Skip to content

Run exec commands on targeted containers instead of all in a pod #213

@mrajagopal

Description

@mrajagopal

Describe the bug
When a customer deploys an NGINX Ingress Controller (NIC) pod with NGINX App-Protect (NAP) the exec commands such as nginx -T and ./nginx-ingress version throw an error when run in containers that are not nginx-ingress such as waf-enforcer and waf-config-mgr.

To Reproduce
Steps to reproduce the behavior:

  1. Deploy NIC+NAP
  2. Run the nginx-supportpkg-for-k8s
....
Running job helm-info.....................  COMPLETED
Running job helm-deployments..............  COMPLETED
Running job exec-nginx-ingress-version.... Files collected so far: [/exec/v5waf-0/v5waf-0-nginx-ingress-controller-5d797d9668-f92bg__nginx-ingress__nginx-ingress-version.txt]
 FAILED: Internal error occurred: error executing command in container: failed to exec in container: failed to start exec "78f64b20ea95c5d3a2db3d7d84ad530c024ba422893678e383827bb6f30920a3": OCI runtime exec failed: exec failed: unable to start container process: exec: "./nginx-ingress": stat ./nginx-ingress: no such file or directory: unknown
Running job exec-nginx-t.................. Files collected so far: [/exec/v5waf-0/v5waf-0-nginx-ingress-controller-5d797d9668-f92bg__nginx-ingress__nginx-t.txt]
 FAILED: Internal error occurred: error executing command in container: failed to exec in container: failed to start exec "58fcf9620fa2c654e66e2eef0d4c1a37728738b8bee8ad4f85856d8ab7b5f47a": OCI runtime exec failed: exec failed: unable to start container process: exec: "/usr/sbin/nginx": stat /usr/sbin/nginx: no such file or directory: unknown
Running job exec-agent-conf............... Files collected so far: [/exec/v5waf-0/v5waf-0-nginx-ingress-controller-5d797d9668-f92bg__nginx-ingress__nginx-agent.conf]
 FAILED: Internal error occurred: error executing command in container: failed to exec in container: failed to start exec "5feab885e2fd18e0478dae5612e4740ab10550435fb5e9e842e9b11b700daf58": OCI runtime exec failed: exec failed: unable to start container process: exec: "cat": executable file not found in $PATH: unknown
Running job exec-agent-version............ Files collected so far: [/exec/v5waf-0/v5waf-0-nginx-ingress-controller-5d797d9668-f92bg__nginx-ingress__nginx-agent-version.txt]
 FAILED: Internal error occurred: error executing command in container: failed to exec in container: failed to start exec "fc249c7c602e0de95c9cd1de71aa2c5b1f732bbda05badf77dc5a47e5db4a098": OCI runtime exec failed: exec failed: unable to start container process: exec: "/usr/bin/nginx-agent": stat /usr/bin/nginx-agent: no such file or directory: unknown
Running job crd-objects...................  COMPLETED
Running job collect-product-platform-info.  COMPLETED
WARNING: 4 failed job(s)
Supportpkg generated with warnings: nic-supportpkg-1759894775.tar.gz

Expected behavior
The supportpkg tool must run exec commands against the target container instead of all containers in the pod

Environment
k8s cluster

Client Version: v1.32.3
Kustomize Version: v5.5.0
Server Version: v1.33.5

Metadata

Metadata

Assignees

Labels

bugSomething isn't working

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions