-
Notifications
You must be signed in to change notification settings - Fork 240
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Pod stuck in ContainerCreating
state with Failed to find the master interface
warning.
#1945
Comments
@tamilmani1989 do you recognize this error? This is CNIv1. |
@rbtr @tamilmani1989 Is there anything I could do to better probe into this challenge? |
@nawazkh apologies for delay. somehow I didn't get notification. The error looks like cni couldn't find master interface(eth0) based on the subnet prefix returned from azure-vnet-ipam. can you share all files starting with azure-vnet prefix
|
Shared access to the above logs via Bastion. |
out-azure-vnet.log Looking at the log at for example 23:08:21 (line 19077 on the azure-vnet log) the azure-vnet did not get a master interface from ipam return call. However, IPAM log shows that it actually returns an Interface (eth1) at that timestamp (line 22652 on ipam log) |
Checked the log. One thing I noticed in the azure-vnet.log is that eth1 does not have the IP address assigned.
The IPAM results an IP address 10.2.0.99 and mask ffff0000 [From the logs]. To find the master interface we loop through all the interfaces and find the first interface with IP address in that subnet[ 10.2.0.0/16]. here. But since there is no interface with the IP address from that subnet, the master interface is not found.( the error we see) All the interfaces from the interfaces i can see in the logs:
@nawazkh eth1 interface is not expected to have ip here? Also you can you please run the cmd mentioned below and share the output with us.
|
Got the following output
|
@nawazkh probably you have to follow-up NMAgent on this. Looks like eth1 IP is not configured in VM which is why CNI failing. eth1 IP is expected to be configured through dhcp by nmagent |
@nawazkh can we close this if you don't have anything else? |
@tamilmani1989 I did not get a chance to follow up with the NMAgent on this. Can I keep this open since it is still a valid issue from CAPZ's perspective? Or can you point me to the repo to transfer the issue to? |
Please open separate support ticket with them via microsoft icm channel |
Makes sense, closing this issue for now :) |
What happened:
webaaw2ci-84cc489f6f-d22gj
is stuck onContainerCreating
status in a self managed cluster.webaaw2ci-84cc489f6f-d22gj
, we see the below events in its description. (Posting the whole description of the pod to point that IP was also not allocated to the pod.)What you expected to happen:
webaaw2ci-84cc489f6f-d22gj
to get scheduled and get intoRunning
state.How to reproduce it:
templates/flavors/azure-cni-v1/patches/azure-machine-template.yaml
update each of thenode-subnet-x
's privateIPConfig value to50
.GINKGO_FOCUS="Azure CNI v1" LOCAL_ONLY=true SKIP_LOG_COLLECTION=true SKIP_CLEANUP=true ./scripts/ci-e2e.sh
webaaw2ci-xxxxx
stuck inContainerCreating
state.Orchestrator and Version (e.g. Kubernetes, Docker):
v1.25.9
Kernel (e.g.
uanme -a
for Linux or$(Get-ItemProperty -Path "C:\windows\system32\hal.dll").VersionInfo.FileVersion
for Windows):Anything else we need to know?:
[Miscellaneous information that will assist in solving the issue.]
The text was updated successfully, but these errors were encountered: