Différences

Ci-dessous, les différences entre deux révisions de la page.

--- admin:infrastructure:backup [2021/01/02 17:47] – [Scripts de pre hooks] fpoulain
+++ admin:infrastructure:backup [2021/01/02 17:53] (Version actuelle) – [Configuration du monitoring] fpoulain
@@ Ligne 262: / Ligne 262: @@
 exit 0
+</code>
+==== Rsyslog ====
+<code conf /etc/rsyslog.d/borgmatic.conf>
+if $programname == 'borgmatic' then /var/log/borgmatic.log
+& stop
 </code>
 ==== Log rotate ====
-<code conf debian/logrotate>
+<code conf debian/borgmatic>
-/var/log/backup.log {
+/var/log/borgmatic.log
-        weekly
+{
-        rotate 52
+  rotate 6
-        compress
+  weekly
-        delaycompress
+  compress
-        missingok
+  missingok
-        notifempty
+  notifempty
-        create 644 root root
 }
 </code>
@@ Ligne 291: / Ligne 297: @@
 On a un script qui parse sur chaque machine le log de backup et qui est déployé par le paquet monitoring-plugins-chapril :
-<code python /usr/lib/nagios/plugins/check_backup>
+<code python /usr/lib/nagios/plugins/check_borgmatic>
-#!/usr/bin/env python
+#!/usr/bin/env python3
-# -*- encoding:utf8 -*-
-import datetime, os, re, locale
+import datetime, itertools, os, re
-today= datetime.datetime.now ()
+now = datetime.datetime.now(datetime.timezone.utc)
-max_backup_delay = datetime.timedelta (1, 7200)
+max_backup_delay = datetime.timedelta(1, 7200)
-def last_backup (log_file):
+def get_name(match):
-    with open(log_file) as s:
+    return match.group('name')
-        logs_ok = re.findall (r'^([ a-zéûA-Z:,0-9]*)( \(UTC\+0[12]00\))?\nReturned 0\n={30}', s.read (), re.MULTILINE)[-1][0]
-        print "Last backup : " + logs_ok
-        try:
-            return datetime.datetime.strptime (logs_ok, '%a %b %d %X %Z %Y')
-        except:
-            locale.setlocale(locale.LC_ALL, 'fr_FR.UTF-8')
-            return datetime.datetime.strptime (logs_ok, '%A %d %B %Y, %X')
+def check_backup(filename):
+    with open(filename) as f:
+        logs = f.read()
+        mixed_statuses = list(re.finditer(r'(?P<status>Succeeded|Failed) (?P<name>\w+) backup at (?P<date>\d\d\d\d-\d\d-\d\dT\d\d:\d\d:\d\d\+\d\d:\d\d)$', logs, re.MULTILINE))
+        for name, statuses in itertools.groupby(sorted(mixed_statuses, key=get_name), key=get_name):
+            last = sorted(statuses, key=lambda x: x.group('date'))[-1]
+            print('{name}: {status} at {date}'.format(**last.groupdict()))
+            last_date = datetime.datetime.fromisoformat(last.group('date'))
+            last_status = last.group('status')
+            if last_status != 'Succeeded' or now - last_date > max_backup_delay:
+                failure.append(name)
+failure = []
 try:
-    last_backup_date= last_backup ("/var/log/backup.log")
+    check_backup ("/var/log/borgmatic.log")
-except:
+except Exception:
-    last_backup_date= last_backup ("/var/log/backup.log.1")
+    check_backup ("/var/log/borgmatic.log.1")
-if today - last_backup_date < max_backup_delay:
+if failure:
-    exit (0)
-else:
     exit (1)
+else:
+    exit (0)
 </code>
@@ Ligne 324: / Ligne 335: @@
 <code conf __admin__/etc/icinga2/zones.d/global-templates/services/backups.conf>
 object CheckCommand "backup" {
-	command = [ PluginDir + "/check_backup" ]
+	command = [ "sudo", PluginDir + "/check_borgmatic" ]
 }
@@ Ligne 336: / Ligne 347: @@
 }
 </code>
 ===== Aspects contrôle d'intégrité =====