Server room power supply failure — three inverter units stopped output (fault F56 DC_VoltLow_Fault)
Affected components
Handling stages and targets
- Investigating30 Jul 2026, 00:46 WITA
- Identified30 Jul 2026, 14:40 WITA
- Monitoringnot recorded
- Resolved30 Jul 2026, 07:27 WITA
| Initial response | 13 h 54 min | target ≤ 15 min | Missed |
| Recovery | 6 h 41 min | target ≤ 4 h | Missed |
Initial response = the first update from the NOC after the outage was detected. An outage that recovered before the initial-response limit counts as handled. Targets follow severity: Critical ≤ 15 min/≤ 4 h, Major ≤ 30 min/≤ 8 h, Minor ≤ 1 h/≤ 24 h.
Timeline
- Identified4 months ago
On the night of Wednesday, 29 July 2026, AC power from the PLN grid stopped reaching the three inverters feeding the server room and network equipment. The entire critical load switched over to the LiFePO4 battery bank without any notification reaching the NOC team.
After roughly 7.5 hours running on battery alone, the stored energy ran out. All three inverters entered fault F56 (DC_VoltLow_Fault) and cut their output simultaneously in the early hours of 30 July 2026, taking down every network and server service. Because fault F56 latches and grid battery charging was not running, the system could not recover on its own until manual intervention that morning.
Service was fully restored at 07:20 WITA, with a total service downtime of 5 hours 42 minutes. The root cause was identified as loss of AC supply upstream of the inverters — strongly suspected to be the SBW-80kVA stabiliser unit, which is a single point of failure for all three inverters — made worse by the absence of an early-warning system and by grid battery charging being inactive. - Resolved4 months ago
Done