Troubleshooting and acceptance
Diagnostic steps by symptom and partner acceptance checks.
Diagnose by symptom
| Symptom | Check |
|---|---|
| Reader does not appear | Check time, DNS, outbound 443, enrollment key permission and account capacity. Identify readers by UUID, not name alone. |
| Private state directory is unavailable or already in use | Check state path, ownership and 0700 permissions. Check for simultaneous standalone and SDK processes. |
| Private enrollment configuration, entropy or identity failed | Check key/seed/identity file ownership, 0600 modes and contents. Do not substitute another reader’s state. |
| Invalid configuration, CA bundle or entropy source | Check the shipped CA bundle, seed and engine configuration. Keep certificate verification enabled. |
| Engine exits after successful start | Inspect poll exit status and stderr. start only confirms spawn. |
| Command is pending | Check online state, owner account and approximately 30-second queue interval. |
| Terminal already in use | End the existing session normally. Do not take over another operator’s session. |
| Dashboard button missing | Check online state and ask your contact to verify dashboard enablement. |
| Dashboard 401/expired | Create a new connection using the StreamZet button. Use OEM credentials for an OEM authentication prompt. |
| Dashboard connection error | Check that the reader's OEM port 9000 service is running, then open it again from StreamZet. |
Foreground diagnostics
start.sh discards output. If diagnostics are needed, first stop StreamZet and confirm exit. Then run the command below in an authorized reader shell to see stderr. Stop with Ctrl+C and restore the normal startup method.
sh
cd /mnt/spp/streamzet-jci
./streamzet-embedded --production \
streamzet-embedded.plitsoft.workers.dev \
/mnt/spp/streamzet-jci/registration.key \
/mnt/spp/streamzet-jci/cacert.pem \
/mnt/spp/streamzet-jci/state 'xSpan R660 001'Partner acceptance checklist
- Does first startup enroll the reader and make it visible to the operator account?
- Have terminal output and successful/failed Fleet results been verified on the physical reader?
- Does the UUID survive app/engine restart, and does connectivity recover after a network outage?
- Do SDK start, poll, graceful stop and child reaping work inside the OEM application?
- During an approved maintenance window, has startup been verified after a full reboot and CAP/firmware update?
- Have CPU, memory, RFID workload and batch execution across multiple physical readers been tested?
- Have dashboard enablement, login, live events and simultaneous terminal use been verified?
When requesting help, provide SDK version, reader model/firmware, UUID, timestamp with timezone, reproduction steps and redacted exit status/logs. Do not send keys, passwords or state/identity.