Google conducts extensive monitoring of all GGC nodes and the GGC Support Team are automatically notified of any faults materially affecting the performance of a GGC node, or of individual GGC machines.
Fault tickets raised by GGC Program Partners or contracted third parties may:
- duplicate alerts already received by the GGC Support Team, which they will attempt to fix remotely without notifying the GGC Program Partner about the fault
- duplicate fault tickets that have already been sent by the GGC Support Team and are waiting for attention by the GGC Program Partner
- refer to problems serving a user prefix that are already indicated on the BGP page
- refer to network performance problems which are already indicated on the Performance page
- refer to users QoE problems that are already implied in the re-transmits per prefix data
To avoid duplicated effort, before creating a new fault or problem ticket GGC Program Partners should:
- ensure that anyone monitoring GGC has access to the ISP Portal
- ensure that there is not already a ticket for the fault or problem they have observed
- ensure that there are no relevant open fault tickets waiting for their response
- ensure that any relevant ISP Portal data has already been reviewed
A dashboard showing the health of Peering links and GGC nodes is available at Asset Health. This dashboard also has links to any open fault tickets for a particular GGC node.
All GGC machines receive regular automated software maintenance, which may result in the machine being restarted. If disk checks are also required, it may be more than an hour before the machine is back online.