Hello, I want to deploy a signoz cluster using the...
# support
v
Hello, I want to deploy a signoz cluster using the helm chart however the documentation is a bit inconsistent and I'm not sure what's true. In the documentation at https://signoz.io/docs/install/kubernetes/gcp/ it says to override the following values to set the storage class
Copy code
clickhouse:
  cloud: gcp
  installCustomStorageClass: true
  persistence:
    storageClass: gce-resizable
But the documentation then links to the github (https://github.com/SigNoz/charts/tree/main/charts/signoz#configuration) but these values are not documented. There are some of them that are similar but on a different component
Copy code
alertmanager.persistence.enabled	Enable volume persistence	true
alertmanager.persistence.storageClass	Set storage class for persistent volume	nil
alertmanager.persistence.accessModes	Set access mode for persistent volume	[ReadWriteOnce]
alertmanager.persistence.size	Set storage size	100Mi
1
Also I'm getting the following error when installing the helm chart : Error: parse error at (signoz/templates/otel-collector-metrics/config.yaml:9): unexpected unclosed action in command
using the command
Copy code
helm --namespace platform install my-release signoz/signoz -f override-values.yaml
p
@Prashant Shahi any insights on this?
p
@Valentin Baert URL for configuration above is for SigNoz components. For clickhouse configurations, please refer to the following: https://github.com/SigNoz/charts/blob/main/charts/clickhouse/values.yaml It will be more streamlines after the following PR is merged. https://github.com/SigNoz/charts/pull/31
v
Thanks, so the doc at https://signoz.io/docs/install/kubernetes/gcp/ may be updated too because it has the wrong linked documentation if I follow you ? It says : "Here's the minimal required
override-values.yaml
that we'll be using later. You can find an overview of the parameters that can be configured during installation under chart configuration." which lead me to the github mentionned earlier. Also do you have insights about this error ? Error: parse error at (signoz/templates/otel-collector-metrics/config.yaml:9): unexpected unclosed action in command
p
Not able to reproduce that error on my machine. Perhaps, your local helm repo points to older version of chart. Try updating and print versions:
Copy code
helm repo update

helm search repo signoz --versions
we had that config error resolve some time back. It was caused by Helm - it happens when using
helm upgrade
along with
-f
flag.
v
I updated the repo but the issue remains
Copy code
$ helm search repo signoz --versions
NAME                    CHART VERSION   APP VERSION     DESCRIPTION                                
signoz/signoz           0.0.16          0.8.0           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.15          0.8.0           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.14          0.8.0           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.13          0.8.0           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.12          0.7.5           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.11          0.7.4           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.10          0.7.1           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.9           0.7.1           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.8           0.6.2           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.7           0.6.1           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.6           0.6.1           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.5           0.6.0           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.4           0.5.4           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.3           0.5.4           SigNoz Observability Platform Helm Chart   
signoz/signoz           0.0.2           0.5.4           SigNoz Observability Platform Helm Chart   
signoz/alertmanager     0.5.2           0.5.0           The Alertmanager handles alerts for SigNoz.
signoz/alertmanager     0.5.1           0.5.0           The Alertmanager handles alerts for SigNoz.
signoz/alertmanager     0.5.0           0.5.0           The Alertmanager handles alerts for SigNoz.
signoz/clickhouse       16.0.5          21.12.3.32      A Helm chart for ClickHouse                
signoz/clickhouse       16.0.4          21.12.3.32      A Helm chart for ClickHouse                
signoz/clickhouse       16.0.3          21.12.3.32      A Helm chart for ClickHouse                
signoz/clickhouse       16.0.2          21.12.3.32      A Helm chart for ClickHouse                
signoz/clickhouse       16.0.1          21.12.3.32      A Helm chart for ClickHouse                
signoz/clickhouse       16.0.0          21.12.3.32      A Helm chart for ClickHouse                
signoz/clickhouse       9.1.0           21.7            A Helm chart for ClickHouse
p
can you share which Helm and K8s versions are you using?
v
kubernetes 1.21.11-gke.900
helm 3.3.0
p
can you upgrade your helm?
v
I have updated to the latest helm 3.9.0, now it worked !
🎉 1
p
Awesome!
v
thanks
p
you are welcome 🙂
v
I have another issue, when I set the clickhouse.user to admin all the pods start properly but I can't login on the frontend because there is a javascript check that blocks the login because the username is expected to be an email If I delete the release and then set the clickhouse.user to an valid email then install the chart again, the pods fail to start because the clickhouse server pod has the following error when starting : <Error> Application: Exception: Failed to merge config with '/etc/clickhouse-server/users.d/chop-generated-users.xml': SAXParseException: Invalid token in '/etc/clickhouse-server/users.d/chop-generated-users.xml', line 12 column 16
Not really sure what to do since I'm stuck in this loop
p
clickhouse.user
is ClickHouseDB-level user, while the frontend user is SigNoz-level user, which is stored in SQLite.
v
p
no worries 🙂