init

monitoring/management/commands/doctor.py

@@ -0,0 +1,35 @@
+"""Check configured infrastructure without changing production data."""
+import json,uuid
+from django.core.management.base import BaseCommand,CommandError
+from django.db import connection
+from django.db.migrations.executor import MigrationExecutor
+from django.core.cache import cache
+from config.celery import app
+
+class Command(BaseCommand):
+    help='Prüft Datenbank, Migrationen, Redis, Broker und optional Worker.'
+    def add_arguments(self,p):p.add_argument('--workers',action='store_true')
+    def handle(self,*args,**options):
+        results={}
+        try:
+            with connection.cursor() as c:c.execute('SELECT 1');c.fetchone()
+            results['database']=connection.vendor
+            executor=MigrationExecutor(connection)
+            pending=executor.migration_plan(executor.loader.graph.leaf_nodes())
+            results['migrations']='pending' if pending else 'ok'
+        except Exception as e:results['database']='ERROR '+type(e).__name__
+        try:
+            key='doctor:'+str(uuid.uuid4());cache.set(key,'ok',30)
+            results['cache']='ok' if cache.get(key)=='ok' else 'ERROR readback'
+            cache.delete(key)
+        except Exception as e:results['cache']='ERROR '+type(e).__name__
+        try:
+            with app.connection_for_read() as conn:conn.ensure_connection(max_retries=0,timeout=5)
+            results['broker']='ok'
+        except Exception as e:results['broker']='ERROR '+type(e).__name__
+        if options['workers']:
+            try:results['workers']='ok' if app.control.ping(timeout=3) else 'ERROR no response'
+            except Exception as e:results['workers']='ERROR '+type(e).__name__
+        self.stdout.write(json.dumps(results,indent=2))
+        if any(v.startswith('ERROR') or v=='pending' for v in results.values()):
+            raise CommandError('Infrastrukturprüfung fehlgeschlagen; keine Zugangsdaten werden ausgegeben.')

monitoring/management/commands/register_agent.py

@@ -0,0 +1,20 @@
+import secrets
+from django.core.management.base import BaseCommand,CommandError
+from django.db import transaction
+from monitoring.models import Segment,Server,AgentCredential,Audit
+from monitoring.auth import digest
+class Command(BaseCommand):
+    help='Register host or rotate its token; prints the secret once. Keep output private.'
+    def add_arguments(self,p):
+        p.add_argument('name');p.add_argument('--segment',required=True);p.add_argument('--address',default='');p.add_argument('--rotate',action='store_true')
+    @transaction.atomic
+    def handle(self,*args,**o):
+        seg,_=Segment.objects.get_or_create(name=o['segment'])
+        s,created=Server.objects.get_or_create(name=o['name'],defaults={'segment':seg,'address':o['address']})
+        if not created and not o['rotate']: raise CommandError('Already registered; use --rotate to revoke old credentials')
+        AgentCredential.objects.filter(server=s).update(active=False)
+        token=secrets.token_urlsafe(48)
+        AgentCredential.objects.create(server=s,digest=digest(token))
+        Audit.objects.create(actor='management-command',action='register_agent',details={'server':str(s.pk),'rotated':not created})
+        self.stdout.write('SERVER_ID='+str(s.pk))
+        self.stdout.write('AGENT_TOKEN='+token)

monitoring/management/commands/seed_demo.py

@@ -0,0 +1,23 @@
+import uuid
+from datetime import timedelta
+from django.core.management.base import BaseCommand,CommandError
+from django.utils import timezone
+from monitoring.models import Segment,Server,Event
+from monitoring.ingest import ingest
+from monitoring.tasks import evaluate_alarms
+class Command(BaseCommand):
+    help='Create isolated, explicitly named demo servers. Never deletes real data.'
+    def handle(self,*args,**kwargs):
+        seg,_=Segment.objects.get_or_create(name='DEMO · Produktion',defaults={'cidr':'Demo-Netz'})
+        for i,name in enumerate(['DEMO-django-01','DEMO-celery-01','DEMO-ai-01']):
+            s,created=Server.objects.get_or_create(name=name,defaults={'segment':seg,'address':'Demodaten'})
+            if not created: continue
+            now=timezone.now();batch=[]
+            for h in range(24):
+                at=now-timedelta(hours=23-h)
+                def event(kind,p):return {'id':uuid.uuid4(),'kind':kind,'occurred_at':at,'payload':p}
+                batch.append(event('metric',{'cpu':89 if i==1 else 25+h%15,'ram':52+i*8,'disk':40+i*5,'services':{'django':True}}))
+                batch.append(event('task-succeeded',{'uuid':str(uuid.uuid4()),'name':'demo.documents.index','hostname':name,'runtime':1.5+i}))
+            ingest(s.pk,batch)
+        evaluate_alarms()
+        self.stdout.write('Demo servers created. Login with a separately created user.')