A
a_user
Hi All,
Not sure if I would consider this a unique situation but its not as straight
forward as I would like.
Currently we have two file file servers servering over 9tb. Storage1 and
storage2. Storage1 is on older hardware with windows 2000(about 3.5tb of
shares). I am trying to migrate this data to storage2 on windows 2003. We do
not currently use dfs but I am intending on implementing it slowly once I
have completed this migration.
My concern/issue is we have hundreds of shares on storage1 that obviously
reference 'storage1' and unfortunately there are even some internal
applications that are mapped to the unc path of storage1. This raises the
issue where simply changing the path in a logon script will accomodate the
users but it will break a lot of other things. I have taken over this
environment and would never base application based storage locations on a
particular server name which is why I want to go with DFS. But that is
neither here nor there right now.
My big question here is this, if I used FSMT to transfer all my shares from
storage1 to storage2 could I then power down/disconnect storage1 from the
network, remove its hostname from DNS and then create an alias of 'storage1'
and point that to storage2? I would then expected any clients looking to
storage1 would pull that same share off of storage2 instead. This is not the
ideal solution but the only one I can think of without introducing lots of
additional time and complexity. I need to ensure this change happens soon and
as seamlessly as possible.
I will then start educating and communicating on the use of DFS and add an
additional file server to the DFS environment for redundancy.
Anyone see anything wrong with this approach? I am just looking for some
reassurance that I am not overlooking something. I dont want to decommission
storage1 for a couple weeks after the transfer to ensure that everything has
been pulled successfully.
Thanks all
Not sure if I would consider this a unique situation but its not as straight
forward as I would like.
Currently we have two file file servers servering over 9tb. Storage1 and
storage2. Storage1 is on older hardware with windows 2000(about 3.5tb of
shares). I am trying to migrate this data to storage2 on windows 2003. We do
not currently use dfs but I am intending on implementing it slowly once I
have completed this migration.
My concern/issue is we have hundreds of shares on storage1 that obviously
reference 'storage1' and unfortunately there are even some internal
applications that are mapped to the unc path of storage1. This raises the
issue where simply changing the path in a logon script will accomodate the
users but it will break a lot of other things. I have taken over this
environment and would never base application based storage locations on a
particular server name which is why I want to go with DFS. But that is
neither here nor there right now.
My big question here is this, if I used FSMT to transfer all my shares from
storage1 to storage2 could I then power down/disconnect storage1 from the
network, remove its hostname from DNS and then create an alias of 'storage1'
and point that to storage2? I would then expected any clients looking to
storage1 would pull that same share off of storage2 instead. This is not the
ideal solution but the only one I can think of without introducing lots of
additional time and complexity. I need to ensure this change happens soon and
as seamlessly as possible.
I will then start educating and communicating on the use of DFS and add an
additional file server to the DFS environment for redundancy.
Anyone see anything wrong with this approach? I am just looking for some
reassurance that I am not overlooking something. I dont want to decommission
storage1 for a couple weeks after the transfer to ensure that everything has
been pulled successfully.
Thanks all