Server lost connection, logs show ena5 connection issues

Question

Our instance i-052cca656183d431d showed running but nothing could connect.  I ended up stopping (let is try for 10 -15 minutes), then forcing to stop it (stopped roughly 5 min later), and once it came back up I saw network issues in the log right before it crashed

ena: Reading reg failed for timeout. expected: req id[10] offset[88] actual: req id[57015] offset[88]
ena: Reg read32 timeout occurred
systemd-networkd[24263]: ens5: Lost carrier
systemd-networkd[24263]: ens5: DHCP lease lost

and then a few minutes later no more logs until the reboot was successful.

Is there an issue with Amazon network here?

Thanks
Jon

Answer

Hi. At this point, I do not see any active issues with EC2 or any other related services which has been reported on the PHD. 
Link- https://health.aws.amazon.com/health/status

There could be multiple elements if you are trying to reach the instance from the on-premise/AWS environment etc. Check local FW settings or the ISP side issues, if any.

You can also check if there was any PHD notifications that you received in your email/registered account.

Checking the instance at this moment using internal tools does not show any issues as per the health checks. It also shows it is running for 45 minutes, however, for any additional checks, I would recommend opening up a support case.

Server lost connection, logs show ena5 connection issues

Relevant content