ENA Driver & Interface Diagnostic Troubleshooter
Use the interactive troubleshooter below to identify your ENA driver or interface error by symptom, review the raw error log, understand the root cause, and apply the recommended fix.
🚨 Step 1: What specific error symptom are you experiencing?
Please click the most accurate description:
Quick Reference Table
| # | Scenario | Key Error Signal | Root Cause | The Fix |
|---|---|---|---|---|
| 1 | Packets destined for proxy services (such as IMDS or VPC DNS) are dropped due to rate limiting. | linklocal_allowance_exceeded: N | The number of proxy packets exceeded the packets per second (PPS) allowance limit specifically allocated for link-local traffic. | N/A |
| 2 | Brief period of traffic loss or intermittent unreachability caused by the network driver abruptly resetting the interface. | Keep alive watchdog timeout | The ENA driver did not receive a keep-alive event from the device, triggering the watchdog mechanism to initiate a device and driver state reset. | aws ec2 modify-instance-attribute --instance-id instance_id --no-ena-support |
| 3 | Failures during ENA device initialization leading to potential network adapter failure or missing interfaces. | reading reg failed for timeout | Memory mapped I/O (MMIO) read operations timed out due to an incompatible driver, a busy hardware device, or physical hardware failure. | N/A |
| 4 | The network interface fails to come up because the OS expects a different MAC address than the one assigned by the VPC fabric. | Device eth0 has different MAC address than expected | There is a hardcoded interface MAC address statically defined in the AMI configuration. | N/A |