Control
The one machine running the scheduler, this UI, Postgres and control-placed runs: its CPU, memory, disk and network. The EC2 instances orca launches for runs are on Fleet; Watchdog lists the rules that pause pipelines.
Containers
| ContainerA Docker container on the control host. | Memory nowThe container's memory now, as docker stats counts it. | Peak in rangeThe most memory the container held over the range shown. |
|---|---|---|
| scheduler | 688 MB | 7.7 GB |
| postgres | 169 MB | 197 MB |
| console | 153 MB | 164 MB |
| cloudflared | 21 MB | 39 MB |
| ui | 19 MB | 45 MB |
| sampler | 14 MB | 14 MB |
Range: 24h 7d 14d (UTC, every 60 minutes; a point with no samples is 0)
CPU
Percent of all cores, user and system.
average peak %
Memory
Percent of memory in use (not counting the page cache).
average peak %
Disk
Percent used of the file system under /var/lib/orca, which holds Postgres' data and the staged logs.
fullest disk %
Network
Throughput over all interfaces but loopback.
in (MB/s) out (MB/s) MB/s
Container memory
Each container’s highest memory in the point, as docker stats counts it.
angry_dhawan cloudflared console postgres sampler scheduler ui MB
Alerts
Alerts go to an ntfy topic (not shown: anyone who knows it can read it); the same alert is sent at most once every 30 minutes. orca notify test sends a test message.
| WhenWhen it happened. | KindHow the step runs: mapped, once per partition, or gather, once over all of them. Hover a kind for more. | SubjectWhat the alert is about: a run, a pipeline, a worker or a host. | PriorityThe alert's priority on the phone: high, default or low. | MessageWhat the alert said. | StatusWhere it stands. Hover a status for what it means. |
|---|---|---|---|---|---|
| 5 h ago | watchdog | failing | low | orca watchdog: failing resolved | resolved |
| 5 h ago | run-failed | 1120/4 | default | orca: mojo_activity run 1120 failed | sent |
| 6 h ago | run-failed | 1117/4 | default | orca: mojo_activity run 1117 failed | sent |