Trends
Capacity forecasts, recurring failure signatures, and fleet-wide correlation.
Capacity forecast
| System | Disk | Memory | Process growth | Changes |
|---|---|---|---|---|
Ubuntu Workstation 60 snapshots | 8d to 95% · high confidence | No sustained growth | worker.service +384 MB | 1 package events |
CI Runner 02 60 snapshots | 8d to 95% · high confidence | No sustained growth | worker.service +384 MB | 1 package events |
Recurring incidents
OOM killer invoked PyTorch training job (PID 38291) consuming 62 GB of 64 GB RAM on Ubuntu, triggered by an unconstrained local training process.
2 events5/12/2026 → 5/20/2026
Fleet correlation
OOM killer invoked PyTorch training job (PID 38291) consuming 62 GB of 64 GB RAM on Ubuntu, triggered by an unconstrained local training process.
2 systemsFingerprint 4be91a6d120f0c82
Healthy versus failed
CPU
Healthy
32.5%
At failure
98%
Memory
Healthy
35%
At failure
99.9%
Disk
Healthy
73.1%
At failure
73%
Free disk
Healthy
264GB
At failure
264GB