commande utiles glusterfs
This commit is contained in:
parent
6c2f9b1e85
commit
bc6803e5ac
|
@ -185,6 +185,7 @@
|
||||||
\include{mse/procedure/03-one-glusterfs-2}
|
\include{mse/procedure/03-one-glusterfs-2}
|
||||||
\include{mse/procedure/03-one-glusterfs-3}
|
\include{mse/procedure/03-one-glusterfs-3}
|
||||||
\include{mse/procedure/03-one-glusterfs-4}
|
\include{mse/procedure/03-one-glusterfs-4}
|
||||||
|
\include{mse/procedure/03-one-glusterfs-5}
|
||||||
|
|
||||||
\section{Le serveur de supervision}
|
\section{Le serveur de supervision}
|
||||||
\begin{frame}{Plan}
|
\begin{frame}{Plan}
|
||||||
|
|
|
@ -0,0 +1,55 @@
|
||||||
|
\begin{frame}
|
||||||
|
\frametitle{Commandes utiles ONE/Glusterfs}
|
||||||
|
\begin{itemize}
|
||||||
|
\item s'assurer de la cohérence du cluster GlusterFS ;
|
||||||
|
\begin{lstlisting}
|
||||||
|
gluster volume heal nomVolume (datastore0 pour MSE) info
|
||||||
|
\end{lstlisting}
|
||||||
|
\begin{lstlisting}
|
||||||
|
Launching heal operation to perform index self heal on volume datastore0 has been successful
|
||||||
|
Use heal info commands to check status.
|
||||||
|
root@mse-pp-worker-a1:~# gluster volume heal datastore0 info
|
||||||
|
Brick worker-a1.replication.lan:/mnt/glusterfs/datastore0
|
||||||
|
Status: Connected
|
||||||
|
Number of entries: 0
|
||||||
|
Brick worker-b1.replication.lan:/mnt/glusterfs/datastore0
|
||||||
|
Status: Connected
|
||||||
|
Number of entries: 0
|
||||||
|
\end{lstlisting}
|
||||||
|
\item Obtenir des information sur le fichier en split-brain :
|
||||||
|
\begin{lstlisting}
|
||||||
|
gluster volume heal datastore0 info split-brain
|
||||||
|
\end{lstlisting}
|
||||||
|
\item identifier le fichier pour lequel les opérations échouent
|
||||||
|
\item fermer les aplications du ficher en question si ce sont des VMs , mettre la vm hors tension
|
||||||
|
\item choisir la bonne copie :
|
||||||
|
\begin{itemize}
|
||||||
|
\item getfattr -d -m. -e hex <chemin-fichier-sur-brique>
|
||||||
|
\item Réinitialisez l'attribut étendu approprié sur la ou les briques contenant la «mauvaise copie» des données / métadonnées du fichier à l'aide de la commande setfattr.
|
||||||
|
\item setfattr -n <nom-attribut> -v <valeur-attribut> <chemin-fichier-sur-brick>
|
||||||
|
\item Déclenchez l'auto-réparation sur le fichier
|
||||||
|
\item ls -l <chemin-fichier-on-gluster-mount>
|
||||||
|
\end{itemize}
|
||||||
|
\item splibrain point de montage :
|
||||||
|
\item gluster volume info datastore0
|
||||||
|
\begin{lstlisting}
|
||||||
|
Volume Name: datastore0
|
||||||
|
Type: Replicate
|
||||||
|
Volume ID: 12f442ba-c6c5-4813-adc8-9f6e6b4966a9
|
||||||
|
Status: Started
|
||||||
|
Snapshot Count: 0
|
||||||
|
Number of Bricks: 1 x 2 = 2
|
||||||
|
Transport-type: tcp
|
||||||
|
Bricks:
|
||||||
|
Brick1: worker-a1.replication.lan:/mnt/glusterfs/datastore0
|
||||||
|
Brick2: worker-b1.replication.lan:/mnt/glusterfs/datastore0
|
||||||
|
Options Reconfigured:
|
||||||
|
performance.client-io-threads: off
|
||||||
|
nfs.disable: on
|
||||||
|
transport.address-family: inet
|
||||||
|
\end{lstlisting}
|
||||||
|
\item si Number of bricks est différent de bricks, redémarrer le service glusterfs sur la bricks non monté
|
||||||
|
\item systemctl status glusterd.service
|
||||||
|
\item systemctl start gusterd.service
|
||||||
|
\end{itemize}
|
||||||
|
\end{frame}
|
Loading…
Reference in New Issue