scheduler: add ping + ssh-script check kinds, metrics refactor, 17 host check scripts
- Refactor executeCheck to return checkResult struct with metrics map - Add ping check kind (ICMP reachability via system ping, macOS+Linux) - Add ssh-script check kind (remote host exec via SSH, allowlisted scripts) - Add threshold evaluation (warn/crit per metric from check config JSONB) - Add inode tracking to disk check - All 4 existing checks now return structured metrics - 17 check scripts: cpu, memory, load, swap, disk_usage, disk_smart, updates, zfs, process, uptime, oom, journal, time, fd, docker_health, caddy_error_rate, backup_freshness - Auto-deploy via tools/setup-checks.sh -> checks/install.sh on git pull - Add ping to OpenAPI CheckKind enum and generated Go types
This commit is contained in:
23
checks/backup_freshness.sh
Normal file
23
checks/backup_freshness.sh
Normal file
@@ -0,0 +1,23 @@
|
||||
#!/usr/bin/env bash
|
||||
# backup_freshness.sh — check that backups exist and are recent.
|
||||
set -euo pipefail
|
||||
|
||||
BACKUP_DIRS="${OIKOS_BACKUP_DIRS:-/var/backups /opt/backups /mnt/backups}"
|
||||
GRACE_HOURS="${OIKOS_BACKUP_GRACE:-48}"
|
||||
STALE=""
|
||||
|
||||
for dir in $BACKUP_DIRS; do
|
||||
[ -d "$dir" ] || continue
|
||||
NEWEST=$(find "$dir" -type f -mmin -$((GRACE_HOURS * 60)) 2>/dev/null | head -1 || true)
|
||||
if [ -z "$NEWEST" ]; then
|
||||
LATEST_TS=$(find "$dir" -type f -printf '%T@ %p\n' 2>/dev/null | sort -rn | head -1 | awk '{print $1}' || echo "0")
|
||||
LATEST_HOURS=$(awk "BEGIN {printf \"%.0f\", ($(date +%s) - ${LATEST_TS:-0})/3600}")
|
||||
STALE="$STALE $dir(${LATEST_HOURS}h)"
|
||||
fi
|
||||
done
|
||||
|
||||
if [ -n "$STALE" ]; then
|
||||
echo "{\"health\":\"degraded\",\"signalKind\":\"backup-stale\",\"evidence\":\"stale backups:$(echo "$STALE" | sed 's/ /, /g')\"}"
|
||||
else
|
||||
echo '{"health":"healthy"}'
|
||||
fi
|
||||
Reference in New Issue
Block a user