Timing out connections to Azure

Question

Timing out connections to Azure

Kerkhof, S 0

The last 4 hours we experience company-wide issues connecting to Azure, the status page of Microsoft tells us all is fine. Still we are timing out on connection attempts, it is becoming a serious issue. More details: a ping-test displays the issue only with Azure/Microsoft servers, 2 other test servers, don't display a packet-loss. Sometimes, after refreshing a few times, a page loads.

0 comments

1 answer

Your answer

Answer 1

Welcome to Microsoft Q&A!

Hello @Kerkhof, S I hope you are doing well,

The public Azure Status page only reflects broad, catastrophic outages affecting entire regions or fundamental platform services. When packet loss and timeouts occur company-wide while other internet targets respond normally, the failure is usually isolated to the transit path between your enterprise network edge and the Microsoft Enterprise Edge (MSEE/AS8075), or to specific regional gateway routing rather than a full cloud outage.

Unlike the public status dashboard, Azure Service Health provides tenant-specific and subscription-level visibility. Sign in to the Azure Portal and check Service Health > Health advisories and Resource health. Regional network edge degradations, BGP flapping, and fiber maintenance alerts are posted here rather than on the public board.

Locate the Drop with Pathping or MTR

Standard ping confirms packet loss but cannot pinpoint the responsible network hop. Run an extended trace from an affected workstation:

Bash

# On Windows CMD or PowerShell:

Loss before Microsoft hops: If latency or packet drops start within your ISP's Autonomous System (AS) or upstream Tier-1 transit providers, the issue is an ISP peering or BGP route issue to Microsoft.

Loss at the Microsoft edge: If packets cleanly traverse your ISP and begin dropping once hitting Microsoft Point of Presence (PoP) or Edge routers, the issue is an ingestion bottleneck on that specific peering edge.

Test Outside Your Corporate Egress

To rule out local firewall state-table exhaustion, MTU black holes, or proxy inspection delays:

Disconnect a test laptop from the corporate network/VPN and connect via an independent cellular hotspot.
If connections succeed immediately on the hotspot, the issue is rooted in your corporate egress firewall, NAT pool exhaustion, or local edge router.
If connections still time out across independent networks, submit an Azure support request under Subscription Management / Technical Support citing your source public IP, target endpoint, and the pathping output so network engineers can check the specific edge route.

Please could you share us if you find something testing outside your corporate network

Help make this community better for everyone: if this answer resolved your issue, please accept it or leave an upvote. If not, share more details in a comment so we can continue the discussion and find the right solution.