Auto commit from /home/doc/genesis-tools

This commit is contained in:
DocTator 2025-04-28 15:03:18 -04:00
parent 4e4f7d797a
commit a7fee7cfee
3 changed files with 3793 additions and 37 deletions

File diff suppressed because it is too large Load Diff

View File

@ -392,3 +392,10 @@ OK: All checks passed.
f.write("<table border='1' cellpadding='5' style='border-collapse: collapse;'>
^
SyntaxError: unterminated string literal (detected at line 181)
Traceback (most recent call last):
File "/home/doc/genesis-tools/miscellaneous/dbcheck1.py", line 187, in <module>
main()
File "/home/doc/genesis-tools/miscellaneous/dbcheck1.py", line 113, in main
res = check_remote_disk(node["host"], node["ssh_user"], disk, node["name"])
^^^^^^^^^^^^^^^^^
NameError: name 'check_remote_disk' is not defined. Did you mean: 'check_remote_logs'?

View File

@ -3,6 +3,7 @@ import requests
import datetime
import paramiko
import time
import psycopg2
# ==== CONFIG ====
MASTODON_INSTANCE = "https://chatwithus.live"
@ -17,13 +18,13 @@ LOG_PATTERNS = ["ERROR", "FATAL", "disk full", "out of memory"]
SUPPRESSED_PATTERNS = ["SomeKnownHarmlessMastodonError"]
NODES = [
{"name": "shredder", "host": "38.102.127.171", "ssh_user": "doc", "services": ["minio.service"], "disks": ["/", "/mnt/raid5"], "type": "remote", "db": False, "raid": True},
{"name": "mastodon", "host": "chatwithus.live", "ssh_user": "root", "services": ["nginx", "mastodon-web"], "disks": ["/"], "type": "remote", "db": False, "raid": False},
{"name": "db1", "host": "cluster.db1.genesishostingtechnologies.com", "ssh_user": "doc", "services": ["postgresql@16-main.service"], "disks": ["/", "/var/lib/postgresql"], "type": "remote", "db": True, "raid": False},
{"name": "db2", "host": "cluster.db2.genesishostingtechnologies.com", "ssh_user": "doc", "services": ["postgresql@16-main.service"], "disks": ["/", "/var/lib/postgresql"], "type": "remote", "db": True, "raid": False}
{"name": "shredder", "host": "38.102.127.171", "ssh_user": "doc", "services": ["minio.service"], "disks": ["/", "/mnt/raid5"], "db": False, "raid": True},
{"name": "mastodon", "host": "chatwithus.live", "ssh_user": "root", "services": ["nginx", "mastodon-web"], "disks": ["/"], "db": False, "raid": False},
{"name": "db1", "host": "cluster.db1.genesishostingtechnologies.com", "ssh_user": "doc", "services": ["postgresql@16-main.service"], "disks": ["/", "/var/lib/postgresql"], "db": True, "raid": False},
{"name": "db2", "host": "cluster.db2.genesishostingtechnologies.com", "ssh_user": "doc", "services": ["postgresql@16-main.service"], "disks": ["/", "/var/lib/postgresql"], "db": True, "raid": False}
]
# ==== Mastodon DM with retry ====
# ==== Mastodon DM ====
def mastodon_dm(message, retries=3):
url = f"{MASTODON_INSTANCE}/api/v1/statuses"
headers = {"Authorization": f"Bearer {MASTODON_TOKEN}"}
@ -35,7 +36,7 @@ def mastodon_dm(message, retries=3):
print(f"Failed to send Mastodon DM (attempt {attempt+1}): {resp.text}")
time.sleep(5)
# ==== SSH command runner ====
# ==== SSH Helper ====
def ssh_command(host, user, cmd):
ssh = paramiko.SSHClient()
ssh.set_missing_host_key_policy(paramiko.AutoAddPolicy())
@ -45,7 +46,7 @@ def ssh_command(host, user, cmd):
ssh.close()
return out
# ==== Emoji chooser ====
# ==== Health Check Helpers ====
def choose_emoji(line):
if "RAID" in line:
return "🧨"
@ -59,18 +60,69 @@ def choose_emoji(line):
return "💥"
return "⚠️"
# ==== Health checks ====
def check_rclone_health(node):
def check_remote_disk(host, user, path, node_name):
try:
result = ssh_command(node["host"], node["ssh_user"], "rclone rc vfs/stats")
if "error" in result.lower() or "failed" in result.lower():
return ("critical", f"[{node['name']}] ERROR: rclone health check failed. Output: {result}")
if "bytesUsed" in result:
bytes_used = int(result.split('"bytesUsed":')[1].split(',')[0].strip())
if bytes_used > 100000000000:
return ("warning", f"[{node['name']}] WARNING: rclone cache usage high: {bytes_used} bytes used.")
cmd = f"df --output=pcent {path} | tail -1 | tr -dc '0-9'"
out = ssh_command(host, user, cmd)
if not out:
return f"[{node_name}] ERROR: Disk {path} not found or could not check disk usage."
percent = int(out)
if percent > (100 - DISK_WARN_THRESHOLD):
return f"[{node_name}] WARNING: Only {100 - percent}% disk free on {path}."
except Exception as e:
return ("critical", f"[{node['name']}] ERROR: Could not check rclone health: {str(e)}")
return f"[{node_name}] ERROR: Disk check failed: {e}"
return None
def check_remote_service(host, user, service, node_name):
try:
cmd = f"systemctl is-active {service}"
out = ssh_command(host, user, cmd)
if out.strip() != "active":
return f"[{node_name}] CRITICAL: Service {service} not running!"
except Exception as e:
return f"[{node_name}] ERROR: Service check failed: {e}"
return None
def check_replication(host, node_name):
try:
conn = psycopg2.connect(host=host, dbname="postgres", user="postgres", connect_timeout=5)
cur = conn.cursor()
cur.execute("SELECT pg_is_in_recovery();")
is_replica = cur.fetchone()[0]
if is_replica:
cur.execute("SELECT EXTRACT(EPOCH FROM (now() - pg_last_xact_replay_timestamp()))::INT;")
lag = cur.fetchone()[0]
if lag is None:
return f"[{node_name}] CRITICAL: Standby not streaming! Replication down."
elif lag >= 60:
return f"[{node_name}] WARNING: Replication lag is {lag} seconds."
cur.close()
conn.close()
except Exception as e:
return f"[{node_name}] ERROR: Replication check failed: {e}"
return None
def check_remote_raid_md0(host, user, node_name):
try:
mdstat = ssh_command(host, user, "cat /proc/mdstat")
lines = mdstat.splitlines()
status = None
inside_md0 = False
for line in lines:
if line.startswith("md0"):
inside_md0 = True
elif inside_md0:
if "[" in line and "]" in line:
status = line[line.index("["):line.index("]")+1]
break
if line.strip() == "" or ":" in line:
break
if status is None:
return f"[{node_name}] CRITICAL: /dev/md0 RAID status string not found!"
if "_" in status:
return f"[{node_name}] WARNING: /dev/md0 RAID degraded! Status: {status}"
except Exception as e:
return f"[{node_name}] ERROR: RAID check failed: {e}"
return None
def check_remote_logs(host, user, node_name):
@ -88,7 +140,7 @@ def check_remote_logs(host, user, node_name):
alerts.append(f"[{node_name}] ERROR: Could not read log {log}: {e}")
return alerts
# ==== Main routine ====
# ==== Main Routine ====
def main():
critical_problems = []
warning_problems = []
@ -97,18 +149,6 @@ def main():
for node in NODES:
status = "✅ Healthy"
if "rclone" in node.get("services", []):
res = check_rclone_health(node)
if res:
level, msg = res
if level == "critical":
critical_problems.append(msg)
status = "🚨 Critical"
else:
warning_problems.append(msg)
if status != "🚨 Critical":
status = "⚠️ Warning"
for disk in node["disks"]:
res = check_remote_disk(node["host"], node["ssh_user"], disk, node["name"])
if res:
@ -132,8 +172,13 @@ def main():
if node.get("db"):
res = check_replication(node["host"], node["name"])
if res:
if "CRITICAL" in res:
critical_problems.append(res)
status = "🚨 Critical"
else:
warning_problems.append(res)
if status != "🚨 Critical":
status = "⚠️ Warning"
if node.get("raid", False):
res = check_remote_raid_md0(node["host"], node["ssh_user"], node["name"])
@ -141,13 +186,14 @@ def main():
if "CRITICAL" in res:
critical_problems.append(res)
status = "🚨 Critical"
elif status != "🚨 Critical":
else:
warning_problems.append(res)
if status != "🚨 Critical":
status = "⚠️ Warning"
logs = check_remote_logs(node["host"], node["ssh_user"], node["name"])
if logs:
warning_problems.extend(logs)
for log_alert in logs:
warning_problems.append(log_alert)
if status != "🚨 Critical":
status = "⚠️ Warning"
@ -172,7 +218,7 @@ def main():
print(msg)
mastodon_dm(msg)
# Write healthcheck HTML dashboard
# Write dashboard
with open(HEALTHCHECK_HTML, "w") as f:
f.write("<html><head><title>Genesis Radio Healthcheck</title><meta http-equiv='refresh' content='60'></head><body>")
f.write(f"<h1>Genesis Radio System Health</h1>")