demand spikes
[8]Radar problem caused mayhem in UK skies on Wednesday
[9]Cloudflare fesses up to config change that borked internet access for all
[10]14-hour+ global blackout at Ingram Micro halts customer orders
We asked Microsoft for clarification yesterday, and despite a spokesperson logging our request, no response has yet landed in our mailbox. We're still waiting for our [11]Crocs too!
The situation follows an outage earlier in July that saw [12]Microsoft Outlook go dark for millions of users worldwide, lasting more than 11 hours.
Earlier in the year, another [13]Azure outage affected users in Norway , hitting businesses and many government websites delivering online services to citizens. ®
Get our [14]Tech Resources
[1] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aJYfFCyOs7CxP-czG1G4ywAAAMQ&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aJYfFCyOs7CxP-czG1G4ywAAAMQ&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aJYfFCyOs7CxP-czG1G4ywAAAMQ&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aJYfFCyOs7CxP-czG1G4ywAAAMQ&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[5] https://azure.status.microsoft/en-us/status
[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aJYfFCyOs7CxP-czG1G4ywAAAMQ&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[7] https://www.theregister.com/2025/08/04/google_ai_datacenter_grid/
[8] https://www.theregister.com/2025/07/30/uk_airspace_outage/
[9] https://www.theregister.com/2025/07/16/cloudflare_fesses_up_to_config/
[10] https://www.theregister.com/2025/07/04/ingram_micro_technical_difficulties/
[11] https://www.theregister.com/2025/08/06/microsoft_reportedly_about_to_unleash/
[12] https://www.theregister.com/2025/07/10/outlook_microsoft_email_outage/
[13] https://www.theregister.com/2025/02/20/microsoft_azure_outage_norway/
[14] https://whitepapers.theregister.com/
Customers using this Microsoft Azure cloud region experienced allocation failures when attempting to create or update virtual machines. The troubles began on July 29, and were supposedly rectified by August 5.
According to details seen by The Register , the issue was caused by a sudden spike in demand for compute resources in the East US region, leading to service management operations failing due to insufficient capacity being available.
[1]
The resource pool for General Compute instances became highly constrained, affecting all associated sets in the region. This surge in utilization actually pushed hardware beyond safe operational thresholds, according to Microsoft, which prevented successful virtual machine allocations.
[2]
[3]
Such problems only appeared to affect a handful of instance types, however, and Microsoft recommended using alternate types or switching to the East US 2 region as workarounds.
A hint to what may have precipitated this came from one of our readers:
[4]
"Capacity issues with several generations of Intel and AMD VMs have resulted in an epidemic of the dreaded ZonalAllocationFailed error for AKS clusters when trying to upgrade them, just during the week when many had to upgrade to k8s 1.31 from 1.30 (which went EOL at the end of the month)," our source said.
Microsoft said in a Mitigation Statement that after a period of monitoring telemetry, it could confirm that the service had now been restored, and customer impact was mitigated. The [5]Azure status page online was also showing no currently active events at the time of publication.
Not according to Reg readers. "Although the incident has been marked resolved, in practice it lingers," one told us.
[6]
"There was also a wave of incidents and cancelled maintenances around the same time in the same region, hinting that the capacity issues may have hit Microsoft's own services as well," they added.
[7]Google agrees to pause AI workloads to protect the grid when power demand spikes
[8]Radar problem caused mayhem in UK skies on Wednesday
[9]Cloudflare fesses up to config change that borked internet access for all
[10]14-hour+ global blackout at Ingram Micro halts customer orders
We asked Microsoft for clarification yesterday, and despite a spokesperson logging our request, no response has yet landed in our mailbox. We're still waiting for our [11]Crocs too!
The situation follows an outage earlier in July that saw [12]Microsoft Outlook go dark for millions of users worldwide, lasting more than 11 hours.
Earlier in the year, another [13]Azure outage affected users in Norway , hitting businesses and many government websites delivering online services to citizens. ®
Get our [14]Tech Resources
[1] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aJYfFCyOs7CxP-czG1G4ywAAAMQ&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aJYfFCyOs7CxP-czG1G4ywAAAMQ&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aJYfFCyOs7CxP-czG1G4ywAAAMQ&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aJYfFCyOs7CxP-czG1G4ywAAAMQ&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[5] https://azure.status.microsoft/en-us/status
[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aJYfFCyOs7CxP-czG1G4ywAAAMQ&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[7] https://www.theregister.com/2025/08/04/google_ai_datacenter_grid/
[8] https://www.theregister.com/2025/07/30/uk_airspace_outage/
[9] https://www.theregister.com/2025/07/16/cloudflare_fesses_up_to_config/
[10] https://www.theregister.com/2025/07/04/ingram_micro_technical_difficulties/
[11] https://www.theregister.com/2025/08/06/microsoft_reportedly_about_to_unleash/
[12] https://www.theregister.com/2025/07/10/outlook_microsoft_email_outage/
[13] https://www.theregister.com/2025/02/20/microsoft_azure_outage_norway/
[14] https://whitepapers.theregister.com/