Repository navigation
RunR 0.6.10
RunR is an open-source, self-hosted serverless platform on Kubernetes and Knative: functions, containers, sandboxes, WebAssembly and durable workflows, with an API, a web console, the runr CLI, SDKs, a Terraform provider and a GitHub Action. You install it on infrastructure you operate, connected or air-gapped. It is developed by BRF Teknoloji and released under the MIT License.
0.6.10 is a security release of the internal network (docs/internal-network.md). From 0.6.0 to 0.6.9, a change of the internal network that the region's API server did not take (refused, failed, not reached) was saved and lost until the next deploy. A project in access mode allowlist could lose the IP allow list of its public host, which then answered every client. A caller removed from a project's allow list, unbound from a model service or deleted with its project, and a later project of the deleted one's name, could stay admitted. 0.6.10 tries such writes again, removes a project's routes only when its Knative Service is gone, takes a deleted or renamed project off the allow lists and model services, and its region Job puts lost callers right.
Upgrade every installation from 0.6.0 on. No migration, no new value: the API changes only. After the upgrade, check that every project in access mode allowlist has its routes (below).
Install
- Kubernetes (Helm): the chart comes from BRF's public Helm repository, without credentials:
See the README and docs/install.md;
helm repo add brf-tech https://helm.brf.sh helm upgrade --install runr brf-tech/runr --namespace runr --create-namespace --set domain=runr.example.com
helm show values brf-tech/runrlists every value. A pinned install adds--version 0.6.10; the API and UI images, the region agent and the event gateway follow the chart version. - Single host (Docker Compose, also air-gapped): compose/README.md, from a checkout of the release tag (
git clone -b v0.6.10 https://github.com/BRF-Tech/runr.git). - CLI:
curl -fsSL https://runr.sh/install.sh | shon Linux and macOS. The archives of every platform are at runr.sh/dl (https://runr.sh/dl/cli/v0.6.10/) and attached below; check them againstchecksums.txt(docs/cli.md#install). - Terraform / OpenTofu:
source = "brf-tech/runr",version = "~> 0.6", from the Terraform Registry;terraform initchecks the archive againstSHA256SUMS, signed with the RunR release key (key ID2588846B241BC095). runr.sh serves the same archives underrunr.sh/brf-tech/runr, for OpenTofu and for configurations that use that address (docs/terraform.md#install). - GitHub Action:
uses: brf-tech/runr/actions/deploy@v0.6.10; it downloads the CLI from runr.sh (docs/github-action.md). - Images: every image is published under
ghcr.io/brf-tech/runrby release-images.yml, for linux/amd64 and linux/arm64, and pulls anonymously.
The documentation is at runr.sh/docs and in the documentation index here.
Upgrading from 0.6.9
Upgrade in place with helm upgrade of the running release; never uninstall and install again, and back up the database first (Procedure in docs/upgrade.md). From 0.6.9 the images keep their names and nothing has to be set (0.6.9 → 0.6.10).
helm repo update brf-tech
helm upgrade runr brf-tech/runr --version 0.6.10 -n runr --reuse-values- What was exposed (0.6.0 to 0.6.9). A change of a project's network that could not read its Knative Service deleted its routes, in access mode
allowlistthe IngressRoutesrunr-access-<service>and-tlsof its public host: it answered every client until the project's next deploy. A removed or deleted caller stayed on the project's allow list in the region until that deploy, and a model service kept admitting a project unbound from it or deleted until the model service's next deploy; a later project of the deleted one's name in the organization was admitted too (What was exposed). - The upgrade puts lost callers right. The region Job
runr-region-reconcile-<revision>(region:reconcile, stepinternal-network) writes a model service's callers and a project's allow list where the region's differ from the database's (What 0.6.10 changes). - A model service's writes are no longer refused for its volume. The pod spec features of the model volume (persistent volume claims) were turned on only by a region's first model service deploy, and Knative's webhook could refuse the service, or a later write to it such as its callers, for tens of seconds after. They are now part of every region's KnativeServing configuration from its install,
region:reconcile(the upgrade's region Job too, stepknative-serving-config) enables them again, and a refusal for a feature enables them before the write is tried again (docs/ai.md). - Check the projects in access mode allowlist. The region Job does not bring back a deleted route of a public host. After the upgrade,
kubectl get ingressroute -A | grep runr-access-lists two routes for every project in access modeallowlist, and its public host answers403to a client outside its ranges; redeploy a project whose routes are missing (After the upgrade). - Until you upgrade: redeploy a project after a change of its allow list or access mode, and a model service after a binding changed or a bound project was deleted or renamed; each deploy writes what a lost change left out.
- The region agent and the event gateway move to 0.6.10. As in 0.6.0 to 0.6.9, after
helm upgradereturned, the Jobrunr-region-reconcile-<revision>moves the regions' agents, queue consumers and topic publishers toghcr.io/brf-tech/runr/agent:0.6.10andqueue-consumer:0.6.10; each rolls once. A region that pulls through a mirror of its own needs these two images there before the upgrade (Order of operations). The same Job writes the model volume's features into each region'sconfig-featuresonce, which restarts nothing; nothing else in the regions moves. - Back to 0.6.9: a
helm rollback; nothing runs on the database, and 0.6.9 loses a write the region does not take again (Back to 0.6.9). - Terraform provider 0.6.10: its resources and data sources are the ones of 0.6.9.
From 0.6.8, upgrade straight to 0.6.10 and read 0.6.8 → 0.6.9 as well: 0.6.9 adds a column (projects.capacity_released_at, and in every tenant schema), a stopped project gives its capacity back in a pay-as-you-go region and is admitted again on resume, and a region with many stopped projects can raise the capacity alarm right after the upgrade. From 0.6.7, also read 0.6.7 → 0.6.8: 0.6.8 adds runtime sources and RunR's signed runtime catalog, which the API reads from ghcr.io (runtimeCatalog.offline: true keeps the bundled one), gives single-tenant Owners platform.manage, moves the region Traefik to chart 42.3.1, and a marketplace that merges the package's default values into an upgrade can bring ingress.tls.mode: letsencrypt to a release that never set it (The console certificate warning). From 0.6.6, also read 0.6.6 → 0.6.7: 0.6.7 changes how sandbox commands travel on Kata and gVisor (a command on gVisor takes about 100 ms more), input that arrives short fails, and a custom sandbox image needs cat and mkfifo. From 0.6.5, also read 0.6.5 → 0.6.6: 0.6.6 re-runs kata-deploy with Kata Containers 4.3.1, which stops the running Kata pods of every region with Kata, the region Traefik moves to chart 42.3.0 and deletes request headers with _ or . in their names, and Reverb refuses client events. From 0.6.4, also read 0.6.4 → 0.6.5: 0.6.5 adds the notification_user_states migration, the Helm value notifications.retentionDays, and its daily task deletes notifications older than 90 days. From 0.6.3, also read 0.6.3 → 0.6.4: 0.6.4 adds the captcha_settings migration and the Helm values captcha.*, and its first upgrade rolls the pods once. From 0.6.2, also read 0.6.2 → 0.6.3; from 0.6.1, also 0.6.1 → 0.6.2. From 0.6.0, also read 0.6.0 → 0.6.1 first: before a rollback to 0.6.0, let the running deploys finish (Rolling back to 0.6.0). From 0.5.10, read 0.5.10 → 0.6.0 in docs/upgrade.md first — in particular what to decide before the upgrade — and the notes of 0.6.0 to 0.6.9 in CHANGELOG.md; everything said there applies. From 0.5.9 or earlier, read the notes of the releases in between as well; from 0.5.8 or earlier, add --set api.image.tag= --set ui.image.tag= to that upgrade once (step 3 of the Procedure).
Changes in 0.6.10 (2026-11-12)
Upgrade notes
- A security release of the internal network: upgrade from 0.6.0 to 0.6.9. On those releases a
project in access modeallowlistcould lose the IP allow list of its public host, which then
answered every client, and a removed, unbound or deleted caller could stay admitted to a project or
a model service, until the next deploy (0.6.9 → 0.6.10 in upgrade.md).
No migration, no new value; the images keep their names. - The region Job of the upgrade puts lost callers right (
region:reconcile, step
internal-network: a model service's callers and a project's allow list where the region's differ
from the database's). It does not bring back a deleted allowlist route: after the upgrade, check
that every project in access modeallowlisthas its IngressRoutesrunr-access-<service>and
-tls, and redeploy one that has not
(After the upgrade). - The same Job writes the model volume's Knative features (persistent volume claims) into each
region'sconfig-featuresonce (region:reconcile, stepknative-serving-config); nothing
restarts (What 0.6.10 changes). - Back to 0.6.9 is a
helm rollback(Back to 0.6.9); nothing runs on
the database.
Fixed
-
A model service's Knative Service is no longer refused for the model volume. The pod spec
features of the model volume (a read-write PVC) were turned on by a region's first model service
deploy, and Knative's webhook went on refusing it ("Persistent volume claim support is
disabled") for tens of seconds while config-features already carried them, and at times again
later: the deploy failed, or a write to the deployed service (its callers on the internal network)
was refused. They are on every region from its install now, and everyregion:reconcileenables
them again (stepknative-serving-config; on a GPU region the GPU features too). A refusal for a
feature enables them again before the write is tried again, a CR that could not be read no
longer hasconfig-featurespatched behind the Knative Operator, and a refused allow list of a
service on the internal network no longer fails the region's internal-network change
(test-internal-network'sin-cross-orgfailed so). -
A project bound to a model service is admitted by it, also when the region did not take the
binding's write. The internal network's authz reads a model service's callers from its Knative
Service, where a binding added or removed was written once: a refusal or an error of the region's
API server, or a failed read before it (taken for "not deployed"), lost it without a word. The
bound project was refused (403) until the model service's next deploy, and an unbound one kept
being admitted. The write is now tried again until the region takes it (for about twelve minutes;
thenregion:reconcile, whoseinternal-networkstep now also writes a model service's callers
when they are not its bindings). A binding made while the model service deploys is written after
the deploy's apply as well.test-internal-network'sin-modelfailed so on GitHub-hosted
runners: in-api was still refused six minutes after its binding. -
A deleted project leaves the model services it was bound to, and a renamed one is its new name
there. Its bindings went with its row without a word to the region: the model service kept
admitting its old name, so a later project of that name in the organization was admitted
without a binding, and a renamed bound project was refused (403) under its new name. Both reach
the model service at once now, tried again like a binding. -
A change of a project's allow list or access mode that the region did not take is tried
again. It was saved and answered with awarning, and waited for the project's next deploy:
meanwhile a removed caller, or one deleted with its project (a later project of its name too),
was still admitted. The change is now tried again for about twelve minutes, and
region:reconcilewrites a project's allow list when its Knative Service has another. -
A failed read of a project's Knative Service no longer removes its routes. A change of its
network took the failed read for a project without a Knative Service and deleted its
runr.internalroute and, in access modeallowlist, the IP allow-list route of its public
host, which then answered every client until the next deploy. Only a missing Knative Service
removes them now; anything else is a change the region did not take (above).
Security
- The dependency bot's secrets are secrets of the GitHub environment
release, which onlymain
deploys to. Its jobsbumpandreleaseare the only ones in it, so a workflow pushed to another
branch with a stolen account cannot read the origin's deploy key, the GitHub App's key or the
bot's signing key. A fork that runs the bot creates the environment and sets the secrets there
(docs/development.md). The dependency audits' workflow is
dependency-audit.yml("Dependency audit"), no longeraudit.yml.