From: jeff barratt-mccartney (jbarratt@COMPSAT.COM)
Date: Wed Mar 03 2004 - 19:22:20 EST
Is this synchronous or asynchronous PPRC? If it is asynch, then it is
operating as designed.
a lot of people use rsync to accomplish the same thing. it has it's issues
as well.
-----Original Message-----
From: IBM AIX Discussion List [mailto:aix-l@Princeton.EDU]On Behalf Of
Jason delaFuente
Sent: Wednesday, March 03, 2004 3:35 PM
To: aix-l@Princeton.EDU
Subject: NFS/AFS/PPRC ?
It seems a number of you have Sharks so I thought this would be a good place
to check...
We currently have two F20 ESS Disk Subsystems in separate geographic
locations. We have about 4 16GB LUNS that we have setup to synchronize
through ESS Copy Services. All of these LUNS are in one Volume Group and
make up one filesystem. We have PPRC setup to only sync track differences.
The filesystem that resides on these disks is where our DB duplicate archive
logs get written to. When FilesystemA at LocationA fills up to a certain
point we break the pprc link and mount the "copy" of FilesystemA at
LocationB. From that point we do some oracle work and delete the files on
the source drives when we are finished with them. The last thing we do is
restart the sync.
The problem that we are having is that occasionally some files do not get
copied over. You can see the file on the source filesystem but when you try
to sync it Copy Services does not recognize it as a track difference even
though the file is not there when you 'ls' the destination side.
My first question is...Has anyone else ran into similar issues using Copy
Services.
My second question....I am investigating the possibility of using an NFS
type solution to do the same thing. I understand that NFS over the WAN is
not secure. I have heard that AFS with Kerberos is a secure alternative but
difficult to setup. Has anyone used this type of setup or know of another
alternative for your basic NFS?
Thanks!
Jason de la Fuente
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 22:17:39 EDT