Versions Compared

Old Version 12

changes.mady.by.user jtel Support

Saved on

compared with

New Version Current

changes.mady.by.user jtel Support

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Content imported from a Scroll Translations translation file.
Sv translation
languageen


Warning
titleAdvanced Topic

Warning - this is an advanced topic. Always backup your data first.

Preparation

Increase the size of the existing disk on both machines by EXACTLY the same size in the virtual environment.

The commands below assume that DRBD is running on /dev/sdb.

Make sure parted is current

Translations Ignore


Code Block
# Update parted
yum update parted



Rescan the SCSI Bus to display the drive

Sometimes the system will not see the new disk immediately. To achieve this without a reboot, run the following command to rescan the SCSI bus.

Translations Ignore


Code Block
titleRescan SCSI Bus
# Check the size of the partition
fdisk -l /dev/sdb

# Expected output - should show the increased size of the disk in the first line
Disk /dev/sdb: 536.9 GB, 536870912000 bytes, 1048576000 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk label type: dos
Disk identifier: 0x00000000
 
# If not, rescan SCSI and repeat above
echo 1 > /sys/block/sdb/device/rescan

# If the above command does not work, rescan SCSI busses individually
echo "- - -" > /sys/class/scsi_host/host0/scan
echo "- - -" > /sys/class/scsi_host/host1/scan
echo "- - -" > /sys/class/scsi_host/host2/scan



Make one node Secondary

Determine which node is servicing the SMB share with pcs status, and then on the non sevicing node:

Translations Ignore


Code Block
# Determine which node is servicing the share
pcs status
 
# Standby the server on which the resources are NOT running
pcs cluster standby acd-store2
 
# Run this command on the STANDBY machine
drbdadm secondary jtelshared



Increase Partition Size and verify - on the secondary node

Translations Ignore


Code Block
# Increase partition size
# Enter "Fix" then "Fix" after entering the command below
parted /dev/sdb print 

# Make sure this has worked
parted /dev/sdb print free

################################################
# DO NOT CONTINUE UNLESS THE FREE SPACE IS SHOWN
################################################

# Increase the partition size
parted /dev/sdb resizepart 1 100%

# Check
parted /dev/sdb print free



Increase Partition Size and verify - on the primary node

Repeat all steps above on the primary node, to increase the partition on the primary node.

Resize DRBD

On the PRIMARY node:

Translations Ignore


Code Block
# Resize DRBD
drbdadm resize jtelshared



Check, and speed up sync

Translations Ignore


Code Block
# Make sure sync is happening
cat /proc/drbd
 
# Speed up sync
drbdadm disk-options --c-plan-ahead=0 --resync-rate=110M jtelshared



When sync finished, resize the file system

This assumes, the DRBD was created with XFS, and is mounted at /dev/drbd1

Translations Ignore


Code Block
# Grow file system
xfs_growfs /dev/drbd1



Check the Results

Translations Ignore


Code Block
df -h



Untune sync and make secondary primary again

Translations Ignore


Code Block
# Untune sync
drbdadm adjust jtelshared

# Make secondary primary again
drbdadm primary jtelshared

# Unstandby the server on which the resources are NOT running in the PCS cluster
pcs cluster unstandby acd-store2




Sv translation
languagede


Warning
titleAdvanced Topic

Warning - this is an advanced topic. Always backup your data first.

Preparation

Increase the size of the existing disk on both machines by EXACTLY the same size in the virtual environment.

The commands below assume that DRBD is running on /dev/sdb.

Make sure parted is current

Translations Ignore


Code Block
# Update parted
yum update parted



Rescan the SCSI Bus to display the drive

Sometimes the system will not see the new disk immediately. To achieve this without a reboot, run the following command to rescan the SCSI bus.

Translations Ignore


Code Block
titleRescan SCSI Bus
# Check the size of the partition
fdisk -l /dev/sdb

# Expected output - should show the increased size of the disk in the first line
Disk /dev/sdb: 536.9 GB, 536870912000 bytes, 1048576000 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk label type: dos
Disk identifier: 0x00000000
 
# If not, rescan SCSI and repeat above
echo 1 > /sys/block/sdb/device/rescan

# If the above command does not work, rescan SCSI busses individually
echo "- - -" > /sys/class/scsi_host/host0/scan
echo "- - -" > /sys/class/scsi_host/host1/scan
echo "- - -" > /sys/class/scsi_host/host2/scan



Make one node Secondary

Determine which node is servicing the SMB share with pcs status, and then on the non sevicing node:

Translations Ignore


Code Block
# Determine which node is servicing the share
pcs status
 
# Standby the server on which the resources are NOT running
pcs cluster standby acd-store2
 
# Run this command on the STANDBY machine
drbdadm secondary jtelshared



Increase Partition Size and verify - on the secondary node

Translations Ignore


Code Block
# Increase partition size
# Enter "Fix" then "Fix" after entering the command below
parted /dev/sdb print 

# Make sure this has worked
parted /dev/sdb print free

################################################
# DO NOT CONTINUE UNLESS THE FREE SPACE IS SHOWN
################################################

# Increase the partition size
parted /dev/sdb resizepart 1 100%

# Check
parted /dev/sdb print free



Increase Partition Size and verify - on the primary node

Repeat all steps above on the primary node, to increase the partition on the primary node.

Resize DRBD

On the PRIMARY node:

Translations Ignore


Code Block
# Resize DRBD
drbdadm resize jtelshared



Check, and speed up sync

Translations Ignore


Code Block
# Make sure sync is happening
cat /proc/drbd
 
# Speed up sync
drbdadm disk-options --c-plan-ahead=0 --resync-rate=110M jtelshared



When sync finished, resize the file system

This assumes, the DRBD was created with XFS, and is mounted at /dev/drbd1

Translations Ignore


Code Block
# Grow file system
xfs_growfs /dev/drbd1



Check the Results

Translations Ignore


Code Block
df -h



Untune sync and make secondary primary again

Translations Ignore


Code Block
# Untune sync
drbdadm adjust jtelshared

# Make secondary primary again
drbdadm primary jtelshared

# Unstandby the server on which the resources are NOT running in the PCS cluster
pcs cluster unstandby acd-store2




Sv translation
languagefr


Warning
titleSujet avancé

Attention - il s'agit d'un sujet avancé. Sauvegardez toujours vos données en premier.

Préparation

Augmenter la taille du disque existant sur les deux machines EXACTEMENT la même taille dans l'environnement virtuel.

Les commandes ci-dessous supposent que le DRBD fonctionne sur /dev/sdb.

Assurez-vous que la pièce est à jour

Translations Ignore


Code Block
# Mise à jour séparée yum mise à jour séparée



Rescanner le bus SCSI pour afficher le lecteur

Parfois, le système ne voit pas le nouveau disque immédiatement. Pour y parvenir sans redémarrer le système, exécutez la commande suivante pour ré-analyser le bus SCSI.

Translations Ignore


Code Block
titleRéanalyser le bus SCSI
# Vérifiez la taille de la partition fdisk -l /dev/sdb # Sortie attendue - devrait montrer la taille accrue du disque sur la première ligne Disque /dev/sdb : 536,9 Go, 536870912000 octets, 1048576000 secteurs Unités = secteurs de 1 * 512 = 512 octets Taille du secteur (logique/physique) : 512 octets / 512 octets Taille des E/S (minimum/optimale) : 512 octets / 512 octets Type d'étiquette de disque : dos Identificateur de disque : 0x00000000 # Sinon, rescanner le SCSI et répéter l'écho ci-dessus 1 > /sys/block/sdb/dev/rescan # Si la commande ci-dessus ne fonctionne pas, rescanner les bus SCSI individuellement écho "- - -" > /sys/class/scsi_host/host0/scan echo "- - -" > /sys/class/scsi_host/host1/scan echo "- - -" > /sys/class/scsi_host/host2/scan



Faire un nœud secondaire

Déterminer quel nœud dessert la part des PME avec statut des pcs, et ensuite sur le nœud de non service :

Translations Ignore


Code Block
# Déterminer quel nœud dessert le share pcs status # Mettre en veille le serveur sur lequel les ressources ne tournent PAS pcs cluster standby acd-store2 # Exécuter cette commande sur la machine STANDBY drbdadm secondary jtelshared



Augmenter la taille de la partition et vérifier - sur le nœud secondaire

Translations Ignore


Code Block
# Augmenter la taille de la partition # Entrez "Fix" puis "Fix" après avoir entré la commande ci-dessous parted /dev/sdb print # Vérifiez que cela a fonctionné parted /dev/sdb print free ################################################ # Ne continuez pas tant que l'espace libre n'est pas affiché ################################################ # Augmentez la taille de la partition parted /dev/sdb resizepart 1 100% # Vérifiez parted /dev/sdb print free



Augmenter la taille de la partition et vérifier - sur le nœud primaire

Répétez toutes les étapes ci-dessus sur le nœud primaire, pour augmenter la partition sur le nœud primaire.

Redimensionner DRBD

Sur le nœud PRIMAIRE :

Translations Ignore


Code Block
# Redimensionner DRBD drbdadm redimensionner jtelshared



Vérifiez et accélérez la synchronisation

Translations Ignore


Code Block
# Vérifier la synchronisation cat /proc/drbd # Accélérer la synchronisation drbdadm disk-options --c-plan-ahead=0 --resync-rate=110M jtelshared



Une fois la synchronisation terminée, redimensionnez le système de fichiers

Cela suppose que le DRBD a été créé avec XFS, et est monté à /dev/drbd1

Translations Ignore


Code Block
# Augmenter le système de fichiers xfs_growfs / dev / drbd1



Vérifiez les résultats

Translations Ignore


Code Block
df -h



Annuler sync et mettre à nouveau le secondaire primaire

Translations Ignore


Code Block
# Untune sync drbdadm adjust jtelshared # Rendre secondaire primaire à nouveau drbdadm primaire jtelshared # Unstandby le serveur sur lequel les ressources ne sont PAS exécutées dans le cluster PCS pcs pcs cluster unstandby acd-store2