Hi all, I have a task on setting up a cluster of 2 servers for providing ftp service. We use RH's Cluster suite. However the cluster setup is lack of shared network storage, so the /var/ftp/pub is mounted on each server. The cluster is set with Master and Slave configuration (when the master is down, slave serves. and when master is up again, slaves gives up the service back to master).
We thought of using rsync to synchronize the two directory on each server.
B -----> A with: rsync -avzgorp -e ssh root@serverB:/var/ftp/pub/ /var/ftp/pub/
A -----> B with: rsync -avzgorp --delete -e ssh /var/ftp/pub/ root@serverB:/var/ftp/pub/ (note: using --delete)
We want those command to make sure that the most current and complete /var/ftp/pub is on A and we put those command on A only.
So, are we on the right direction? Or is there any better way? Thank you very much,
Pls read the below link. It may also help you. Pls note that when you ssh to a serverB, try it as a normal user without a password.
What is the Master node? Server A or Server B?
http://www.howtoforge.com/mirroring_with_rsync
---------- Forwarded message ---------- From: Fajar Priyanto fajarpri@cbn.net.id Date: Jan 17, 2007 2:14 PM Subject: [CentOS] OOT: Advice on failover rsync setup To: CentOS list centos@centos.org
Hi all, I have a task on setting up a cluster of 2 servers for providing ftp service. We use RH's Cluster suite. However the cluster setup is lack of shared network storage, so the /var/ftp/pub is mounted on each server. The cluster is set with Master and Slave configuration (when the master is down, slave serves. and when master is up again, slaves gives up the service back to master).
We thought of using rsync to synchronize the two directory on each server.
B -----> A with: rsync -avzgorp -e ssh root@serverB:/var/ftp/pub/ /var/ftp/pub/
A -----> B with: rsync -avzgorp --delete -e ssh /var/ftp/pub/ root@serverB:/var/ftp/pub/ (note: using --delete)
We want those command to make sure that the most current and complete /var/ftp/pub is on A and we put those command on A only.
So, are we on the right direction? Or is there any better way? Thank you very much, -- Fajar Priyanto | Reg'd Linux User #327841 | Linux tutorial http://linux2.arinet.org 3:44pm up 6:53, 2.6.16.13-4-default GNU/Linux Let's use OpenOffice. http://www.openoffice.org
_______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
On Wednesday 17 January 2007 16:38, Indunil Jayasooriya wrote:
Pls read the below link. It may also help you. Pls note that when you ssh to a serverB, try it as a normal user without a password.
What is the Master node? Server A or Server B?
Hi Indunil, The master node is server A. Thank you.
Fajar Priyanto wrote:
I have a task on setting up a cluster of 2 servers for providing ftp service. We use RH's Cluster suite. However the cluster setup is lack of shared network storage, so the /var/ftp/pub is mounted on each server. The cluster is set
So, are we on the right direction? Or is there any better way? Thank you very much,
have you considered using drbd ?
On Wednesday 17 January 2007 04:41, Karanbir Singh wrote:
Fajar Priyanto wrote:
I have a task on setting up a cluster of 2 servers for providing ftp
service.
We use RH's Cluster suite. However the cluster setup is lack of shared
network
storage, so the /var/ftp/pub is mounted on each server. The cluster is set
So, are we on the right direction? Or is there any better way? Thank you very much,
have you considered using drbd ?
Well, I was going to suggest using RAID1 with one of the partitions being a network block device, but this looks even better!
Rsync is great for period backups (not more than once per day or so) but if you're already looking into clustering, HA and/performance issues are already a concern, so it's just not really up to the task. Rsync takes a terrible performance hit, unmanageable on a busy system if it's IDE-based.
It's great, though, for Distaster Recovery off-site backups. See http://www.effortlessis.com/backupbuddy for an example I cooked up some time back.
Go with DRBD, but I've seen RAID1 on a virtual network block device work reasonably well if both systems are on the same network. (I'd suggest putting the block device network traffic on its own dedicated hub/network separate from other traffic)
If I recall correctly, this is the network block device we used: (it was a while back) http://www.it.uc3m.es/~ptb/nbd/
-Ben
Karanbir Singh : http://www.karan.org/ : 2522219@icq _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.