Version of Helm and Kubernetes:
v1.9.2+coreos.0
Which chart:
sonatype-nexus-1.1.1
What happened:
Since the proxy container was added, I am not able to access nexus. Going to the IP address of the load balancer results in Internal Server Error.
What you expected to happen:
How to reproduce it (as minimally and precisely as possible):
This is my configuration:
# configuration for sonatype/nexus
nexus:
imageTag: 3.6.1
nodeSelector:
failure-domain.beta.kubernetes.io/zone: de-nbg6-1b
serviceType: LoadBalancer
ingress:
enabled: false
persistence:
enabled: true
accessMode: ReadWriteOnce
storageClass: default-de-nbg6-1b
storageSize: 50Gi
nexusBackup:
#nexusAdminPassword: t0ps3kr3t
persistence:
enabled: false
#storageClass: default-de-nbg6-1b
#storageSize: 3Gi
I deployed with:
helm upgrade -f nexus.yml nude-nexus stable/sonatype-nexus
I believe my problem is due to the fact that I didn't set
nexusProxy:
env:
nexusDockerHost: container.example.com
nexusHttpHost: nexus.example.com
How ever, at the moment I don't know how to create a redirect from the dns record to the correct IP (which I still don't know because the Pod isn't scheduled before I do helm apply).
I would appreciate any help here.
having the same issue on helm chart v 1.4.0 +1
Same issue too. v. sonatype-nexus-1.5.1
Log file:
kubectl logs nexus-release-sonatype-nexus-654df8f789-t7rbs nexus-proxy
2018-08-15 15:52:41,586 [WARN ] [vert.x-eventloop-thread-0] [io.netty.util.internal.MacAddressUtil] Failed to find a usable hardware address from the network interfaces; using random bytes: fa:ae:94:83:de:8a:90:c9
2018-08-15 15:52:41,663 [INFO ] [vert.x-eventloop-thread-0] [com.travelaudience.nexus.proxy.BaseNexusProxyVerticle] Listening at 0.0.0.0:8080.
Same issue.
2018-09-07 02:37:02,912 [WARN ] [vert.x-eventloop-thread-0] [io.netty.util.internal.MacAddressUtil] Failed to find a usable hardware address from the network interfaces; using random bytes: f5:8a:ee:64:e4:4a:4f:32
2018-09-07 02:37:02,963 [INFO ] [vert.x-eventloop-thread-0] [com.travelaudience.nexus.proxy.BaseNexusProxyVerticle] Listening at 0.0.0.0:8080.
2018-09-07 02:38:35,657 [ERROR] [vert.x-eventloop-thread-0] [io.vertx.ext.web.impl.RoutingContextImplBase] Unexpected exception in route
java.lang.NullPointerException: null
FYI anyone seeing this could try the resolution found here. These guys actually maintain the helm chart.
https://github.com/travelaudience/nexus-proxy/issues/4
try using imageTag 2.2.0 for nexusProxy
There is no chance to disable nexus-proxy at all and use only nexus.
Is it possible to add such feature to disable proxy and point the service directly to nexus or create service for direct access to nexus?
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Any further update will cause the issue/pull request to no longer be considered stale. Thank you for your contributions.
This issue is being automatically closed due to inactivity.
There is no chance to disable nexus-proxy at all and use only nexus.
Is it possible to add such feature to disable proxy and point the service directly to nexus or create service for direct access to nexus?
I'm having the same issue. I'm doing SSL termination outside kubernetes in a haproxy installation. I was able to connect to proxy with this config but then all assets are not found since nexus HTML uses absolute urls.
frontend https-in
bind :443
[...]
# Define hosts for virtual hosting
acl nexus hdr(host) -i nexus.example.com
# Routing rules, letsencrypt must be the first rule
use_backend NEXUS if nexus
# Add host header needed by K8S ingress
http-request set-header Host nexus.lan.example.com if nexus
http-request set-header X-Forwarded-Host nexus.example.com
backend NEXUS
option httpchk GET / HTTP/1.1\r\nHost:\ nexus.lan.example.com
server nexus nexus.lan.example.com:80 check inter 60s fastinter 1s downinter 2s
How to get rid of nexus proxy and get a plain service to nexus3?? (BTW what is its purpose)
Or alternatively is there any way to make nexus produce relative links instead of absolute ones?
I'm facing the same issue. All over the internet, I don't find a solution. I don't need this proxy at all. Please did you find a solution ? I just need to access my nexus using loadbalancer for tests.
There is no chance to disable nexus-proxy at all and use only nexus.
Is it possible to add such feature to disable proxy and point the service directly to nexus or create service for direct access to nexus?I'm having the same issue. I'm doing SSL termination outside kubernetes in a haproxy installation. I was able to connect to proxy with this config but then all assets are not found since nexus HTML uses absolute urls.
frontend https-in bind :443 [...] # Define hosts for virtual hosting acl nexus hdr(host) -i nexus.example.com # Routing rules, letsencrypt must be the first rule use_backend NEXUS if nexus # Add host header needed by K8S ingress http-request set-header Host nexus.lan.example.com if nexus http-request set-header X-Forwarded-Host nexus.example.com backend NEXUS option httpchk GET / HTTP/1.1\r\nHost:\ nexus.lan.example.com server nexus nexus.lan.example.com:80 check inter 60s fastinter 1s downinter 2sHow to get rid of nexus proxy and get a plain service to nexus3?? (BTW what is its purpose)
Or alternatively is there any way to make nexus produce relative links instead of absolute ones?
See if this works:
"Update the Base URL to use https in your repository manager configuration using the Base URL capability."
https://help.sonatype.com/repomanager3/security/configuring-ssl#ConfiguringSSL-HowtoEnabletheHTTPSConnector
I'm trying to find a way to edit the base url without first logging in, since my SSL-terminating load balancer does 301s HTTP requests to HTTPS.
I solved using nginx-configuration to forward the X-Forwarded-Proto using
https://github.com/kubernetes/ingress-nginx/issues/3192#issuecomment-483281931
Most helpful comment
having the same issue on helm chart v 1.4.0 +1