Open kalyanteja19 opened 1 month ago
@jcmoraisjr , We need your help to resolve this issue at earliest, can you please assist. Currently, all of our 100+ haproxy applications are facing this issue.
Hi, is there any error message you can share, so we can better identify where the problem is.
/start.sh permission denied
Hi @jcmoraisjr ,
We are getting error as below when we check the logs of haproxy ingress pod.
/start.sh permission denied
Wondering how you're building your customized image. It seems it's working as expected when I build the image myself:
$ ~> cd ~/.../haproxy-ingress/
$ ~/j/g/s/g/j/haproxy-ingress (master)> make image
...
Successfully tagged localhost/haproxy-ingress:latest
$ ~/j/g/s/g/j/haproxy-ingress (master)> docker images |head
REPOSITORY TAG IMAGE ID CREATED SIZE
localhost/haproxy-ingress latest 0ebcc31e4278 8 seconds ago 92.6MB
...
$ ~/j/g/s/g/j/haproxy-ingress (master)> docker run -ti --rm localhost/haproxy-ingress:latest sh
/ $ id
uid=99(haproxy) gid=99(haproxy) groups=99(haproxy)
/ $ ls -la /start.sh
-rwxr-xr-x 1 root root 1089 Jun 6 22:27 /start.sh
/ $
$ ~/j/g/s/g/j/haproxy-ingress (master)> grep -A4 ' image:' haproxy-ingress-values.yaml
image:
registry: localhost
repository: haproxy-ingress
tag: latest
pullPolicy: Never
$ ~/j/g/s/g/j/haproxy-ingress (master)> helm install haproxy-ingress haproxy-ingress/haproxy-ingress --create-namespace --namespace ingress-controller --version 0.15.0-alpha.2 --devel -f haproxy-ingress-values.yaml
NAME: haproxy-ingress
...
$ ~/j/g/s/g/j/haproxy-ingress (master)> kubectl -n ingress-controller logs pods/haproxy-ingress-557cdf96b9-jpcbd -f
I0610 13:15:12.107244 7 config.go:143] "version info" logger="config" controller-publicname="HAProxy Ingress" controller-release="local" controller-build="git-660edd4c" controller-repo="https://github.com/jcmoraisjr/haproxy-ingress.git"
...
I0610 13:15:21.912228 7 services.go:268] "finish haproxy update" logger="services" id=2 parse-ingress-ms=1.25852 write-maps-ms=0.094819 total-ms=1.353339
Wondering how you're building your customized image. It seems it's working as expected when I build the image myself:
$ ~> cd ~/.../haproxy-ingress/ $ ~/j/g/s/g/j/haproxy-ingress (master)> make image ... Successfully tagged localhost/haproxy-ingress:latest $ ~/j/g/s/g/j/haproxy-ingress (master)> docker images |head REPOSITORY TAG IMAGE ID CREATED SIZE localhost/haproxy-ingress latest 0ebcc31e4278 8 seconds ago 92.6MB ... $ ~/j/g/s/g/j/haproxy-ingress (master)> docker run -ti --rm localhost/haproxy-ingress:latest sh / $ id uid=99(haproxy) gid=99(haproxy) groups=99(haproxy) / $ ls -la /start.sh -rwxr-xr-x 1 root root 1089 Jun 6 22:27 /start.sh / $ $ ~/j/g/s/g/j/haproxy-ingress (master)> grep -A4 ' image:' haproxy-ingress-values.yaml image: registry: localhost repository: haproxy-ingress tag: latest pullPolicy: Never $ ~/j/g/s/g/j/haproxy-ingress (master)> helm install haproxy-ingress haproxy-ingress/haproxy-ingress --create-namespace --namespace ingress-controller --version 0.15.0-alpha.2 --devel -f haproxy-ingress-values.yaml NAME: haproxy-ingress ... $ ~/j/g/s/g/j/haproxy-ingress (master)> kubectl -n ingress-controller logs pods/haproxy-ingress-557cdf96b9-jpcbd -f I0610 13:15:12.107244 7 config.go:143] "version info" logger="config" controller-publicname="HAProxy Ingress" controller-release="local" controller-build="git-660edd4c" controller-repo="https://github.com/jcmoraisjr/haproxy-ingress.git" ... I0610 13:15:21.912228 7 services.go:268] "finish haproxy update" logger="services" id=2 parse-ingress-ms=1.25852 write-maps-ms=0.094819 total-ms=1.353339
@jcmoraisjr, sure. We are using make docker-build command as in the readme file. Here is the error when we run make docker-build.
We will try with make image option and let you know.
Also, we want to deploy it on our production clusters. As seen in notes, alpha is not recommended, please suggest any timelines for next production release.
Description of the problem Initially, we are having issues with leader election. After pod restart, haproxy is working fine. Hence, we build the image based on the code from master branch (based on pull request 1087 merge). Now, we are getting /start.sh permission denied during initialization of pod. Pod is going into crashloopbackoff state.
Expected behavior
Pod should run without any issues.
Steps to reproduce the problem
Environment information
HAProxy Ingress version:
based on master code