andreas/vanlink
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

deploy: --apply-network changes nothing itself; helper does install/apply, two-stage gate, checked rollback, status file

Browse Source 
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KqhjP7nud5gdBg25zCiRk9
This commit is contained in:
andreas
2026-09-23 18:44:59 -04:00
co-authored by Claude Sonnet 5
parent eaec8eb8b9
commit d46fd5a1e4
2 changed files with 265 additions and 62 deletions
Download Patch File Download Diff File Expand all files Collapse all files
deploy.sh
+164 -54
View File
@@ -11,8 +11,10 @@
# ./deploy.sh --classify-radio # reads `iw dev <if> info` text on stdin, prints beaconing|silent (no ssid line = silent; used by post_check; for tests) # ./deploy.sh --classify-radio # reads `iw dev <if> info` text on stdin, prints beaconing|silent (no ssid line = silent; used by post_check; for tests)
# ./deploy.sh --render-wan-netplan # rendered /etc/netplan/50-van-wan.yaml (ethernet part + wifis stanza when WIFI_WAN_IFACE is set) # ./deploy.sh --render-wan-netplan # rendered /etc/netplan/50-van-wan.yaml (ethernet part + wifis stanza when WIFI_WAN_IFACE is set)
# ./deploy.sh --wan-plan [conf] # what the NetworkManager WAN provisioning would do (read-only, no root) # ./deploy.sh --wan-plan [conf] # what the NetworkManager WAN provisioning would do (read-only, no root)
# ./deploy.sh --apply-network # root: install the WAN netplan, apply it from a detached unit, roll back automatically if the uplink does not return # ./deploy.sh --apply-network [--force] # root: a detached helper installs the WAN netplan, applies it, gates on uplink health (2 stages) and rolls back
# # automatically on failure; status in /var/lib/vanlink/apply-network.status. No-op when already applied unless --force.
# ./deploy.sh --print-apply-helper [conf] # print the detached apply/rollback helper script (read-only, no root; for review and tests) # ./deploy.sh --print-apply-helper [conf] # print the detached apply/rollback helper script (read-only, no root; for review and tests)
# # test-only flags: --backup --netplan-dir --log --status --sysnet --rendered --deadlines "s1 s2 rc" --stamp
# ./deploy.sh render <file> # preview a templated file on stdout # ./deploy.sh render <file> # preview a templated file on stdout
# ./deploy.sh --redact # filter stdin, masking secret values (used by --diff; for tests) # ./deploy.sh --redact # filter stdin, masking secret values (used by --diff; for tests)
# VANLINK_NM_DIR=<dir> # test override: where existing NM *.nmconnection files are looked up (default /etc/NetworkManager/system-connections) # VANLINK_NM_DIR=<dir> # test override: where existing NM *.nmconnection files are looked up (default /etc/NetworkManager/system-connections)
@@ -277,68 +279,158 @@ post_check() { # read-only live checks; every probe is guarded, warns instead o
else warn "post-deploy: default route present but 1.1.1.1 does not answer ping"; fi else warn "post-deploy: default route present but 1.1.1.1 does not answer ping"; fi
} }
apply_helper_script() { # apply_helper_script <backup-dir> — the script the detached unit runs # --apply-network time budgets (seconds), in one place. The helper's worst case is
local bak=$1 # apply + stage 1 + restart + stage 2 + rollback (apply + restart + recheck); the
# Unquoted heredoc on purpose: $bak and $STARLINK_IFACE expand now; \$... stay for run time. # foreground waits that long plus a margin. van-failover starts optimistic and
# probes every WAN at once (probe_timeout 3 s), enforcing the route in its first
# loop, so a healthy default route shows up within seconds of a restart; the 120 s
# stage 2 deadline is a wide margin for a slow Starlink/cellular probe.
AN_APPLY_T=90 AN_RESTART_T=30 AN_S1=60 AN_S2=120 AN_RC=60
AN_WAIT=$(( 2 * AN_APPLY_T + 2 * AN_RESTART_T + AN_S1 + AN_S2 + AN_RC + 30 ))
# apply_helper_script <backup-dir> [netplan-dir] [log] [status-file] [sysnet-dir] [rendered-file] [deadlines "s1 s2 rc"] [run-stamp]
# The script the detached unit runs. Paths are parameters only so tests can bake
# sandbox paths into the text; the production helper reads no environment.
apply_helper_script() {
local bak=$1 nd=${2:-/etc/netplan} log=${3:-/var/log/vanlink-apply-network.log}
local st=${4:-/var/lib/vanlink/apply-network.status} sn=${5:-/sys/class/net}
local rf=${6:-/run/vanlink-50-van-wan.yaml} dl=${7:-"$AN_S1 $AN_S2 $AN_RC"} stamp=${8:-manual}
local d1 d2 d3
read -r d1 d2 d3 <<< "$dl"
# Unquoted heredoc on purpose: values expand now; \$... stays for run time.
cat <<HELPER cat <<HELPER
#!/bin/bash #!/bin/bash
# Generated by deploy.sh --apply-network: apply the WAN netplan, verify the uplink, roll back if it did not come back. # Generated by deploy.sh --apply-network; runs detached. Installs the rendered WAN netplan,
exec >>/var/log/vanlink-apply-network.log 2>&1 # applies it, verifies the uplink in two stages and restores the pre-run netplan on any failure.
echo "== apply-network \$(date -Is) (backup: $bak)" # Stage 1 (${d1}s): the apply succeeded and, if the Starlink port exists, NM shows it connected
netplan generate && netplan apply || echo "netplan apply failed" # via a netplan-* profile with an IPv4 address whose gateway answers on-link (the profile is
ok=0 # never-default; van-failover owns default routes).
for i in \$(seq 1 60); do # Stage 2 (${d2}s): van-failover restarted; default route and internet ping.
have_default=0; ip route show default | grep . >/dev/null && have_default=1 bak='$bak'
nm_ok=1 nd='$nd'
if [ -e /sys/class/net/$STARLINK_IFACE ]; then log='$log'
nm_ok=0 st='$st'
nmcli -t -f DEVICE,STATE device 2>/dev/null | grep -Fx '$STARLINK_IFACE:connected' >/dev/null && nm_ok=1 sn='$sn'
rf='$rf'
exec >>"\$log" 2>&1
status() { mkdir -p "\$(dirname "\$st")"; echo "\$1 \$(date -Is)\${2:+ \$2}" > "\$st"; }
echo "== apply-network $stamp \$(date -Is) (backup: \$bak)"
conn_ok() { ip route show default | grep . >/dev/null && ping -c1 -W2 1.1.1.1 >/dev/null 2>&1; }
gate1() {
[ -e "\$sn/$STARLINK_IFACE" ] || return 0
[ "\$(nmcli -g GENERAL.STATE device show '$STARLINK_IFACE' 2>/dev/null)" = "100 (connected)" ] || return 1
case "\$(nmcli -g GENERAL.CONNECTION device show '$STARLINK_IFACE' 2>/dev/null)" in netplan-*) ;; *) return 1 ;; esac
a=\$(nmcli -g IP4.ADDRESS device show '$STARLINK_IFACE' 2>/dev/null)
g=\$(nmcli -g IP4.GATEWAY device show '$STARLINK_IFACE' 2>/dev/null)
[ -n "\$a" ] && [ "\$a" != -- ] && [ -n "\$g" ] && [ "\$g" != -- ] || return 1
ping -c1 -W2 -I '$STARLINK_IFACE' "\$g" >/dev/null 2>&1
}
restore() { # restore <reason>: put the pre-run netplan back, re-check connectivity, always exit 1
local rb=ok c=down
echo "apply-network FAILED (\$1) - restoring \$bak"
if [ -d "\$bak" ]; then
rm -f "\$nd"/* "\$nd"/.[!.]*
cp -a "\$bak"/. "\$nd"/ || rb=CP-FAILED
else
rb=BACKUP-MISSING
fi fi
if [ \$have_default = 1 ] && [ \$nm_ok = 1 ] && ping -c1 -W2 1.1.1.1 >/dev/null 2>&1; then ok=1; break; fi netplan generate && timeout $AN_APPLY_T netplan apply || { [ "\$rb" != ok ] || rb=APPLY-FAILED; }
timeout $AN_RESTART_T systemctl restart van-failover || true
end=\$((SECONDS+$d3))
while :; do
conn_ok && { c=ok; break; }
[ "\$SECONDS" -ge "\$end" ] && break
sleep 1 sleep 1
done done
if [ \$ok = 1 ]; then if [ "\$rb" = ok ] && [ "\$c" = ok ]; then
echo "apply-network OK" echo "apply-network ROLLED BACK (restored, connectivity ok)"
status rolled-back "\$1"
else else
echo "apply-network FAILED — restoring $bak" echo "apply-network ROLLBACK FAILED: \$rb / connectivity \$c"
rm -f /etc/netplan/* status rollback-failed "\$1; \$rb / connectivity \$c"
cp -a $bak/. /etc/netplan/ [ "\$c" = ok ] || { echo "no connectivity after rollback - rebooting"; systemctl reboot; }
netplan generate && netplan apply
echo "apply-network ROLLED BACK"
exit 1
fi fi
exit 1
}
applied=0
if install -m0600 "\$rf" "\$nd/50-van-wan.yaml" \\
&& { [ ! -f "\$nd/10-dhcp-all-interfaces.yaml" ] || mv "\$nd/10-dhcp-all-interfaces.yaml" "\$nd/10-dhcp-all-interfaces.yaml.vanlink-disabled"; } \\
&& netplan generate && timeout $AN_APPLY_T netplan apply; then
applied=1
else
echo "netplan install/generate/apply failed"
fi
s1=0
if [ "\$applied" = 1 ]; then
end=\$((SECONDS+$d1))
while :; do
gate1 && { s1=1; break; }
[ "\$SECONDS" -ge "\$end" ] && break
sleep 1
done
fi
[ "\$s1" = 1 ] || restore "stage 1: apply or uplink gate failed"
timeout $AN_RESTART_T systemctl restart van-failover || restore "van-failover restart failed"
s2=0
end=\$((SECONDS+$d2))
while :; do
conn_ok && { s2=1; break; }
[ "\$SECONDS" -ge "\$end" ] && break
sleep 1
done
[ "\$s2" = 1 ] || restore "stage 2: no default route or internet"
status ok
echo "apply-network OK"
exit 0
HELPER HELPER
} }
apply_network() { # ./deploy.sh --apply-network — explicit, root; see spec §4 apply_network() { # ./deploy.sh --apply-network [--force] — explicit, root; changes nothing under /etc/netplan itself (the detached helper does); see spec §4
local force=0
[ "${1:-}" = --force ] && force=1
[ "$(id -u)" = 0 ] || die "--apply-network needs root (sudo)" [ "$(id -u)" = 0 ] || die "--apply-network needs root (sudo)"
[ "$ENABLE_NM_WAN" = 1 ] || die "ENABLE_NM_WAN is 0 in $VANLINK_CONF — nothing to apply" [ "$ENABLE_NM_WAN" = 1 ] || die "ENABLE_NM_WAN is 0 in $VANLINK_CONF — nothing to apply"
dpkg -s network-manager >/dev/null 2>&1 || die "network-manager is not installed — run ./deploy.sh first" dpkg-query -W -f='${Status}' network-manager 2>/dev/null | grep -q 'install ok installed' \
local stamp bak tmp || die "network-manager is not installed — run ./deploy.sh first"
stamp=$(date +%Y%m%d-%H%M%S) systemctl is-active --quiet NetworkManager || die "NetworkManager is not running — handing it the uplink port would flap it"
bak=/var/backups/vanlink-netplan-$stamp systemctl is-active --quiet vanlink-apply-network && die "an apply is already running — see /var/log/vanlink-apply-network.log"
install -d -m0700 "$bak" if [ "$force" = 0 ] && [ "$(nm_wan_state_netplan)" = matches ] && [ ! -e /etc/netplan/10-dhcp-all-interfaces.yaml ]; then
cp -a /etc/netplan/. "$bak/" echo "nothing to apply (WAN netplan already matches; use --force to re-apply)"
tmp=$(mktemp) return 0
render_wan_netplan > "$tmp"
check_rendered ap/50-van-wan.yaml "$tmp" || { rm -f "$tmp"; exit 1; }
install -m0600 "$tmp" /etc/netplan/50-van-wan.yaml
rm -f "$tmp"
if [ -f /etc/netplan/10-dhcp-all-interfaces.yaml ]; then
mv /etc/netplan/10-dhcp-all-interfaces.yaml /etc/netplan/10-dhcp-all-interfaces.yaml.vanlink-disabled
fi fi
: > /var/log/vanlink-apply-network.log local rendered=/run/vanlink-50-van-wan.yaml stamp bak log=/var/log/vanlink-apply-network.log st=/var/lib/vanlink/apply-network.status
apply_helper_script "$bak" > /run/vanlink-apply-network.sh (umask 077; render_wan_netplan > "$rendered") || die "render failed"
check_rendered ap/50-van-wan.yaml "$rendered" || { rm -f "$rendered"; exit 1; }
stamp=$(date +%Y%m%d-%H%M%S)
bak=$(mktemp -d "/var/backups/vanlink-netplan-$stamp.XXXX") || die "cannot create backup dir"
cp -a /etc/netplan/. "$bak/" || die "backup failed — nothing changed"
if [ ! -e /var/backups/vanlink-netplan-orig ] && [ ! -e /etc/netplan/50-van-wan.yaml ]; then
install -d -m0700 /var/backups/vanlink-netplan-orig
cp -a /etc/netplan/. /var/backups/vanlink-netplan-orig/ || { rm -rf /var/backups/vanlink-netplan-orig; die "pristine backup failed — nothing changed"; }
fi
install -d -m0755 /var/lib/vanlink
rm -f "$st"
apply_helper_script "$bak" /etc/netplan "$log" "$st" /sys/class/net "$rendered" "$AN_S1 $AN_S2 $AN_RC" "$stamp" > /run/vanlink-apply-network.sh
chmod 0700 /run/vanlink-apply-network.sh chmod 0700 /run/vanlink-apply-network.sh
echo "backup: $bak log: /var/log/vanlink-apply-network.log" echo "backup: $bak log: $log"
echo "applying from a detached unit (survives an SSH drop); waiting up to 90 s ..." echo "applying from a detached unit (survives an SSH drop); waiting up to $AN_WAIT s ..."
systemctl reset-failed vanlink-apply-network 2>/dev/null || true systemctl reset-failed vanlink-apply-network 2>/dev/null || true
systemd-run --unit=vanlink-apply-network --collect /bin/bash /run/vanlink-apply-network.sh >/dev/null systemd-run --unit=vanlink-apply-network --collect -p RuntimeMaxSec=900 /bin/bash /run/vanlink-apply-network.sh >/dev/null \
local t=0 || die "systemd-run failed — nothing was changed (backup: $bak)"
while systemctl is-active --quiet vanlink-apply-network && [ "$t" -lt 90 ]; do sleep 1; t=$((t + 1)); done SECONDS=0
tail -n 6 /var/log/vanlink-apply-network.log 2>/dev/null || true while systemctl is-active --quiet vanlink-apply-network && [ "$SECONDS" -lt "$AN_WAIT" ]; do sleep 1; done
grep -q 'apply-network OK' /var/log/vanlink-apply-network.log 2>/dev/null \ awk -v s="== apply-network $stamp" 'index($0, s) == 1 { p = 1 } p' "$log" 2>/dev/null | tail -n 8 || true
|| die "apply-network did not report OK — see /var/log/vanlink-apply-network.log (previous config: $bak)" if systemctl is-active --quiet vanlink-apply-network; then
echo "still running — check $log"
return 2
fi
local verdict
verdict=$(cut -d' ' -f1 "$st" 2>/dev/null || true)
case "$verdict" in
ok) return 0 ;;
rolled-back|rollback-failed) echo "apply-network: $(cat "$st") (previous config: $bak)" >&2; return 1 ;;
*) echo "apply-network: no status file — see $log (previous config: $bak)" >&2; return 1 ;;
esac
} }
case "${1:-}" in case "${1:-}" in
@@ -386,13 +478,31 @@ case "${1:-}" in
--redact) # ./deploy.sh --redact — stdin through redact_secrets (no config, no root) --redact) # ./deploy.sh --redact — stdin through redact_secrets (no config, no root)
redact_secrets redact_secrets
exit 0 ;; exit 0 ;;
--apply-network) # ./deploy.sh --apply-network — root; applies the WAN netplan with automatic rollback --apply-network) # ./deploy.sh --apply-network [--force] — root; applies the WAN netplan from a detached helper with automatic rollback
load_conf "$VANLINK_CONF" load_conf "$VANLINK_CONF"
apply_network apply_network "${2:-}"
exit 0 ;; exit $? ;;
--print-apply-helper) # for review/tests: the helper script, with a placeholder backup dir; no root, no changes --print-apply-helper) # [conf] the helper script for review; test-only flags: --backup D --netplan-dir D --log L --status S --sysnet N --rendered R --deadlines "s1 s2 rc" --stamp T
load_conf "${2:-$VANLINK_CONF}" shift; pc= a_bak=/var/backups/vanlink-netplan-EXAMPLE a_nd= a_log= a_st= a_sn= a_rf= a_dl= a_stamp=
apply_helper_script /var/backups/vanlink-netplan-EXAMPLE while [ $# -gt 0 ]; do
case "$1" in
--backup|--netplan-dir|--log|--status|--sysnet|--rendered|--deadlines|--stamp)
[ $# -ge 2 ] || die "$1 needs a value" ;;
esac
case "$1" in
--backup) a_bak=$2; shift 2 ;;
--netplan-dir) a_nd=$2; shift 2 ;;
--log) a_log=$2; shift 2 ;;
--status) a_st=$2; shift 2 ;;
--sysnet) a_sn=$2; shift 2 ;;
--rendered) a_rf=$2; shift 2 ;;
--deadlines) a_dl=$2; shift 2 ;;
--stamp) a_stamp=$2; shift 2 ;;
*) pc=$1; shift ;;
esac
done
load_conf "${pc:-$VANLINK_CONF}"
apply_helper_script "$a_bak" "$a_nd" "$a_log" "$a_st" "$a_sn" "$a_rf" "$a_dl" "$a_stamp"
exit 0 ;; exit 0 ;;
--render-wan-netplan) load_conf "$VANLINK_CONF"; render_wan_netplan; exit 0 ;; --render-wan-netplan) load_conf "$VANLINK_CONF"; render_wan_netplan; exit 0 ;;
--wan-plan) # ./deploy.sh --wan-plan [conf] — what the NM WAN provisioning would do; changes nothing --wan-plan) # ./deploy.sh --wan-plan [conf] — what the NM WAN provisioning would do; changes nothing
tests/check-templates.sh
+99 -6
View File
@@ -257,16 +257,109 @@ rm -rf "$nmd" "$nme"
# 17c. NM must not auto-create "Wired connection" profiles # 17c. NM must not auto-create "Wired connection" profiles
[ "$(grep -v '^[[:space:]]*#' ap/nm-no-auto-default.conf 2>/dev/null | grep -v '^[[:space:]]*$')" = "$(printf '[main]\nno-auto-default=*')" ] && ok "nm-no-auto-default.conf is [main] + no-auto-default=*" || bad "ap/nm-no-auto-default.conf missing or wrong" [ "$(grep -v '^[[:space:]]*#' ap/nm-no-auto-default.conf 2>/dev/null | grep -v '^[[:space:]]*$')" = "$(printf '[main]\nno-auto-default=*')" ] && ok "nm-no-auto-default.conf is [main] + no-auto-default=*" || bad "ap/nm-no-auto-default.conf missing or wrong"
# 18. the detached apply helper: syntax, ordering, rollback branch # 18. the detached apply helper: syntax, then behaviour against stubbed system commands in a sandbox
h=$(VANLINK_CONF=hosts/van.conf ./deploy.sh --print-apply-helper 2>&1) || { bad "--print-apply-helper failed"; echo "$h" | sed 's/^/ /'; h=; } h=$(VANLINK_CONF=hosts/van.conf ./deploy.sh --print-apply-helper 2>&1) || { bad "--print-apply-helper failed"; echo "$h" | sed 's/^/ /'; h=; }
if [ -n "$h" ]; then if [ -n "$h" ]; then
echo "$h" | bash -n && ok "apply helper: bash -n" || bad "apply helper: bash -n" echo "$h" | bash -n && ok "apply helper: bash -n" || bad "apply helper: bash -n"
echo "$h" | grep -q 'netplan generate' && echo "$h" | grep -q 'netplan apply' && ok "apply helper applies netplan" || bad "apply helper lacks netplan generate/apply"
echo "$h" | grep -q 'restoring' && ok "apply helper has a rollback branch" || bad "apply helper lacks rollback"
echo "$h" | grep -q 'enP2p33s0' && ok "apply helper waits on the Starlink port" || bad "apply helper does not reference STARLINK_IFACE" echo "$h" | grep -q 'enP2p33s0' && ok "apply helper waits on the Starlink port" || bad "apply helper does not reference STARLINK_IFACE"
# rollback must come after the wait loop, and the backup is restored wholesale echo "$h" | grep -q 'systemctl restart van-failover' && ok "apply helper restarts van-failover" || bad "apply helper lacks van-failover restart"
a=$(echo "$h" | grep -n 'netplan apply' | head -1 | cut -d: -f1); r=$(echo "$h" | grep -n 'restoring' | head -1 | cut -d: -f1) stubs=$(mktemp -d)
[ "$a" -lt "$r" ] && ok "apply helper: apply precedes rollback" || bad "apply helper: rollback ordering" for c in netplan nmcli ip ping systemctl reboot; do
printf '#!/bin/bash\necho "%s $*" >> "$STUB_CALLS"\n' "$c" > "$stubs/$c"
done
cat >> "$stubs/netplan" <<'STUB'
if [ "$1" = apply ]; then
n=$(cat "$STUB_SB/napply" 2>/dev/null || echo 0); echo $((n + 1)) > "$STUB_SB/napply"
[ "${STUB_FAIL_APPLY:-}" = first ] && [ "$n" = 0 ] && exit 1
fi
exit 0
STUB
cat >> "$stubs/nmcli" <<'STUB'
case "$2" in
GENERAL.STATE) echo "${NM_STATE-100 (connected)}" ;;
GENERAL.CONNECTION) echo "${NM_CONN-netplan-enP2p33s0}" ;;
IP4.ADDRESS) echo "${NM_ADDR-192.168.100.10/24}" ;;
IP4.GATEWAY) echo "${NM_GW-192.168.100.1}" ;;
esac
exit 0
STUB
cat >> "$stubs/ip" <<'STUB'
case "${STUB_ROUTE:-yes}" in
yes) echo "default via 192.168.100.1 dev x" ;;
after-rollback) [ -e "$STUB_ND/50-van-wan.yaml" ] || echo "default via 192.168.100.1 dev x" ;;
esac
exit 0
STUB
cat >> "$stubs/ping" <<'STUB'
case " $* " in *" -I "*) exit "${STUB_GW_PING:-0}" ;; esac
exit "${STUB_NET_PING:-0}"
STUB
printf '#!/bin/bash\nexec /bin/sleep 0.05\n' > "$stubs/sleep"
printf '#!/bin/bash\nshift\nexec "$@"\n' > "$stubs/timeout"
printf '#!/bin/bash\nexit 0\n' >> "$stubs/systemctl"
chmod +x "$stubs"/*
stubs_ok=1
for c in netplan nmcli ip ping systemctl reboot sleep timeout; do [ -x "$stubs/$c" ] || stubs_ok=0; done
[ "$stubs_ok" = 1 ] && ok "apply helper sandbox: all system commands stubbed" || bad "apply helper sandbox: stub missing"
an_run() { # an_run [VAR=val ...] — fresh sandbox, run the helper; sets sb rc
sb=$(mktemp -d); mkdir -p "$sb/netplan" "$sb/sysnet"
printf 'armbian\n' > "$sb/netplan/10-dhcp-all-interfaces.yaml"; chmod 0600 "$sb/netplan/10-dhcp-all-interfaces.yaml"
printf 'old\n' > "$sb/netplan/60-other.yaml"
cp -a "$sb/netplan" "$sb/bak"
[ -n "${AN_NOPORT:-}" ] || mkdir "$sb/sysnet/enP2p33s0"
[ -z "${AN_NOBAK:-}" ] || rm -rf "$sb/bak"
printf 'network:\n version: 2\n' > "$sb/rendered.yaml"
VANLINK_CONF=hosts/van.conf ./deploy.sh --print-apply-helper --backup "$sb/bak" --netplan-dir "$sb/netplan" \
--log "$sb/log" --status "$sb/status" --sysnet "$sb/sysnet" --rendered "$sb/rendered.yaml" \
--deadlines "1 1 1" --stamp T > "$sb/helper.sh"
env PATH="$stubs:/usr/bin:/bin" STUB_CALLS="$sb/calls" STUB_SB="$sb" STUB_ND="$sb/netplan" "$@" bash "$sb/helper.sh" && rc=0 || rc=$?
}
a_is() { [ "$2" = "$3" ] && ok "apply helper $1" || bad "apply helper $1 (got '$2', want '$3')"; }
st_word() { cut -d' ' -f1 "$sb/status" 2>/dev/null; }
# (1) success
an_run
a_is "success: exit 0" "$rc" 0; a_is "success: status" "$(st_word)" ok
grep -q 'apply-network OK' "$sb/log" && ok "apply helper success: log OK" || bad "apply helper success: log"
a_is "success: 50-van-wan.yaml mode" "$(stat -c %a "$sb/netplan/50-van-wan.yaml" 2>/dev/null)" 600
[ -e "$sb/netplan/10-dhcp-all-interfaces.yaml.vanlink-disabled" ] && [ ! -e "$sb/netplan/10-dhcp-all-interfaces.yaml" ] && ok "apply helper success: Armbian file moved aside" || bad "apply helper success: Armbian file not moved aside"
grep -q '^systemctl restart van-failover' "$sb/calls" && ok "apply helper success: van-failover restarted" || bad "apply helper success: no van-failover restart"
rm -rf "$sb"
# (2) stage 1 fails: connected (externally)
an_run NM_STATE="100 (connected (externally))"
a_is "stage1: exit 1" "$rc" 1; a_is "stage1: status" "$(st_word)" rolled-back
grep -q 'ROLLED BACK (restored, connectivity ok)' "$sb/log" && ok "apply helper stage1: log ROLLED BACK" || bad "apply helper stage1: log"
a_is "stage1: Armbian file restored mode" "$(stat -c %a "$sb/netplan/10-dhcp-all-interfaces.yaml" 2>/dev/null)" 600
[ ! -e "$sb/netplan/50-van-wan.yaml" ] && [ ! -e "$sb/netplan/10-dhcp-all-interfaces.yaml.vanlink-disabled" ] && [ -e "$sb/netplan/60-other.yaml" ] && ok "apply helper stage1: netplan dir restored wholesale" || bad "apply helper stage1: netplan dir not restored"
rm -rf "$sb"
# (2b) stage 1 fails: wrong connection name; gateway unpingable
an_run NM_CONN="Wired connection 1"; a_is "stage1 conn name: status" "$(st_word)" rolled-back; rm -rf "$sb"
an_run STUB_GW_PING=1; a_is "stage1 gateway ping: status" "$(st_word)" rolled-back; rm -rf "$sb"
an_run NM_GW=; a_is "stage1 no gateway: status" "$(st_word)" rolled-back; rm -rf "$sb"
# (3) netplan apply fails, port absent
AN_NOPORT=1 an_run STUB_FAIL_APPLY=first
a_is "apply failure: exit 1" "$rc" 1; a_is "apply failure: not ok" "$(st_word)" rolled-back
[ -e "$sb/netplan/10-dhcp-all-interfaces.yaml" ] && [ ! -e "$sb/netplan/50-van-wan.yaml" ] && ok "apply helper apply failure: restored" || bad "apply helper apply failure: not restored"
rm -rf "$sb"
# (3b) port absent and apply ok: stage 1 needs only the apply
AN_NOPORT=1 an_run; a_is "port absent, apply ok: status" "$(st_word)" ok; rm -rf "$sb"
# (4) rollback cp fails (backup dir missing)
AN_NOBAK=1 an_run NM_STATE=bad
a_is "rollback cp failure: exit 1" "$rc" 1; a_is "rollback cp failure: status" "$(st_word)" rollback-failed
grep -q 'ROLLBACK FAILED' "$sb/log" && ok "apply helper rollback failure: loud log" || bad "apply helper rollback failure: log"
grep -q '^reboot\|systemctl reboot' "$sb/calls" && bad "apply helper rollback failure with connectivity: must not reboot" || ok "apply helper rollback failure with connectivity: no reboot"
rm -rf "$sb"
# (5) stage 2 fails: no default route until rolled back
an_run STUB_ROUTE=after-rollback
a_is "stage2: exit 1" "$rc" 1; a_is "stage2: status" "$(st_word)" rolled-back
[ ! -e "$sb/netplan/50-van-wan.yaml" ] && [ -e "$sb/netplan/10-dhcp-all-interfaces.yaml" ] && ok "apply helper stage2: restored" || bad "apply helper stage2: not restored"
rm -rf "$sb"
# (6) no connectivity even after rollback: last-resort reboot
an_run STUB_ROUTE=none
a_is "no connectivity: status" "$(st_word)" rollback-failed
grep -q '^systemctl reboot' "$sb/calls" && ok "apply helper: reboots when connectivity is not back after rollback" || bad "apply helper: no reboot"
rm -rf "$sb"
rm -rf "$stubs"
fi fi
exit "$fail" exit "$fail"