Skip to content

feat(B-0812 / B-0835 Bug 4): iter-5.4.1 self-registration commit+push (Step 6.9) — opens registration PR per node-install; fixes CORE REQUIREMENT failure #5352

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
AceHack merged 1 commit into from
May 26, 2026
Merged

feat(B-0812 / B-0835 Bug 4): iter-5.4.1 self-registration commit+push (Step 6.9) — opens registration PR per node-install; fixes CORE REQUIREMENT failure #5352

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
174 changes: 174 additions & 0 deletions full-ai-cluster/usb-nixos-installer/zeta-install.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -671,6 +671,161 @@ else
fi
echo

# ── Step 6.9: iter-5.4.1 self-registration commit+push (B-0812) ──
# B-0794 sub-target 3 full implementation. After iter-5.4.0 captures
# operator's gh-auth foothold + ssh pubkeys, this step:
# 1. Probes hardware (CPU/RAM/cores/GPU/storage/network/MAC)
# 2. Composes a ClusterNode YAML matching the provisional schema
# 3. Opens a PR on the Zeta repo registering this node under
# maintainers/<operator-gh-user>/cluster-nodes/<hostname>/node.yaml
#
# Operator (or peer agent) merges the PR from anywhere (phone-merge OK).
# ArgoCD then watches maintainers/*/cluster-nodes/** and reconciles
# the node into the cluster (B-0813 iter-5.4.2; tracked separately).
#
# Skip conditions (cascade with iter-5.4.0):
# - GH_AUTH_OK != 1 (gh auth login was skipped or failed)
# - hostname unknown (iter-5.2 hostname injection also skipped)
#
# Empirical anchor: operator 2026-05-26 physical hardware-support test
# verified self-registration did NOT happen — maintainers/aaron/cluster-
Comment thread
AceHack marked this conversation as resolved.
# nodes/ doesn't exist on the repo. This Step 6.9 implements the missing
# substrate to fix B-0835 Bug 4 (CRITICAL per operator's CORE REQUIREMENT
# of post-boot fully-operational chain without operator login).
SELF_REG_OK=0
SELF_REG_PR_URL=""
if [ "$GH_AUTH_OK" = 1 ]; then
echo "[iter-5.4.1] ── self-registration commit+push (B-0812) ──"
echo "[iter-5.4.1] Composing ClusterNode YAML + opening registration PR..."

# Resolve operator GH user (used for the per-maintainer subtree path).
MAINTAINER=$(gh api /user --jq .login 2>/dev/null || echo "")
if [ -z "$MAINTAINER" ]; then
echo "[iter-5.4.1] WARN: gh api /user failed; cannot resolve operator GH login; skipping"
else
# Resolve installed hostname (iter-5.2 substrate writes to
# /mnt/etc/zeta/cluster-node-id). Fallback to flake-default $HOST
# if the iter-5.2 file is absent (means iter-5.2.2 generation was
# skipped or failed — graceful degradation; warn loudly).
if [ -f "$HOSTNAME_DST" ]; then
NODE_HOSTNAME=$(cat "$HOSTNAME_DST" | tr -d '[:space:]')
else
NODE_HOSTNAME="$HOST"
echo "[iter-5.4.1] WARN: $HOSTNAME_DST absent; using flake-host '$HOST' as node-name"
echo "[iter-5.4.1] (may produce naming collision if multiple nodes use this flake-host)"
fi
echo "[iter-5.4.1] maintainer: $MAINTAINER"
echo "[iter-5.4.1] node-name: $NODE_HOSTNAME"

# ── hardware probe ──
# Emits the inner fields of the ClusterNode `hardware:` block.
# Each field is best-effort; absent fields are omitted from YAML
# rather than emitting empty-string values (ArgoCD/k8s consumers
# prefer absent over empty).
CPU_MODEL=$(grep 'model name' /proc/cpuinfo 2>/dev/null | head -1 | cut -d: -f2- | sed 's/^[[:space:]]*//' | sed 's/"//g' || echo "")
MEM_TOTAL=$(free -h --si 2>/dev/null | awk '/Mem:/{print $2}' || echo "")
CPU_CORES=$(nproc 2>/dev/null || echo "")
GPU_LINE=$(lspci -nn 2>/dev/null | grep -iE 'vga|3d|display' | head -1 | sed 's/"//g' || echo "")
IP_ADDR=$(ip -4 -o addr 2>/dev/null | awk '/inet/ && !/lo/{print $4; exit}' || echo "")
MAC_ADDR=$(ip -o link 2>/dev/null | awk '/state UP/ && !/lo/{print $(NF-2); exit}' || echo "")
Comment thread
AceHack marked this conversation as resolved.
STORAGE_LINES=$(lsblk -ndo NAME,SIZE,TYPE -e7 2>/dev/null | awk '$3=="disk"{print " - \"/dev/" $1 " " $2 "\""}' || echo "")
REG_TIMESTAMP=$(date -u +"%Y-%m-%dT%H:%M:%SZ")
FLAKE_COMMIT=$(git -C /mnt/etc/zeta rev-parse HEAD 2>/dev/null | head -c 12 || echo "unknown")

# ── compose node.yaml ──
NODE_YAML="apiVersion: zeta.lucent-financial-group.com/v1
kind: ClusterNode
metadata:
name: $NODE_HOSTNAME
namespace: zeta-cluster
annotations:
zeta.lucent-financial-group.com/registered-at: \"$REG_TIMESTAMP\"
zeta.lucent-financial-group.com/flake-commit: \"$FLAKE_COMMIT\"
zeta.lucent-financial-group.com/flake-host: \"$HOST\"
zeta.lucent-financial-group.com/registered-via: \"iter-5.4.1\"
spec:
hostname: $NODE_HOSTNAME
role: $HOST
maintainer: $MAINTAINER
Comment thread
AceHack marked this conversation as resolved.
hardware:"
[ -n "$CPU_MODEL" ] && NODE_YAML="$NODE_YAML
cpu: \"$CPU_MODEL\""
[ -n "$MEM_TOTAL" ] && NODE_YAML="$NODE_YAML
memory: \"$MEM_TOTAL\""
[ -n "$CPU_CORES" ] && NODE_YAML="$NODE_YAML
cores: $CPU_CORES"
[ -n "$GPU_LINE" ] && NODE_YAML="$NODE_YAML
gpu: \"$GPU_LINE\""
[ -n "$STORAGE_LINES" ] && NODE_YAML="$NODE_YAML
storage:
$STORAGE_LINES"
Comment thread
AceHack marked this conversation as resolved.
if [ -n "$IP_ADDR" ] || [ -n "$MAC_ADDR" ]; then
NODE_YAML="$NODE_YAML
network:"
[ -n "$IP_ADDR" ] && NODE_YAML="$NODE_YAML
ip: \"$IP_ADDR\""
[ -n "$MAC_ADDR" ] && NODE_YAML="$NODE_YAML
mac: \"$MAC_ADDR\""
fi

# ── clone repo to temp; write node.yaml; commit + open PR ──
WORK_DIR=$(mktemp -d -t zeta-self-register.XXXXXX)
REG_BRANCH="register-${NODE_HOSTNAME}-$(date -u +%Y%m%dT%H%M%SZ)"
if gh repo clone Lucent-Financial-Group/Zeta "$WORK_DIR" -- --depth 1 --quiet 2>&1 | tail -3; then
NODE_DIR="$WORK_DIR/maintainers/$MAINTAINER/cluster-nodes/$NODE_HOSTNAME"
mkdir -p "$NODE_DIR"
printf '%s\n' "$NODE_YAML" > "$NODE_DIR/node.yaml"
(
cd "$WORK_DIR"
# commit-author = gh-auth'd operator (no shipped credentials;
# clean attribution chain). Configure user.{name,email} from gh.
OP_NAME=$(gh api /user --jq .name 2>/dev/null || echo "$MAINTAINER")
OP_EMAIL=$(gh api /user/emails --jq '.[] | select(.primary == true) | .email' 2>/dev/null \
| head -1 || echo "${MAINTAINER}@users.noreply.github.com")
git config user.name "$OP_NAME"
git config user.email "$OP_EMAIL"
git checkout -b "$REG_BRANCH"
git add "maintainers/$MAINTAINER/cluster-nodes/$NODE_HOSTNAME/"
git commit -m "feat(node-register): $NODE_HOSTNAME self-registers via iter-5.4.1

Auto-generated by zeta-install.sh Step 6.9 on the node during install.
Registers ${NODE_HOSTNAME} under maintainers/${MAINTAINER}/cluster-nodes/.
ArgoCD watches maintainers/*/cluster-nodes/** + reconciles per B-0813.

flake-host: ${HOST}
flake-commit: ${FLAKE_COMMIT}
registered-at: ${REG_TIMESTAMP}
" >/dev/null 2>&1
git push -u origin "$REG_BRANCH" >/dev/null 2>&1
SELF_REG_PR_URL=$(gh pr create \
--title "feat(node-register): $NODE_HOSTNAME self-registers via iter-5.4.1" \
--body "Self-registration PR opened by zeta-install.sh on the node during install. Composes with B-0812 iter-5.4.1 + B-0813 iter-5.4.2 ArgoCD reconciliation. Review + merge to bring the node into the cluster." \
--base main \
--head "$REG_BRANCH" 2>&1 | tail -1)
echo "$SELF_REG_PR_URL" > /tmp/zeta-self-reg-pr-url
)
Comment thread
AceHack marked this conversation as resolved.
if [ -s /tmp/zeta-self-reg-pr-url ]; then
SELF_REG_PR_URL=$(cat /tmp/zeta-self-reg-pr-url)
SELF_REG_OK=1
echo "[iter-5.4.1] SUCCESS — registration PR opened: $SELF_REG_PR_URL"
echo "[iter-5.4.1] Operator merges from anywhere (phone-merge OK)."
echo "[iter-5.4.1] ArgoCD reconciles after merge per B-0813 iter-5.4.2."
else
echo "[iter-5.4.1] WARN: gh pr create did not return a URL; check $WORK_DIR for state"
fi
else
echo "[iter-5.4.1] WARN: gh repo clone failed; skipping self-registration"
echo "[iter-5.4.1] (operator can re-run manually post-install)"
fi
# Cleanup: temp dir is operator-owned + safe to remove
rm -rf "$WORK_DIR" /tmp/zeta-self-reg-pr-url 2>/dev/null || true
fi
else
echo "[iter-5.4.1] skipped — iter-5.4.0 gh-auth was skipped or failed; no auth foothold for commit+push"
echo "[iter-5.4.1] (operator can re-run manually post-install via tools/cluster/register-node.ts when that ships)"
fi
echo

echo "Running nixos-install --flake /mnt/etc/zeta/full-ai-cluster#$HOST ..."
sudo nixos-install --flake "/mnt/etc/zeta/full-ai-cluster#$HOST" --no-root-password

Expand Down Expand Up @@ -702,6 +857,25 @@ if [ "$GH_AUTH_OK" = 1 ] && [ "$GH_KEY_COUNT" != "0" ]; then
echo " your registered-with-GitHub SSH keys:"
echo " ssh zeta@\$(hostname).local"
echo

# B-0812 iter-5.4.1: surface the self-registration PR URL if Step 6.9
# opened one. This is the operator's call-to-action — merge the PR
# from anywhere (phone OK) to bring the node into the cluster via
# ArgoCD reconciliation (B-0813 iter-5.4.2).
if [ "$SELF_REG_OK" = 1 ] && [ -n "$SELF_REG_PR_URL" ]; then
echo " iter-5.4.1 SELF-REGISTRATION: SUCCESS"
echo " Node-registration PR opened:"
echo " $SELF_REG_PR_URL"
echo " Review + merge → ArgoCD reconciles → node joins cluster"
echo " (phone-merge OK — no laptop kubectl required)"
echo
else
echo " iter-5.4.1 SELF-REGISTRATION: SKIPPED (see diagnostics above)"
echo " Manual fallback: tools/cluster/register-node.ts (when shipped)"
echo " OR push commit to maintainers/<your-gh-user>/cluster-nodes/<hostname>/node.yaml"
echo
fi

echo " AFTER FIRST LOGIN:"
echo " 1. (password already set per iter-5.3 prompt — or unchanged"
echo " if iter-5.3 was skipped; rotate via 'passwd zeta' anytime)"
Expand Down
Loading