On Boarding Steps (Prime High Availability)
Go-live essentials for Prime High Availability. Start with Top customer deliverables, then four scopes.
Top customer deliverables
Ordered by priority — what the customer must provide for go-live:
Prio | Deliverable | Owner (customer) |
|---|---|---|
1 | Two production-capable hosts + Galera plan — IPs, hostnames, SQL cluster design | DevOps |
2 | Network / firewall rules — peer ports for DHCP failover, SSH, rsyslog, monitoring as agreed | Network / Security |
3 | CMTS / access-node HA settings — helper-address, NTP peers for both NMS hosts | Network |
4 | Named HA owners — who may change ProvHA GUI and who runs failover drills | IT / Ops |
5 | Maintenance window for acceptance failover drill | Ops |
Scopes
# | Scope | Goal |
|---|---|---|
1 | Access & enable | Package on both hosts; roles set |
2 | Main decisions | Load ratio, rebuild interval, monitoring HA scope |
3 | Initial data / peer config | Failover.conf, GUI peers, SSH sync |
4 | Acceptance | Failover drill signed |
1. Access & enable
NMS Prime brings | Customer brings |
|---|---|
Package | Two hosts, Galera, root/SSH access |
ProvHA GUI | Firewall change windows |
Decide: who is operational owner of master vs slave?
2. Main decisions
NMS Prime brings | Customer brings |
|---|---|
Recommended defaults (rebuild interval, load ratio) | Target load split master/slave |
Install guide for Cacti/Icinga HA options | Whether Cacti/Icinga HA is in go-live scope |
Decide:
3. Initial peer config
NMS Prime brings | Customer brings |
|---|---|
Official Documentation + nested Installation | Confirmed IPs, secrets, CMTS statements |
Sync / cleanup commands | Owner who verifies journal and sync |
Decide: who runs first production sync and who signs peer health?
4. Acceptance
NMS Prime brings | Customer brings |
|---|---|
DevOps + provisioning testers |
Decide: which checklist rows are release-blocking? (Recommend: #1–#5, #8.)