Hello,
Your experience is similar to our, we have implemented all 3 technologies
with Netapp.
1. FC
2. ISCSI
3. CIFS
Time and time again, we had to raise volume size due to increased use and
it's easy as it can get, we also have VMWARE luns on Netapp and it has
scalled\performed well.
I have not seen any performance bottlenecks so far, we are getting close
to filling up our filer and the only disadvantage is you have to keep
adding storage, cluster keeps filer A going while Filer B is being
services, i have configured teaming on Network as well as ISCSI
connections for better throughput with GB links, virus software and
scanning is the most problematic area as the symentec product we have does
not seems to be living up to it's reputation.
We have a odd failure with onboard NIC failure and for that support has
been arranged to swap out the MB.
We updated the OS recently with Disk firmware and that went without any
issues.
We are utilizing Snapshot in very limited fashion due to possibility of
out of control snapshot issues, applications hosted so far include Lotus
Notes Domino server that is performing a lot better after the local
storage was moved to Netapp FC drives via ISCSI, i utilize Jumbo Frame
private vlan to host the ISCSI on Cisco switches.
As far as Networker and backups, we use NDMP for individual CIFS and
backup luns on the servers where it's assigned, NDMP backups have been
very stable and provide level backups with ability to do file level
restore.Configuring NDMP initially was a challenge, now with the knowledge
gained its not a problem.Typical NDMP backups average apx 60-90 Mb\m on
FAS3050C, we use VTL as well as LTO2. There is over the Network NDMP
backup method that i have tested and worked well as well as provide
multiplexing, i currently do not use that method as we have FC attached
drives to the Filer.
One thing you have to be careful about is restore, as NDMP restores can
only be directed to NDMP target and no matter what option you select,
restores overwrite existing file, to get around that i usually, only do
restores to alternate location on the filer( restore volume\share) , even
if you get the path wrong the data could end up in your root vol ( can be
deleted later, worse yet if you are not careful you can fill up your root
vol).This is a major problem and i hope future releases of NDMP filer OS
will correct the issue.
One other problem was with BINDVIEW , we used it to monitor permission's
etc on the CIFS shares, for some reason that functionality was lost due to
accidentally restoring something in the root vol ( as described above),
once that happened i can never get bindview to query the filer's cifs
permisons like i did before, however the alternative was to create a D$,
E$ shares for each CIFS share to allow BindView back into the MIX.
I have recovered and tested filer at DR and there has not been any
issues.
I use DFM ( NetApp Tool) to get some reporting etc, however, i do suggest
a automatic process should be developed that netapp lacks right now out of
box to upload and save current configurations for DR purpose.
We have had couple of application issues but with increased awareness
related to Netapp i have fixed them, i.e OPS Lock and Citrix, Max-MPX
connections, Failover settings & Antivirus headaches with Cab files
problems.
Preliminary support is addiquate and not decisive, i hope NetApp trains
their Level 1 tech's a little better.
HTH
To sign off this list, send email to listserv AT listserv.temple DOT edu and
type "signoff networker" in the body of the email. Please write to
networker-request AT listserv.temple DOT edu if you have any problems with this
list. You can access the archives at
http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER
|