gsadmin/PSProxmoxVE
1
0
Fork 0
mirror of https://github.com/GoodOlClint/PSProxmoxVE.git synced 2026-09-04 03:05:32 +00:00
Code Issues Packages Projects Releases Wiki Activity

ci: storage VM setup diagnoses its own network and logs phases

Browse Source 
Run 168 burned 13 minutes to report 'unable to locate package tgt' when
the real failure was dead egress from the CI VLAN (apt lists never
updated). The setup script now probes gateway/internet/DNS from inside
the VM and prints the verdicts, fails apt fast with the real error, and
marks each phase.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
goodolclint-claude[bot]
2026-08-31 20:39:02 +00:00
committed by GitHub
parent e219766452
commit 5d0b384508
1 changed files with 30 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/infrastructure/scripts/setup-storage-server.sh
+30 -4
View File
@@ -38,8 +38,21 @@ done
# sshd accepts logins before cloud-init's first-boot work (and its apt/dpkg
# locks) is finished; a degraded cloud-init result is fine for our purposes,
# and a hung cloud-init must fail fast, not eat the job timeout.
echo "Waiting for cloud-init to finish (up to 300s)..."
timeout 300 ssh ${SSH_OPTS} "ubuntu@${STORAGE_IP}" \
"sudo cloud-init status --wait" || true
"sudo cloud-init status --wait" \
|| echo "WARNING: cloud-init did not finish cleanly within 300s — continuing"
echo "Network probe from the storage VM:"
ssh ${SSH_OPTS} "ubuntu@${STORAGE_IP}" '
ip -4 -brief addr; ip route show default
ping -c1 -W2 "$(ip route show default | awk "{print \$3; exit}")" >/dev/null 2>&1 \
&& echo "gateway ping: OK" || echo "gateway ping: FAIL"
ping -c1 -W2 1.1.1.1 >/dev/null 2>&1 \
&& echo "internet ping: OK" || echo "internet ping: FAIL"
getent hosts archive.ubuntu.com >/dev/null 2>&1 \
&& echo "DNS resolve: OK" || echo "DNS resolve: FAIL (resolv.conf: $(grep ^nameserver /etc/resolv.conf | tr "\n" " "))"
' || true
echo "Copying answer files..."
ssh ${SSH_OPTS} "ubuntu@${STORAGE_IP}" \
@@ -53,11 +66,23 @@ ssh ${SSH_OPTS} "ubuntu@${STORAGE_IP}" \
set -euo pipefail
export DEBIAN_FRONTEND=noninteractive
# cloud-init's first-boot apt activity can hold the dpkg lock briefly
for i in $(seq 1 30); do
apt-get update -qq >/dev/null 2>&1 && break
# cloud-init's first-boot apt activity can hold the dpkg lock briefly;
# a network/DNS failure must fail fast with the real apt error, not fall
# through to a misleading "package not found" from empty lists.
echo "Updating apt package lists..."
for i in $(seq 1 12); do
if apt_out=$(apt-get update -qq 2>&1); then
break
fi
echo "apt-get update attempt $i/12 failed"
if [ "$i" -eq 12 ]; then
echo "$apt_out" >&2
echo "ERROR: apt-get update never succeeded — check CI VLAN egress/DNS" >&2
exit 1
fi
sleep 5
done
echo "Installing nfs-kernel-server, tgt, docker.io..."
apt-get install -y -qq nfs-kernel-server tgt docker.io >/dev/null
# Ubuntu's nfs-kernel-server package does not create /etc/exports.d
@@ -80,6 +105,7 @@ systemctl enable --now nfs-kernel-server tgt >/dev/null || {
}
systemctl restart tgt
echo "Starting answer server container..."
docker rm -f pvetest-answer-server >/dev/null 2>&1 || true
docker run -d --name pvetest-answer-server --restart unless-stopped \
--network host \