monky-deployd

mdella/monky-deployd

Fork 0

mirror of https://scm.tikali.ai/tikali/applications/monky/monky-deployd.git synced 2026-09-18 06:16:16 +00:00

Commit Graph

Author	SHA1	Message	Date
mdella	43615a6fda	fix(onboarding): identity read that survives a rewrite, disk refused before the pull — 0.1.8 Three faults from one agent-managed onboarding (env-dev-08, 2026-09-09), each of which pointed the diagnosis away from the actual fault. 1. install.sh granted the agent's read on the ziti identity with a POSIX ACL. ziti-edge-tunnel rewrites that file on a controller config update and the rewrite drops the ACL: the agent applied cleanly at 01:21 and was failing every tick by 01:32. Group membership survives the rewrite (the file stays ziti:ziti 0640), so install.sh and the package postinstall now add monky-deployd to the `ziti` group, and a default ACL on the identity directory carries the grant onto a newly created file. The explicit ACLs stay for the boxes that need them. 2. openziti.load() accepts an unreadable or malformed identity: the C SDK logs "configuration is invalid" and returns a context that only fails at dial, as a bare TypeError, which the transport reported as a missing intercept or a policy gap. The SDK transport now reads and parses the identity itself and names the real fault first. 3. The disk pre-flight ran only when the bundle declared disk_need_bytes, so a bundle without one died mid-pull with containerd's "no space left on device" — which reads as a registry fault. A bundle that declares no size now has to clear the headroom floor, and the pre-flight measures containerd's root as well as the docker data-root: docker 29 keeps image layers in the containerd image store, and on env-dev-08 those sat on different filesystems (93 GiB free where the agent looked, 2.8 GiB where the pull wrote). Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KLB7jieMNRkTsJ2epr4Ds1	2026-09-09 01:48:49 +00:00
mdella	31586c3058	fix: tenancy.port is the intercept port (443); clear sdk dial error; 0.1.5 install.sh wrote tenancy.port 8081 (the in-pod port) but the monky.tenancy.deploy intercept is 443, so the SDK found no intercept and the check-in failed with 'service not available' followed by the SDK's bare TypeError. Also wrap non-OSError SDK failures in a TransportError that names the host:port and what to check. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KLB7jieMNRkTsJ2epr4Ds1	2026-09-07 06:05:21 +00:00
mdella	1c42e913a8	feat: monky-deployd v0.1.0 — pull agent over the mesh (ADR-0028) Stdlib-only Python 3.12 agent for docker VMs and laptops: flock → checkin (bearer = the agent's OpenBao token, bootstrapped from the install kit's jwt-tenancy deploy grant) → action apply\|none\|down → bundle (sha256 verified) → refusal checks (unresolved ${VAR} names only, manifest paths pinned to monky/data/<env>/see/, privileged/host-network, rollback, disk need×1.5+headroom) → lease → POST /v1/auth/jwt-tenancy/login → KV reads → .env 0600 → promote → compose pull/up → wait healthy → report; finally renew-self / re-lease before max TTL, scrub. Exit 0/75/78/1. Redactor log filter. Transports sdk (openziti) / proxy (ziti tunnel proxy 18443/18200) / system. Laptop mode. Packaging: hardened oneshot + 60 s timer + proxy unit, nfpm .deb with /opt/monky-deployd/venv, install.sh for Ubuntu 26.04 (Gitea release download, enrol, ACLs, bootstrap from stdin), ansible role skeleton for osg1-07. CI: lint/test on every change; wheel (openziti on ubuntu:26.04) and package (nfpm) allow_failure until runner egress is proven; GitLab release + release:gitea on v* tags. Docs: README, PROTOCOL, OPERATIONS, CHANGELOG, CLAUDE/AGENTS. Divergence noted: monky-tenancy main (MR !15) still ships the AppRole lease and kit; this agent implements the plan's Gate 1 RESULT (login_jwt, no unwrap) and refuses an AppRole lease loudly (LEASE_SHAPE). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KLB7jieMNRkTsJ2epr4Ds1	2026-09-05 08:01:36 +00:00

Author

SHA1

Message

Date

mdella

43615a6fda

fix(onboarding): identity read that survives a rewrite, disk refused before the pull — 0.1.8

Three faults from one agent-managed onboarding (env-dev-08, 2026-09-09), each of which pointed
the diagnosis away from the actual fault.

1. install.sh granted the agent's read on the ziti identity with a POSIX ACL. ziti-edge-tunnel
rewrites that file on a controller config update and the rewrite drops the ACL: the agent
applied cleanly at 01:21 and was failing every tick by 01:32. Group membership survives the
rewrite (the file stays ziti:ziti 0640), so install.sh and the package postinstall now add
monky-deployd to the `ziti` group, and a default ACL on the identity directory carries the
grant onto a newly created file. The explicit ACLs stay for the boxes that need them.

2. openziti.load() accepts an unreadable or malformed identity: the C SDK logs "configuration is
invalid" and returns a context that only fails at dial, as a bare TypeError, which the
transport reported as a missing intercept or a policy gap. The SDK transport now reads and
parses the identity itself and names the real fault first.

3. The disk pre-flight ran only when the bundle declared disk_need_bytes, so a bundle without one
died mid-pull with containerd's "no space left on device" — which reads as a registry fault.
A bundle that declares no size now has to clear the headroom floor, and the pre-flight measures
containerd's root as well as the docker data-root: docker 29 keeps image layers in the
containerd image store, and on env-dev-08 those sat on different filesystems (93 GiB free where
the agent looked, 2.8 GiB where the pull wrote).

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KLB7jieMNRkTsJ2epr4Ds1

2026-09-09 01:48:49 +00:00

mdella

31586c3058

fix: tenancy.port is the intercept port (443); clear sdk dial error; 0.1.5

install.sh wrote tenancy.port 8081 (the in-pod port) but the
monky.tenancy.deploy intercept is 443, so the SDK found no intercept and the
check-in failed with 'service not available' followed by the SDK's bare
TypeError. Also wrap non-OSError SDK failures in a TransportError that names
the host:port and what to check.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KLB7jieMNRkTsJ2epr4Ds1

2026-09-07 06:05:21 +00:00

mdella

1c42e913a8

feat: monky-deployd v0.1.0 — pull agent over the mesh (ADR-0028)

Stdlib-only Python 3.12 agent for docker VMs and laptops: flock → checkin
(bearer = the agent's OpenBao token, bootstrapped from the install kit's
jwt-tenancy deploy grant) → action apply|none|down → bundle (sha256
verified) → refusal checks (unresolved ${VAR} names only, manifest paths
pinned to monky/data/<env>/see/, privileged/host-network, rollback, disk
need×1.5+headroom) → lease → POST /v1/auth/jwt-tenancy/login → KV reads →
.env 0600 → promote → compose pull/up → wait healthy → report; finally
renew-self / re-lease before max TTL, scrub. Exit 0/75/78/1. Redactor log
filter. Transports sdk (openziti) / proxy (ziti tunnel proxy 18443/18200) /
system. Laptop mode.

Packaging: hardened oneshot + 60 s timer + proxy unit, nfpm .deb with
/opt/monky-deployd/venv, install.sh for Ubuntu 26.04 (Gitea release
download, enrol, ACLs, bootstrap from stdin), ansible role skeleton for
osg1-07. CI: lint/test on every change; wheel (openziti on ubuntu:26.04) and
package (nfpm) allow_failure until runner egress is proven; GitLab release +
release:gitea on v* tags. Docs: README, PROTOCOL, OPERATIONS, CHANGELOG,
CLAUDE/AGENTS.

Divergence noted: monky-tenancy main (MR !15) still ships the AppRole lease
and kit; this agent implements the plan's Gate 1 RESULT (login_jwt, no
unwrap) and refuses an AppRole lease loudly (LEASE_SHAPE).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KLB7jieMNRkTsJ2epr4Ds1

2026-09-05 08:01:36 +00:00

3 Commits