Commit Graph

6 Commits

Author SHA1 Message Date
Ruben Fiszel b656dc6cdc feat(nsjail): optional disk-backed /tmp via instance setting (#9272)
* feat(nsjail): optional disk-backed /tmp via instance setting

* test(nsjail): unit-test tmp mount resolver and narrow visibility

* refactor(nsjail): switch tmp backing to select + conditional UI

* ui(nsjail): make tmpfs the visible default in /tmp backing select

* fix(nsjail): refuse preexisting jail_tmp to block symlink escape

* fix(nsjail): allow jail_tmp reuse on sequential nsjail calls

Codex flagged that python/ruby/rust executors invoke nsjail twice per
job_dir (install then run). The previous resolver treated any preexisting
jail_tmp as hostile and silently fell back to tmpfs on the second call,
so disk-backed mode never reached the main script run for those langs.

Use symlink_metadata().is_dir() to distinguish a real directory left by
an earlier call in the same job_dir (safe to reuse) from a symlink or
other entity (still refused, as the codebase-tar escape requires).

Also loosen the frontend visibility predicate: only hide nsjail settings
when job_isolation is explicitly 'none' or 'unshare', so deployments
that enable nsjail via DISABLE_NSJAIL=false with no DB setting can
still see the controls.
2026-05-21 15:34:49 +00:00
Ruben Fiszel 9111f8908d feat(nsjail): make tmpfs size configurable via instance setting (#9261)
* feat(nsjail): make tmpfs size configurable via instance setting

Adds a new `nsjail_tmpfs_size_mb` instance setting that overrides the
size of the `/tmp` tmpfs mount inside the nsjail sandbox across all
languages. When unset, the existing per-language defaults (500MB or
800MB) continue to apply, so no behavior change for existing
deployments.

The setting is exposed under Settings → Jobs and is read at job
execution time, so changes take effect on the next job without a
restart.

Fixes WIN-1963

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* refactor(nsjail): unify default tmpfs size to 800MB

Previously each executor passed its own per-language default (500MB or
800MB) to resolve_nsjail_tmpfs_size. Unify on a single
DEFAULT_NSJAIL_TMPFS_SIZE_BYTES constant (800MB) so the placeholder
behavior is consistent across languages.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(nsjail): resolve tmpfs size outside ruby download closure

The download.ruby config render runs inside a sync closure passed to
par_install_language_dependencies_seq, so `.await` on
resolve_nsjail_tmpfs_size() was a compile error under the `ruby`
feature. Resolve the size once before the closure and capture the
string instead.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* docs(nsjail): rename resolver to *_bytes and clarify fallback

Addresses CI review feedback:
- Rename `resolve_nsjail_tmpfs_size` to `resolve_nsjail_tmpfs_size_bytes`
  so the returned unit is unambiguous at the call site (cubic P2).
- Fix the `NSJAIL_TMPFS_SIZE_MB` doc comment that still said "per-language
  default" — there is no per-language fallback anymore, all unset
  values resolve to the unified 800MB `DEFAULT_NSJAIL_TMPFS_SIZE_BYTES`
  (codex/pi P2).
- Expand the resolver doc to call out that `Some(0)` and negative values
  also fall back, since the match arm is `Some(mb) if mb > 0`.

No behavior change.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-20 13:24:49 +00:00
Ruben Fiszel f8467f38c8 fix: prevent cross-tenant DNS poisoning via writable /etc in nsjail (#9194)
* fix: bind /etc resolver files read-only in nsjail sandboxes

* docs(nsjail): explain why per-file /etc resolver binds are load-bearing

The explicit /etc/hosts, /etc/resolv.conf and /etc/hostname binds look
like removable duplication of the read-only /etc bind above them. They
are not: on Kubernetes those files are separate kubelet bind-mounts on
top of /etc and nsjail's read-only remount is non-recursive, so without
these shadow binds they stay writable and a job can persist cross-tenant
DNS poisoning for the pod lifetime. Comment guards against a future
"dedup cleanup" silently reintroducing the vulnerability.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* docs(nsjail): shorten the load-bearing-bind comment to 3 lines

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude <noreply@anthropic.com>
2026-05-17 12:54:38 +00:00
hugocasa 65a92d9899 fix: set nsjail time_limit from job timeout so configured defaults are respected (#8389)
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-16 12:06:31 +00:00
Ruben Fiszel e7e27259ee nsjail skip_setsid for pg propagation + faster exit 2025-07-13 11:48:01 +00:00
pyranota 59740c0478 feat: add java support (#5458)
* feat: add nu (nushell) support

* add worker tests

* deactivate tables and non-any types below top-level

full support will come in V1
for V0 it's better to keep things minimal and simple

* add syntax highlighting

used python's grammar, since nushell isn't supported by monaco nor svelte-highlights

for V1 nu will get it`s own grammar

* add logo

* partially implement plugin support

* change logo + ability to deploy + nsjail draft

* static variables + get_resource + get_variable

* lsp/dev.nu + initial nu lsp (not working yet)

* make it work with nsjail

* nullguard

* Much more flexible signature parsing and better error-messages

* add init script

* rename nulsp to nu

* install nu to dockerfile

* fix merge

* implement Default for MainArgSignature

* stage NU_CACHE_DIR

* improve dockerfiles

* dev.nu for parser-wasm + flake.nix

* update code for windows

* add nushell to flake

* upload Cargo.lock

* make build.sh work on nixos

* build wasm cli parsers

* add docs to README_DEV.md

* add helper script docker/dev.nu

* improve docker/dev.nu

* fix windows

* commit frontend/package(lock).json

* update cargo.lock

* correctly update cargo.lock

* remove lsp

* update flake.nix to include svelte server and nushell

* Revert base.sql to main

* remove PLUGIN_USE_RE

* make CARGO_PATH private

* add nu to cli

* Change flags to build wasm-nu-parser

* remove flake.nix from parser-wasm

* update wasm-build target

* remove unused import

* add cli support for nu

* update github workflows

* wasm-build 0.17 -> 0.19

* update build script

* update cargo.lock

* Fix typographical error

* start working on java

* do java boilerplate

* implement parser for java

* update Cargo.lock

* update ENV_SETTINGS

* use published nu parser

* update package.lock

* java is S3 + Caching enabled

* install nsjail backup

* commit v0

* fix nsjail

* v0.1

* rewrite parser in tree-sitter

* implement parser from scratch

* polishing

* change init script to match new parser

* fix imports

* fix cli build

* fix cli build

* refactor install phase

* implement .valid.windmill atomic verification

* implement java init functionality

* remove quick-xml

* fix windows not recognizing 'mvn'

* create empty settings.xml if there is no config provided

* clean up

* change default settings.xml

* change classpath format for windows

* docs to helper

* java copy bin cache instead of symlink

* remove comments

* merge

* fix package.json

* fix package.json 2

* minor fixing

* migrate to Coursier

* update misc

* Http(s) Proxy + CA certs

* remove unused .wasm

* make requirements insensitive to spaces

* update handle_child refs

* rework save_cache for directories

* fix s3 bug

* compile .wasm for cli

* remove uuid import

* fix compilation

* use reference

* fix zero-dep failure

* removing unsafe stuff

* remove unneeded imports

* revert: we still need winapi

* remove nix store from nsjail

* do not create cache_nomount

* add java to dnt

* remove duplicated dependency in init script

* fix typos

* fix CI

* use published parser
2025-03-27 15:12:55 +01:00