Need help for data transfer

[praveen..]

Hi,
I have two servers.
1. HP-UX B.11.23
2. RHEL AS4 Server

I need to transefer the data (near about 1000GB) from HP server to IBM server means server no.1 to server no.2

these servers are in in different -different networks means IPs are of diff. diff. subnet

what will be the best method to move the data from server 1 to 2.

scp, ftp, NFS configuration or any other method: which one should i choose?

what will be the steps to configure the NFS in these two servers?

Please suggest

[spex]

praveen,

It really depends on your particular network configuration. Of the three options you gave, FTP is probably the easiest to set up. Just make sure you transfer the data in chunks, so that if an error occurs, you don't have to start over from the beginning. If you are transferring many files, you should satisfy this requirement. However, if you're transferring only a few large files, consider using 'split' to split them into smaller pieces before transfer. Then 'cat' them back together on the other end.

PCS

[spex]

praveen,

Also consider using 'cksum' on both sides to catch any data corruption.

PCS

[Jaime Bolanos Rojas.]

Praven, also you might want to look at Storage Replicator or volume replicator from Veritas, I know they work on hp-ux and ibm servers.

You might want to check the site

http://www.symantec.com/enterprise/products/index.jsp

Regards,

Jaime.

[Peter Godron]

Praveen,
if you want to use NFS:
1. Set up one server as host by exporting the filesystem
2. Set up the second server as target by importing the filesystem.
Then use the normal copy command.

See man nfs as a starting point, or use SAM

P.S.: How is your preparation going for CSA exam ?
( http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=1042847 )

[Ivan Ferreira · Accepted solution]

ftp is not recommended because you cannot transfer directories.

scp could do the job but it adds an overhead to the transfer because the encription of the data.

NFS probably is your best option if you have enough access in the routers/firewalls.

Maybe a backup/restore to tape device could be faster.

[praveen..]

I am trying to configure the NFS, but when i give the # mount server_ip:/server_directory /client_directory

it takes too much time and could not completed and gives this error:

nfs mount: get_fh: server_ip:: RPC: Rpcbind failure - RPC: Timed out


and when i try to start the service, this process gets hanged.

# /sbin/init.d/nfs.server start
starting NFS SERVER networking

starting up the rpcbind daemon
rpcbind already started, using pid: 1038
Reading in /etc/exports
starting up the mount daemon
rpc.mountd already started, using pid: 2048
starting up the NFS daemons
nfsd(s) already started, using pid(s): 2064 2063 2061 2065 2059 2060 2066 2068 2070 2072 2073 2074 2075 2076 2077 2078 2079
starting up the Status Monitor daemon
rpc.statd already started, using pid: 1081
starting up the Lock Manager daemon
rpc.lockd already started, using pid: 1087

::::::
::::::
HANG

[Ivan Ferreira]

Please post the results of the following command:

From the NFS server itselv

rpcinfo -p
showmount -e


From the NFS client

rpcinfo -p servername_or_ip_address
showmount -e servername_or_ip_address

[praveen..]

NFS Server:

# rpcinfo -p
program vers proto port
100000 2 tcp 111 portmapper
100000 2 udp 111 portmapper
100024 1 udp 32769 status
100024 1 tcp 32768 status
100007 2 udp 701 ypbind
100007 1 udp 701 ypbind
100007 2 tcp 704 ypbind
100007 1 tcp 704 ypbind
100011 1 udp 787 rquotad
100011 2 udp 787 rquotad
100011 1 tcp 790 rquotad
100011 2 tcp 790 rquotad
100003 2 udp 2049 nfs
100003 3 udp 2049 nfs
100003 4 udp 2049 nfs
100003 2 tcp 2049 nfs
100003 3 tcp 2049 nfs
100003 4 tcp 2049 nfs
100021 1 udp 32779 nlockmgr
100021 3 udp 32779 nlockmgr
100021 4 udp 32779 nlockmgr
100021 1 tcp 32911 nlockmgr
100021 3 tcp 32911 nlockmgr
100021 4 tcp 32911 nlockmgr
100005 1 udp 803 mountd
100005 1 tcp 806 mountd
100005 2 udp 803 mountd
100005 2 tcp 806 mountd
100005 3 udp 803 mountd
100005 3 tcp 806 mountd

# showmount -e
Export list for server_hostname:
/vipin_test client_ip
#



NFS Client:

# rpcinfo -p server_ip
rpcinfo: can't contact portmapper: RPC: Rpcbind failure - RPC: Failed (unspecified error)
# showmount -e server_ip
showmount: server_ip: RPC: Rpcbind failure - RPC: Timed out

[Ivan Ferreira]

From the output, I can see that the client cannot contact to the server, there should be some firewall blocking in the middle.

You should be able to telnet from the client to the server to the port 111:

telnet nfsserver 111


If not, stop the firewall on the linux server:

service iptables stop

And check another firewall/router configuration that may be blocking the access.

[praveen..]

NFS Server:

# service iptables status
Firewall is stopped.
# service iptables stop


NFS Client:

# telnet nfs_server 111
Trying...
telnet: Unable to connect to remote host: Connection timed out
#

[Ivan Ferreira]

I would like to know if there is another firewall on your network that may be blocking the connection. Maybe a filter rule in your router. Are you the network administrator also?

Please run from Red Hat Client:

nmap nfsserver

To check open ports. Can you access other services on this server? for example SSH?

[Ivan Ferreira]

As a last option, you may take your Red Hat server to the same network and do the data transfer ;).

[spex]

praveen,

Can you FTP to the server? If you're hesitant to use FTP because of a complex directory structure, tar the files (using relative paths). Then transfer the tar archive via FTP. You can pipe the tar through compress and split, if so desired.

If FTP doesn't work, is there another host which is mutually accessible from the client and server? If so, use it as an intermediary.

PCS