Export not Collecting metric lbvserver

[estefancaique]

After configuring netscaler-metrics-exporter it is not returning VPX lbvserver

[estefancaique]

What message is this that appears in the log

019-09-04T20:45:26+0000 INFO Collecting metric lbvserver for xxxxxx 2019-09-04T20:45:26+0000 INFO metrices for lbvserver with k8sprefix "VIP" are not fetched 2019-09-04T20:45:26+0000 WARNING Counter stats for toleratingttlbtransactionsrate not enabled in netscalar xxxxxx, so could not add to lbvserver 2019-09-04T20:45:26+0000 WARNING Counter stats for totcltttlbtransactions not enabled in netscalar xxxxxx, so could not add to lbvserver 2019-09-04T20:45:26+0000 WARNING Counter stats for frustratingttlbtransactions not enabled in netscalar xxxxxx, so could not add to lbvserver 2019-09-04T20:45:26+0000 WARNING Counter stats for toleratingttlbtransactions not enabled in netscalar xxxxxx, so could not add to lbvserver 2019-09-04T20:45:26+0000 WARNING Gauge stats for frustratingttlbtransactionsrate not enabled in netscalar xxxxxx, so could not add to lbvserver 2019-09-04T20:45:26+0000 WARNING Gauge stats for cltresponsetimeapdex not enabled in netscalar xxxxxx, so could not add to lbvserver 2019-09-04T20:45:26+0000 WARNING Gauge stats for avgcltttlb not enabled in netscalar xxxxxx, so could not add to lbvserver 2019-09-04T20:45:26+0000 WARNING Gauge stats for cltttlbtransactionsrate not enabled in netscalar xxxxxx, so could not add to lbvserver

[aroraharsh23]

Can you confirm the environment you are running this VPX with, i meant whether the exporter is run as : script/docker/k8s setup/k8s with CIC. This message is for the case when the exporter is run with Citrix ingress controller in k8s environment, there you need to provide the prefix used by CIC(will update the details in READMe). For now, check exporter help for that. If you are not in CIC setup then, we can debug the issue.

[aroraharsh23]

We have updated the repo name and image directory, for any further testing please confirm on quay.io/citrix/citrix-adc-metrics-exporter:1.1 or the new "exporter.py" script. Thanks.

[marcoapvilhena]

@aroraharsh23 we aren't using VPX as ingress controller. Our VPX is used only for Load Balance for non Kubernetes applications.

[aroraharsh23]

@marcoapvilhena Thanks for confirming, You can try with the updated script or container image, and let us know, if the issue stills persists or not. quay.io/citrix/citrix-adc-metrics-exporter:1.1

[marcoapvilhena]

@aroraharsh23 We got the test after your update but we had the same issue. We found "metrices for lbvserver with k8sCICprefix "VIP" are not fetched" in the log

[aroraharsh23]

@marcoapvilhena Thanks for testing. Just to clarify again, you are not running exporter as a pod as you mentioned, then, this log should not appear as it's placed under "KUBERNETES_SERVICE_HOST" environment variable. Is this variable set by any chance in your case? That should not be the case in of non-k8s. Or you can try deleting the docker image/script and pulling it again, which i am suspecting is still the previous one.

[HatBeardZero]

I have the same issue running it as a cron/nohup setup.

2019-09-11T18:39:37+0000 INFO metrices for lbvserver with k8sprefix "redacted" are not fetched

[marcoapvilhena]

@aroraharsh23 we are running this exporter as a pod, but our VPX doesn't work as ingress, only to balancing for non-kubernetes aplication.

The "KUBERNETES_SERVICE_HOST" variable is not setted. Bellow has the pod describe:

Name: netscaler-metrics-exporter Namespace: monitoring Priority: 0 PriorityClassName: Node: xxxxxx/xxxxx Start Time: Thu, 05 Sep 2019 11:40:59 -0300 Labels: app=exporter Annotations: kubectl.kubernetes.io/last-applied-configuration: {"apiVersion":"v1","kind":"Pod","metadata":{"annotations":{},"labels":{"app":"exporter"},"name":"netscaler-metrics-exporter","namespace":"... kubernetes.io/psp: grafana-test Status: Running IP: xxxxxxx Containers: exporter: Container ID: containerd://5092e68bddbc0c759f14e8fe9d83432b7081799b27024638f5c658cf98add040 Image: quay.io/citrix/citrix-adc-metrics-exporter:1.1 Image ID: sha256:b4343090e213bc85844bb29c8024b9bda741c4f65c99464292a97825687c0819 Port: Host Port: Args: --target-nsip=xxxxxx:80 --port=8888 --username=root --password=xxxxxx State: Running Started: Thu, 05 Sep 2019 11:41:08 -0300 Ready: True Restart Count: 0 Environment: Mounts: /var/run/secrets/kubernetes.io/serviceaccount from default-token-lnvbq (ro) Conditions: Type Status Initialized True Ready True ContainersReady True PodScheduled True Volumes: default-token-lnvbq: Type: Secret (a volume populated by a Secret) SecretName: default-token-lnvbq Optional: false QoS Class: BestEffort Node-Selectors: Tolerations: node.kubernetes.io/not-ready:NoExecute for 600s node.kubernetes.io/unreachable:NoExecute for 600s Events:

[aroraharsh23]

@HatBeardZero If you are not running exporter as a k8s pod, issue is not the same. As per the log, you are at an older version, Please just update to image "quay.io/citrix/citrix-adc-metrics-exporter:1.2" and it will get resolved.

This issue only happens if exporter is run as a pod while Citrix ingress controller is not used. We are working on it.

[aroraharsh23]

@marcoapvilhena Looks like that we misunderstood your deployment. So, it's a pod after all But non-CIC. In that case, that was a known issue, which we have fixed in 1.2, please update the same: "quay.io/citrix/citrix-adc-metrics-exporter:1.2"

[marcoapvilhena]

@aroraharsh23 now is working fine. Thank you for help!!

[aroraharsh23]

@marcoapvilhena Thanks for confirming, closing the issue.