Getting Support
The fastest route to an answer is one email with the right details. This page tells you who to contact, what to attach, and how to rule out the usual causes first.
Best for anything with logs, versions, or values attached. Goes straight to the support team.
Contact formPrefer a form? Use the contact page on our website.
PrerequisitesMost install problems trace back to something on this checklist.
Talk to us before you act
Contact us first, rather than after something goes wrong, for:
- Initial setup of your deployment after the first install.
- Any major MongoDB version upgrade.
- Rotating the token-signing or password-hashing secrets.
- Enabling the
archiver,analytics, ordailyMetricsjobs. - Running on a platform or Istio mode other than the reference setup.
- Getting registry credentials, plus the image tags, security report, and SBOMs that match your chart version.
- Arranging outbound system email.
What to include
A request that carries these details can usually be answered without a round trip:
- What you were doing, the exact command, and the exact error.
- How urgent it is and what it blocks.
- The chart version, image tags, and your values file with anything sensitive removed.
- Versions of Kubernetes, Istio, cert-manager, and both operators, plus your node kernel version, CNI, and StorageClass.
- Pod state, recent events, and logs from the affected components.
These commands collect most of it:
helm list -n glyph
helm get values glyph-app -n glyph
kubectl version
kubectl get pods -n glyph -o wide
kubectl get events -n glyph --sort-by=.lastTimestamp
kubectl get mongodbcommunity,rabbitmqcluster -n glyph
kubectl describe pod <pod> -n glyph
kubectl logs <pod> -n glyph --all-containers --tail=200
Common problems
| Symptom | Likely cause | What to check |
|---|---|---|
helm install fails with no matches for kind "Gateway" or "VirtualService" | Istio CRDs are not installed | Install the Istio base chart first |
helm install fails with no endpoints available for service "cluster-operator-webhook-service" | The RabbitMQ operator webhook is not routable yet | Wait for endpoints in rabbitmq-system, then retry |
Server pod in CreateContainerConfigError | The application Secret or one of its keys is missing | kubectl describe pod, then compare keys with Prerequisites |
Server pod stuck in Init | MongoDB is not Running yet | kubectl get mongodbcommunity -n glyph |
| MongoDB crash loops on a new cluster | Node kernel affected by the MongoDB 8 startup check | uname -r on the nodes |
ErrImagePull or ImagePullBackOff | Wrong repository path, missing registry credentials, or a malformed reference from global.imageRegistry | kubectl describe pod, then see Configuration |
| Browser shows a certificate error or no response | TLS Secret missing from istio-system, wrong name, or no ingress gateway labeled istio: ingress | Check the Secret name against ingress.tls.secretName and the gateway pod labels |
| Page loads but the app cannot sign in or resolve its host | Hostname mismatch, or global.subdomain left empty (renders .glyphapp.io) | kubectl get gateway -n istio-system -o yaml |
| Browser reports service workers unavailable | The site is not served over HTTPS | Confirm you reach it on HTTPS with a trusted certificate |
| Nobody can sign in on a fresh install | Initial setup has not been completed | Contact support |
| Invitation or password reset emails never arrive | Outbound email has not been arranged | See Email |
| Meetings fail to connect | Placeholder Vonage credentials | Replace them with a real Vonage Video API application |
| Pods cannot reach each other | A NetworkPolicy or mesh policy is blocking traffic | Review policies in the application namespace |
For problems end users report, such as sign-in or sync issues, see Troubleshooting.
What support can help with
- Chart installation, upgrade, and configuration questions.
- Image tags, initial setup, and how GLYPH behaves once deployed.
- Requests for the security report and SBOMs.
- Platform bugs in the chart or the GLYPH images.
Problems inside your cluster platform, such as the CNI, storage, DNS, or Istio itself, are usually best raised with the team or vendor that runs them. Send us what you have tried and we will help work out where the problem sits.