View Source

Gestion du module

Sur réception du signal SIGUSR1 le module va effectuer un dump de sa mémoire, pour tout autre signal, le module va s'éteindre :

[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ MANAGE SIGNAL ] The worker with the pid XXXX received a signal XX

Arrêt critique

Quand le processus de pilotage s'arrête de façon inopinée :

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ WORKER:XXXX  ] I am a worker with pid: XXXX and my master process YYYY is dead, I exit.

Demande d'un dump de la mémoire

Le dump est fait

Python 2.6

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ WORKER pid=XXXX ] (support-only) MEMORY DUMP (to be sent to the support):
xxxxxxxx
xxxxxxxx
xxxxxxxx

Python 2.7

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ WORKER pid=XXXX ] (support-only) Memory information dumped to file FFFFFFF (to be sent to the support)

Le dump a échoué

Python 2.6

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ WORKER pid=XXXX ] MEMORY DUMP: FAIL check if guppy lib is installed

Python 2.7

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ WORKER pid=XXXX ] (support-only) MEMORY DUMP: FAIL check if meliae lib is installed

Connexion à la base de données

Pour la connexion à la base de données, trois SOUS-SECTION existent:

Section	Description
LOAD RETENTION	Correspond au chargement de la rétention
DELETE OLD RETENTION	Correspond à la suppression des anciennes rétentions
SAVE WORKER XXXX	Correspond aux workers utilisés pour la sauvegarde en rétention

Connexion normale

[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ SOUS-SECTION ] We are creating mongo connection [uri=mongodb://192.168.1.120/?safe=false] [database=shinken] [ssh=True]
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ SOUS-SECTION ] Connection created in : 0.200s

Il y indique :

L'URL utilisée
La base de données (peut être différente du défaut "shinken" comme ici)
Si un tunnel SSH va être utilisé ou pas
Le temps prit pour se connecter à la base mongo

La connexion échoue

[YYYY-MM-DD HH:MM:SS] WARNING: [ SCHEDULERNAME ] [ MongodbRetention ] [ SOUS-SECTION ] Mongo connection failed 1/X time, we will try again
[YYYY-MM-DD HH:MM:SS] WARNING: [ SCHEDULERNAME ] [ MongodbRetention ] [ SOUS-SECTION ] Mongo connection failed Y/X times, we will try again
[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ SOUS-SECTION ] Mongo connection failed X/X times, we stop trying

La connexion a été perdue ou n'existe pas

[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ SOUS-SECTION ] We need to create a mongo connection

La connexion n'a pas pu être établie

[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ SOUS-SECTION ] Could not create mongo connection

Erreur de configuration du module

Si plusieurs url mongo sont précisées

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MODULES-MANAGER ] The instance MongodbRetention raised an error: Multiples urls were found in the module's configuration file. I disable it and set it to restart it later

Sauvegarde en rétention

Pour la sauvegarde de la rétention, trois SOUS-SECTION existent:

Section	Description
SAVE GLOBAL	Correspond au processus global de la sauvegarde
SAVE WORKERS	Correspond à un sous-processus de SAVE GLOBAL, qui s'occupe de la file d'attente des différents workers de la sauvegarde
SAVE WORKER X	C'est un sous-processus de SAVE WORKERS, correspondant à un worker numéroté X qui permet de sauvegarder une partie des informations du scheduler en base. Le nombre de workers est paramétrable dans les paramètres du module. ( voir Rétention en base de données centralisée par royaume ( Module MongodbRetention ) )

SAVE GLOBAL

Les logs SAVE GLOBAL donnent des informations relatives au fonctionnement global du module ou de sa configuration.

Avant de faire la rétention, le module nous informe de l'URI utilisé ainsi que du nombre total d'hôtes et de checks à sauvegarder.

[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE GLOBAL      ] Starting to save retention data with X worker(s). [YY:hosts] [ZZ:checks] (Database used = mongodb://127.0.0.1/?safe=false, use ssh = True/False)

Dans l'exemple :

X : Le nombre de workers lancés en parallèle pour effectuer la sauvegarde.
YY : Le nombre d'hôtes et clusters qui vont être sauvegardés.
ZZ : Le nombre de checks qui vont être sauvegardés.

[YYYY-MM-DD HH:MM:SS] INFO : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE GLOBAL ] SUCCESS Retention data was saved into mongodb. Total time X.XXs

Erreurs

Les erreurs lors de la sauvegarde de la rétention sont aussi enregistrées dans les logs sous cette forme:

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MODULES-MANAGER ] The instance MongodbRetention raised an error: ERROR MESSAGE. Total time XX.XXs. I disable it and set it to restart it later

Exemples

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MODULES-MANAGER  ] The instance MongodbRetention raised an error: [ SAVE GLOBAL ] FAILED Retention data could not be saved in mongodb. Total time 22.20s. I disable it and set it to restart it later

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MODULES-MANAGER  ] The instance MongodbRetention raised an error: [ SAVE GLOBAL ] FAILED Retention data could not be saved in mongodb because mongo is unreachable. Total time 2.11s. I disable it and set it to restart it later

SAVE WORKERS

Les logs SAVE WORKERS donnent l'état de chaque worker de sa création à son succès/échec.

[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKERS ] Starting worker X with pid XXXX. Try: X/X
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKERS ] The worker X did SUCCESS (after X try)

La préparation des données à sauvegarder a été longue :

[YYYY-MM-DD HH:MM:SS] WARNING: [ SCHEDULERNAME ] [ MongodbRetention ] [ PERF ] [ X.XXXs ] atomization duration

Des erreurs empêchent le bon déroulé de la sauvegarde :

[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKERS ] some workers did fail to exit or encountered an error. The retention save can be incomplete.

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKERS ] Too many tries failed

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKERS ] Cannot start the XXXX worker process as there is not enough memory

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKERS ] Cannot start the worker XXXX process: XX. Exiting the retention save, killing all currently launched workers

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKERS ] ERROR MESSAGE
[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKERS ] "EXCEPTION PYTHON"

SAVE WORKER X

Les logs SAVE WORKER X donne pour le worker ayant l'identifiant X, les statistiques sur les sauvegardes qu'il a effectuées.

[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKER X ] Updating retention with elements: checks [ XXX ] -- hosts [ XX ] in mongodb
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKER X ]  Retention data saved into mongodb in X.XXX seconds

Nous sommes donc informés de :

Du nombre d'éléments mis à jour dans la base ( nombre de checks et nombre d'hôtes )
Le temps que la sauvegarde a pris

Erreurs

[YYYY-MM-DD HH:MM:SS] WARNING: [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKER X ] The worker (pid:XXXX | try:XX) did not exit on time (XX s). We are restarting it.

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKER X ] Failed connection with the following message : ERROR MESSAGE

Perte de connexion à la base de données

[YYYY-MM-DD HH:MM:SS] WARNING: [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKER X ] worker has been disconnected of mongo. Will retry [1/X]
[YYYY-MM-DD HH:MM:SS] WARNING: [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKER X ] worker has been disconnected of mongo. Will retry [Y/X]
[YYYY-MM-DD HH:MM:SS] WARNING: [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKER X ] worker has been disconnected of mongo. Will retry [X/X]
[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKER X ] After X tries, worker could not connect to mongo :[ERROR MESSAGE]
[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKER X ] (pid=XXXX) "EXCEPTION PYTHON"

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKER X ] Worker has an error: [ ERROR MESSAGE ]
[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ SAVE WORKER X ] (pid=XXXX) "EXCEPTION PYTHON"

Chargement de la rétention

Les logs fournissent des informations liées au chargement de la rétention, permettant de suivre son avancée et l'état sur la connexion à Mongo.

[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ LOAD RETENTION ] [ HOSTS / CLUSTERS ] [ X.XXXs ] We took X    hosts/clusters  from the retention [ in scheduler hosts/clusters : without retention=X    / total=1    ]
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ LOAD RETENTION ] [ HOSTS / CLUSTERS ] No host/cluster are needed for retention load (scheduler already have all X    hosts retention data).
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ LOAD RETENTION ] [ CHECKS           ] [ X.XXXs ] We took X    checks          from the retention [ in scheduler checks         : without retention=XX   / total=XX   ]
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ LOAD RETENTION ] [ CHECKS           ] No checks       are needed for retention load (scheduler already have all X    checks retention data).
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ LOAD RETENTION ] [ X.XXXs]  Total number of elements load from mongo database: X    ( scheduler have a total of XX   elements )
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ LOAD RETENTION ] [ X.XXXs ] SUCCESS Retention data loaded successfully.

Erreurs

Les erreurs lors du chargement de la rétention sont aussi enregistrées dans les logs sous cette forme:

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ LOAD RETENTION ] FAILED Retention data could not be loaded from mongodb: ERROR MESSAGE DETAILS

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ LOAD RETENTION ] error querying host entries: ERROR MESSAGE. Module exiting.

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ LOAD RETENTION ] error querying checks entries: ERROR MESSAGE. Module exiting.

Suppression des anciennes rétentions

Les logs de suppression permettent de voir le nombre d'objets supprimés (triés par hôtes et checks) ainsi que la date à partir de laquelle la rétention est conservée.

[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION   ] We will delete all retention data that were saved before the XXXX-XX-XX XX:XX UTC (X days)
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION   ]  - Deleting XXX hosts from old retention [XXXX by XXXX]
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION   ]  - XXX  - hosts deleted in X.XXXs
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION   ]  - Deleting XXX services from old retention [XXXX by XXXX]
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION   ]  - XXX  - services deleted in X.XXXs
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION   ] Total time for deleting XXXX entries = X.XXXs

[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION   ] We will delete all retention data that were saved before the XXXX-XX-XX XX:XX UTC (X days)
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION   ]  - There is no data to delete
[YYYY-MM-DD HH:MM:SS] INFO   : [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION   ] Total time for deleting 0 entries = X.XXXs

Erreur : perte de connexion à la base de données

[YYYY-MM-DD HH:MM:SS] WARNING: [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION ] We have been disconnected of mongo. Will retry [1/3]
[YYYY-MM-DD HH:MM:SS] WARNING: [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION ] We have been disconnected of mongo. Will retry [2/3]
[YYYY-MM-DD HH:MM:SS] WARNING: [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION ] We have been disconnected of mongo. Will retry [3/3]
[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION ] After 3 tries, we couldn't connect to mongo

[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION ] We have an error:[ERROR MESSAGE]
[YYYY-MM-DD HH:MM:SS] ERROR  : [ SCHEDULERNAME ] [ MongodbRetention ] [ DELETE OLD RETENTION ] "EXCEPTION PYTHON"