Skip to content

Automated Sync from main to stable - #40

Open
rhods-devops-app[bot] wants to merge 650 commits into
stablefrom
main
Open

rhods-devops-app[bot] wants to merge 650 commits into
stablefrom
main

Conversation

@rhods-devops-app

@rhods-devops-app rhods-devops-app Bot commented Oct 24, 2026 •

Copy link
Copy Markdown

Automated Sync from main to stable

This PR automatically syncs the main branch to the stable branch by opening a pull request from main into stable.

Sync Summary

  • Latest commit: b7a5c6734 Merge remote-tracking branch 'upstream/main'

  • Total commits to sync: 639

  • Source: https://github.com/red-hat-data-services/OpenShell.git @ main

  • Target: https://github.com/red-hat-data-services/OpenShell.git @ stable

  • PR head: main

Commits to be synced

Merging

GitHub automerge is enabled for this pull request once required checks pass.

Comment thread .github/workflows/codex-security.yml Dismissed
Comment thread .github/workflows/release.yml Fixed
Comment thread .github/workflows/snap-publish.yml Dismissed
@rhods-devops-app
rhods-devops-app Bot enabled auto-merge October 24, 2026 01:13
@coderabbitai

coderabbitai Bot commented Oct 24, 2026 •

Copy link
Copy Markdown

Important

Review skipped

Bot user detected.

To trigger a single review, invoke the @coderabbitai review command.

Run configuration

  • Source mode: Auto
  • Repository: Unavailable
  • Branch: Unavailable
  • File: No configuration file from this repository applied
  • Commit: Unavailable
  • Layers: Organization UI
  • Review profile: CHILL
  • Plan: Advanced
  • Run ID: 30d83377-7c9d-4b92-9a91-d33f2354ea9c

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

  • Trigger review

Comment thread .github/workflows/release-auto-tag.yml Dismissed
Comment thread .github/workflows/release-prerelease.yml Dismissed
Comment thread .github/workflows/release.yml Dismissed
matthewgrossman and others added 3 commits November 3, 2026 23:14
get.k3s.io serves the latest install.sh from k3s master, so the pinned
checksum broke when upstream changed the script. Download install.sh from
the release tag matching k3s_version instead and pin its checksum.

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>
…yed agent traces (NVIDIA#5153)

The observability overview had no picture of which component exports
which signal to which collector, agent traces appeared only in the
gateway configuration reference, and the Traces page did not say that
the agent's own spans travel separately.

Add an OTLP Collectors section to the overview with a diagram, drawn in
the style of the architecture images, of every lane: gateway and driver
spans, supervisor spans and OCSF events, and agent traces relayed
through the supervisor. Summarize the per-collector protocol, header,
and TLS settings, add an Agent Traces section with an MLflow example,
and link to the overview from the Traces page and the configuration
reference.

Signed-off-by: Kris Hicks <khicks@nvidia.com>
…#4863)

* docs(review): propose area and topic reviewer registry

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* docs(review): keep registry focused on areas and rosters

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* docs(review): match roster tables to maintainer format

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* feat(review): add offline reviewer registry parser

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* feat(reviewers): reconcile area and topic review requests

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* fix(reviewers): derive request state from PR metadata

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* refactor(reviewers): simplify roster parsing and reconciliation

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* ci(reviewers): route reviews on readiness and label changes

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* docs(reviewers): expand initial area and topic nominations

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* chore(reviewers): protect registry and routing with maintainer ownership

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* docs(reviewers): incorporate roster review feedback

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* style(reviewers): format reviewer routing scripts

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* docs(reviewers): add observability reviewers

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* docs(reviewers): add extensions topic roster

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* fix(reviewers): validate paginated API responses

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* docs(reviewers): incorporate community roster requests

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

* docs(reviewers): alphabetize roster members by first name

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>

---------

Signed-off-by: Jim Meyer <jimeyer@nvidia.com>
FrostGod and others added 4 commits November 4, 2026 02:16
NVIDIA#5052)

* feat(server): place pool sandbox sessions with their supervisor's replica

A shared supervisor pool sandbox's session now goes to the gateway
replica serving its owning member's stream, instead of the replica the
ring picks by sandbox ID. Each pool reconciliation pass publishes the
sandbox-to-replica map; sessions fall back to the ring when the parent
is unknown, not a live peer, or this replica is shutting down.

Signed-off-by: divesh <dgude@nvidia.com>

* feat(server): move live pool sessions to their supervisor's replica

A pool sandbox session served by a replica other than the one serving its owning member is redirected there on its next heartbeat, reusing the shutdown handoff so the sandbox stays Ready. This re-homes sessions left on the wrong replica by a rolling restart.

Signed-off-by: divesh <dgude@nvidia.com>

* refactor(server): drop gateway-side pool session placement

Removes the per-sandbox parent map and the heartbeat relocation; pool sessions will dial their supervisor's gateway directly instead.

Signed-off-by: divesh <dgude@nvidia.com>

* feat(supervisor): dial pool sandbox sessions to the host's gateway replica

HostAccepted now carries the accepting replica's peer endpoint. A shared host's sandbox sessions dial it directly, so each session runs on the replica serving its supervisor. A stopping replica sheds its host streams, and redirected sessions follow the host to its next replica. Closes NVIDIA#5051.

Signed-off-by: divesh <dgude@nvidia.com>

* test(e2e): keep pool sessions on their supervisor's gateway through failover

Add the shared-supervisor-tree lane: two gateway replicas on PostgreSQL with a
two-member supervisor pool. After start, a gateway crash, a gateway rolling
restart, and a supervisor crash, every pool sandbox's session runs on the
replica serving its supervisor and exec reaches every sandbox.

Signed-off-by: divesh <dgude@nvidia.com>

* fix(supervisor): move established pool sessions when their host changes gateway

A host whose stream ends, for example when its supervisor token expires, can
reattach to another replica. Its sandbox sessions now watch for that and
reconnect to the host's new replica instead of staying on the old one.

Signed-off-by: divesh <dgude@nvidia.com>

* test(e2e): fold the supervisor tree checks into the shared-supervisor-ha lane

Run the shared-supervisor-ha lane with two gateway replicas on PostgreSQL and
move the gateway crash, rolling restart, and supervisor crash colocation test
into it, replacing the separate shared-supervisor-tree lane.

Signed-off-by: divesh <dgude@nvidia.com>

* fix(supervisor): redial the accepting gateway replica after a host stream ends

A host's stream ends routinely, for example when its supervisor token
expires. Redialing the replica that accepted it, before falling back to the
Service, keeps the host and its sandbox sessions on that replica instead of
moving them all on every reconnect.

Signed-off-by: divesh <dgude@nvidia.com>

* fix(supervisor): follow a host move that lands during session setup

An established session marked its host's current gateway as seen before
watching it, so a move that landed while the session was connecting was
never followed. Compare the current gateway first, then wait for changes.

Signed-off-by: divesh <dgude@nvidia.com>

* fix(supervisor): dial the configured gateway at once when the host's replica has no peer endpoint

Validate the advertised peer endpoint scheme with the shared supervisor, and document that pool sessions follow their supervisor's gateway.

Signed-off-by: divesh <dgude@nvidia.com>

* test(e2e): harden the supervisor tree check's port-forward and time limit

Signed-off-by: divesh <dgude@nvidia.com>

---------

Signed-off-by: divesh <dgude@nvidia.com>
…VIDIA#5221)

The root allow-list was implemented twice: open_root_allowlist and
RootEntryHandle in linux/landlock.rs were documented as the audited
primitive but had no production caller, while the capability-free
baseline enumerates the root itself in sandbox/linux/landlock.rs. Remove
the unused helper, its handle type, their tests, and the dead imports,
so the audited module describes only what runs. abi_version stays; it
backs probe_availability.

Reword the baseline's private-root comment to match the code: the
descriptor backs only the eager O_DIRECTORY|O_NOFOLLOW validation and is
dropped before enumeration, which excludes the private name and opens
every entry with O_NOFOLLOW.

Closes NVIDIA#5190

Signed-off-by: Emilien Macchi <emacchi@redhat.com>
Signed-off-by: konflux-internal-p02 <170854209+konflux-internal-p02[bot]@users.noreply.github.com>
Co-authored-by: konflux-internal-p02[bot] <170854209+konflux-internal-p02[bot]@users.noreply.github.com>
…ic (NVIDIA#5219)

* test(testkit): add cleanup hooks that run after pass, failure, or panic

Tests can register cleanup with cx.on_cleanup(name, hook). Hooks run last
registered first after the body, each with CLEANUP_TIMEOUT and caught
panics. A failing, panicking, or timed-out hook is reported, the rest still
run, and a test whose body passed fails.

Sandbox tracking now registers a delete-and-wait hook keyed by sandbox
name, so tracking then adopting registers it once and Sandbox::delete
removes it. Leftover sandboxes are deleted one at a time rather than
concurrently.

The Keycloak provider-refresh test registers its provider and profile
deletes right after creating them, instead of deleting them inline after
the body, so cleanup also runs on panic and a failed delete fails the test.

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>

* test(testkit): register a keyed cleanup hook under one lock

Check for and add a keyed hook under one guard, so concurrent tracking of
the same sandbox registers one delete. Rename the double-tracking test and
count the deletes it reports rather than relying on the error text.

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>

* test(keycloak): register provider and profile deletes before creating them

Both deletes succeed when there is nothing to delete, so registering them
first also cleans up a create the gateway applied but the CLI reported as
failed. Show the same order in the on_cleanup docs.

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>

---------

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>
afourniernv and others added 22 commits November 17, 2026 05:21
…NVIDIA#5214)

* feat(observability): correlate HTTP receipts, audit records and traces

Closes NVIDIA#5393

Refs NVIDIA#5210

Signed-off-by: Alex Fournier <afournier@nvidia.com>

* fix(observability): narrow receipt validation and simplify tests

Signed-off-by: Alex Fournier <afournier@nvidia.com>

* refactor(observability): preserve per-control HTTP audit records

Correlate original control records at emission with a request-scoped ID and sequence. Remove primary-record aggregation and broad response-head validation while retaining protected receipts and request tracing.

Signed-off-by: Alex Fournier <afournier@nvidia.com>

* fix(observability): preserve HTTP correlation through relay failures

Avoid parsing queries solely for correlation. Carry request attribution through terminal errors and policy reloads, and clear HTTP scope at WebSocket upgrade.

Preserve per-control audit records while reporting bounded stage outcomes and filtering receipt trailers consistently. Remove JSON receipt additions and the unused correlation setter.

Signed-off-by: Alex Fournier <afournier@nvidia.com>

* fix(observability): finish stage results and isolate correlation fixtures

Signed-off-by: Alex Fournier <afournier@nvidia.com>

* fix(observability): report handled middleware stage failures

Signed-off-by: Alex Fournier <afournier@nvidia.com>

* fix(observability): complete proxy request correlation paths

Signed-off-by: Alex Fournier <afournier@nvidia.com>

---------

Signed-off-by: Alex Fournier <afournier@nvidia.com>
…licas (NVIDIA#5436)

* feat(supervisor): spread shared supervisors evenly across gateway replicas (closes NVIDIA#5434)

Signed-off-by: divesh <dgude@nvidia.com>

* fix(supervisor): redirect only hosts that accept a redirect

Signed-off-by: divesh <dgude@nvidia.com>

* fix(supervisor): count in-flight placements and keep handoff redirects after a redirect

Signed-off-by: divesh <dgude@nvidia.com>

* fix(supervisor): count every host a gateway keeps, including hosts that refuse redirects

Signed-off-by: divesh <dgude@nvidia.com>

* docs(supervisor): document best-effort gateway spreading

Signed-off-by: divesh <dgude@nvidia.com>

---------

Signed-off-by: divesh <dgude@nvidia.com>
* chore(deps): defer chrono advisory pending time API migration

Record RUSTSEC-2026-0543 for chrono 0.4.45. No patched release exists; replacing direct time APIs and transitive Kubernetes, OAuth2, and OCI consumers requires a coordinated migration. Keep all other cargo-deny policies unchanged.

Signed-off-by: Adrien Langou <alangou@nvidia.com>

* chore(deps): record crypto policy blocker for RSA advisory

russh 0.65.2 now accepts patched rsa 0.10.0, but rsa adds a mandatory hmac dependency outside the current wrapper policy. Retain RUSTSEC-2023-0071 until that policy change is reviewed and authorized, then upgrade and validate SSH and OIDC test-fixture consumers.

Signed-off-by: Adrien Langou <alangou@nvidia.com>

---------

Signed-off-by: Adrien Langou <alangou@nvidia.com>
* ci(e2e): build the Branch E2E deb without the VM driver

Add an include-vm-driver input to deb-package.yml (default true) and an
explicit OPENSHELL_DEB_OMIT_VM_DRIVER=1 opt-out in package-deb.sh, so
release builds still fail without the VM driver. Branch E2E builds its
deb without it and no longer waits for build-vm-driver-linux.

Run the deb job on hosted ubuntu-24.04 without the CI container. The
script only stages files and takes the target arch from
OPENSHELL_DEB_ARCH, so both architectures build on amd64.

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>

* ci(e2e): keep the VM driver in every deb

Drop the include-vm-driver input and the OPENSHELL_DEB_OMIT_VM_DRIVER
opt-out so every caller builds the same deb. Most of the delay came from
the deb job's self-hosted runner queue and CI container start, which the
hosted runner already removes.

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>

---------

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>
Signed-off-by: Simon Scatton <sscatton@nvidia.com>
…tions (NVIDIA#5611)

Mutation records for creates, updates, and lifecycle changes named no
resource, so an operator could not tell which sandbox was started, which
template was created, or which workspace member was added. A provider
attach that was already in place, a replayed request_id, and a profile
import that failed validation all looked like ordinary successful changes.

Hook the create, update, and lifecycle RPCs for sandboxes, templates,
services, provider attachments, provider profiles, workspaces, and rootfs
staging to record the workspace and the requested resources before the
call starts, completed with the name and ID of each resource the gateway
produced. Record no_op:unchanged and no_op:already_absent for attach and
detach calls that changed nothing, no_op:replayed for answers served from
a stored request_id result, and rejected:invalid_argument for provider
profile imports and updates that failed validation, including their
replays. A record lists at most 16 resources and counts the rest.

Refs NVIDIA#2911

Signed-off-by: Kris Hicks <khicks@nvidia.com>
…DIA#5635)

* feat(server): authenticate postgres with azure workload identity

Signed-off-by: Taylor Mutch <taylormutch@gmail.com>

* docs(gateway): clarify azure postgres schema grants

Signed-off-by: Taylor Mutch <taylormutch@gmail.com>

* fix(server): reuse busy postgres sessions during entra outages

Signed-off-by: Taylor Mutch <taylormutch@gmail.com>

* docs(server): clarify idle-session polling

Signed-off-by: Taylor Mutch <taylormutch@gmail.com>

* fix(postgres): preserve transport timeout error classification

Signed-off-by: Taylor Mutch <taylormutch@gmail.com>

---------

Signed-off-by: Taylor Mutch <taylormutch@gmail.com>
…rite (NVIDIA#5666)

* fix(compute): retry watch-driven sandbox removal after a concurrent write

A driver deletion event is the only signal that removes a sandbox record
after a driver-confirmed delete. Its removal compared the record version
once and silently gave up when another replica wrote the record in
between, such as the endpoint-status reset when the supervisor session
ends. The record then stayed until the orphan sweep, about five minutes
after creation.

Re-read and retry the removal, and log driver deletion events, retries,
retained records, and Kubernetes sandbox watcher relists.

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>

* fix(compute): clarify watch deletion retry logs and assert driver cleanup

The removal helper returns false for more than a version conflict, so
log the retry neutrally and only before attempts remain. Assert that the
successful retry reclaims driver-owned resources.

The orphan-sweep removal log also gains the sandbox phase, to tell a
stalled delete apart from a sandbox that never reached the driver.

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>

---------

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>
…DIA#5667)

After a retained sandbox's main process exits, CreateSshSession and
ForwardTcp allow access only while the supervisor is connected. The check
looked only at the receiving replica's session map, so on a multi-replica
gateway a request that landed on another replica was refused with
"sandbox is not ready". This broke `sandbox create -- <cmd>` when the
command finished before the CLI attached.

Check the supervisor owner record too, the same way relay opening already
routes to the owning replica. Run the host-gateway client as the sandbox's
main process again, so the k3s-ha row covers the attach.

Signed-off-by: Piotr Mlocek <pmlocek@nvidia.com>
* feat(policy): lint authored policy files offline

Validate authored files before gateway selection or authentication. Share creation safety checks and report file-only violations together while deferring checks that require live gateway state.

Closes NVIDIA#4035

Signed-off-by: Shiju <shiju@nvidia.com>

* fix(policy): preserve gateway size errors and refresh example locks

Signed-off-by: Shiju <shiju@nvidia.com>

* fix(policy): reject oversized input before lint diagnostics

Signed-off-by: Shiju <shiju@nvidia.com>

* fix(policy): synchronize standalone lockfile dependencies

Signed-off-by: Shiju <shiju@nvidia.com>

---------

Signed-off-by: Shiju <shiju@nvidia.com>
moulalis and others added 7 commits November 18, 2026 04:07
* ci(setup-nix): disable HTTP/2 for Nix downloads

A failed HTTP/2 connection fails every NAR multiplexed on it. Nix resumes NARs that had fully arrived at EOF, gets HTTP 416, and fails the substitution (NixOS/nix#16722). Separate HTTP/1.1 connections confine a failure to one transfer.

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>

* ci: add temporary Nix HTTP/2 download experiment

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>

* ci: keep the Nix HTTP/2 experiment running on zero matches

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>

* ci: remove temporary Nix HTTP/2 download experiment

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>

---------

Signed-off-by: Matthew Grossman <mgrossman@nvidia.com>
…5650)

* fix(network): refresh expired policy DNS mappings on connect

Signed-off-by: Drew Newberry <anewberry@nvidia.com>

* fix(network): revalidate DNS pins and preserve refresh eligibility

Signed-off-by: Drew Newberry <anewberry@nvidia.com>

* refactor(network): centralize policy DNS connect lookups

Signed-off-by: Drew Newberry <anewberry@nvidia.com>

* test(network): consolidate DNS refresh scenarios

Signed-off-by: Drew Newberry <anewberry@nvidia.com>

---------

Signed-off-by: Drew Newberry <anewberry@nvidia.com>
Signed-off-by: Taylor Mutch <taylormutch@gmail.com>
…ool lock holders (NVIDIA#4696)

* fix(server): bound data-pool lock holders

The SSH identity and provider-refresh locks each hold their PostgreSQL
advisory lock on a data-pool connection, and their holders then query
the same pool. At a small database_max_connections, a few holders could
take every data connection and wait for one more until sqlx's 30-second
acquire timeout, stalling every database request on the replica.

Each PostgresStore now has max(1, max_connections / 2) data-pool lock
slots. A holder takes a slot before its connection and keeps it until
that connection has closed, so every holder can still get a second
connection for its queries. A slot that stays busy for the 10-second
mutation lock timeout fails with a lock timeout and a warning. The
floor rationale in the config docs, man page and startup error now names
both holders, and mutation_lock.rs documents the slot-sharing cycle the
timeouts break at small pools.

Part of NVIDIA#4307.

Signed-off-by: Emilien Macchi <emacchi@redhat.com>

* fix(server): scope the SSH identity lock per sandbox

SSH host identity provisioning and cleanup took one fleet-wide advisory
key and a process-wide mutex, so create, start, restart and delete ran
that step for one sandbox at a time across the fleet, with each waiting
create holding a lock-pool connection.

The SSH identity guard now takes the legacy key in shared mode and then
an exclusive key derived from the sandbox id, on one data-pool
connection and one data-pool lock slot. The local mutex is keyed by
sandbox id too. Steps for different sandboxes run at once, steps for the
same sandbox still exclude each other on every replica, and gateways
from earlier releases, which hold the legacy key exclusively, keep
excluding new ones during a rolling upgrade.

One 10-second deadline now covers a data-pool lock holder's slot, its
connection and every lock statement, as it already did for mutation
locks, so the second lock statement doesn't extend the wait under
create's mutation guard. A lock that times out returns a lock timeout.

Part of NVIDIA#4307.

Closes NVIDIA#4686

Signed-off-by: Emilien Macchi <emacchi@redhat.com>

* fix(server): release the create guard before the SSH identity step

Sandbox creation held its PostgreSQL mutation guard while it waited for the
SSH identity keys, so each replica admitted at most one in-flight create per
lock-pool connection. The per-sandbox SSH identity lock already fences a
concurrent delete, so release the guard before that step.

Data-pool lock connections were closed on every release, which opened a new
PostgreSQL session per create. Unlock and reset an idle session and return it
to the pool; close it only when a lock statement may still be waiting.

Signed-off-by: divesh <dgude@nvidia.com>

---------

Signed-off-by: Emilien Macchi <emacchi@redhat.com>
Signed-off-by: divesh <dgude@nvidia.com>
Co-authored-by: divesh <dgude@nvidia.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.