EC2: eth0: Could not set DHCPv4 address: Connection timed out

0

Hi all,

our EC2 instance today crashed with the following logs in the syslog:

Nov 01 20:56:55 ip-10-1-0-236 systemd-networkd[1695219]: eth0: Could not set DHCPv4 address: Connection timed out
Nov 01 20:56:56 ip-10-1-0-236 systemd-networkd[1695219]: eth0: Failed
-- Reboot --
Nov 01 21:15:55 ip-10-1-0-236 kernel: Linux version 5.15.0-1049-aws (buildd@lcy02-amd64-042) (gcc (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0, GNU ld (GNU Binutils for Ubuntu) 2.34) #54~20.04.1-Ubuntu SMP Fri Oct 6 22:04:33 UTC >
Nov 01 21:15:55 ip-10-1-0-236 kernel: Command line: BOOT_IMAGE=/boot/vmlinuz-5.15.0-1049-aws root=PARTUUID=24ca9e81-01 ro console=tty1 console=ttyS0 nvme_core.io_timeout=4294967295 panic=-1

there is a similar question on re:Post, but the answer sends the OP to an archived page, so the information is no longer visible. Any idea why the EC2 stalled?

pbocan
已提问 7 个月前511 查看次数
2 回答
0

Hello.

What kind of applications are you running on EC2?
Also, is it possible that the application has increased the load on EC2?
Please check the CPU usage rate and memory usage rate.

profile picture
专家
已回答 7 个月前
  • A handful of services written in Go plus a datadog agent. I am afraid we don't have memory usage rate metrics exported to CW, but the CPU metrics show 1.3-1.4 CPU used. We use CodeDeploy to deploy onto this EC2 instance...

    EDIT: I do see a CloudWatch alarm being raised for the 90+% utilisation for an extended period (5+ minutes).

  • I suspect that networkd may not be able to communicate with the kernel due to CPU or memory load.

  • Yes, I think that one of the services spiraled and crashed.

0

if you have nmap installed you can try

sudo nmap --script broadcast-dhcp-discover -e eth0

to see what is going on.

Also maybe check the status of dhcp service and restart it.

emrah
已回答 7 个月前
  • I am guessing that will not help me post mortem. This, I guess works for when it is happening.

您未登录。 登录 发布回答。

一个好的回答可以清楚地解答问题和提供建设性反馈,并能促进提问者的职业发展。

回答问题的准则