Elamur: **Mova** · [Parse](https://mova.elamur.ai/free/en/parse/) · [Dictionary](https://md.dictionary.elamur.ai/free/en/) · [Translate](https://md.translate.elamur.ai/free/en/) · [Research](https://md.research.elamur.ai/free/en/) · [MCP](https://mcp.elamur.ai/) · [Groupants](https://md.groupants.elamur.ai/free/en/)

[Mova](https://md.mova.elamur.ai/free/en/) · [this page in HTML](https://mova.elamur.ai/free/en/docs/tools/jobs/)

[home](https://md.mova.elamur.ai/free/en/) / [docs](https://md.mova.elamur.ai/free/en/docs/) / [tools](https://md.mova.elamur.ai/free/en/docs/tools/) / jobs — a heavy-job runner with limits and a log

# jobs — a heavy-job runner with limits and a log

`jobs` runs one command, samples the whole process tree every second (peak RSS and CPU time from `/proc`), stops only that job when it passes its memory or time limit, and logs estimate against actual.

## Run

```
cargo build --release -p jobs
target/release/jobs run --name N [--mem-gb G] [--secs S] [--est-mem-gb E] [--est-secs T] -- command args…
target/release/jobs log
```

| Flag | Default | Meaning |
| --- | --- | --- |
| `--name` | the command | job name in the log |
| `--mem-gb` | 16 | stop when the tree's RSS exceeds it |
| `--secs` | 7200 | stop when wall time exceeds it |
| `--est-mem-gb`, `--est-secs` | — | your estimate, logged next to the actual values |

A stop sends SIGTERM to the tree, then SIGKILL after 5 s; the exit code is then -9. Otherwise `jobs` exits with the command's code.

`jobs log` prints jobs.jsonl: one JSON line per job with `name`, `cmd`, `est_mem_gb`, `est_secs`, `peak_mem_gb`, `cpu_secs`, `wall_secs`, `code`, `reason`.

## Limits

Linux only (`/proc`). Sampling is once per second, so a short spike can pass unseen. `$MOVA_DATA` defaults to `~/mova-data`.

- [← previous hbo](https://md.mova.elamur.ai/free/en/docs/tools/hbo/)
- [next → kaggle — teacher models](https://md.mova.elamur.ai/free/en/docs/tools/kaggle/)
- [↑ back to tools](https://md.mova.elamur.ai/free/en/docs/tools/)