Hi Ehren,
ok - there are so many other factors which can influence this, it is
currently hard to identify the root cause without more details around
setup/config....
When you had validated your cluster before you had configured cluster disks
and all tests was passed than we can exclude here any issues with
hardware/software/networking related miss configurations.
The validation wizard is really very smart and do check many many things for
you especially in the storage section like a.e. failover shared disks to
verify correct zoning....
If possible please take your CSV disk offline and re-run the validation test
(minimum sections network, storage, system) ?
How did you implemented the multiple ISCSI paths to your storage, MCS ?
Otherwise from networking side it looks ok - you could also specify a
designated network for the CSV communication. BUT normally this is not
required as the cluster service will find automatically the right/best path
to the storage ->
http://technet.microsoft.com/en-us/l...35(WS.10).aspx - If I
remember correctly the cluster service will use the network with the lowest
metric for CSV communication. The network metrics can be only checked with
powershell (Get-ClusterNetwork | ft Name, Metric, AutoMetric, Role).
You should also have a look on your cluster / event logs if there is
anything which can be a pointer.
a.e. Do you see any 5121,5136 or 5137 events which can be a pointer for
redirected access ?
More details can be found here ->
http://technet.microsoft.com/en-us/l...27(WS.10).aspx
Also check the following article to verify your networking environment which
is really important generally in clusters and especially as you're using
ISCSI here !
http://blogs.technet.com/askcore/arc...g-details.aspx
Hope that helps to show you ways to troubleshoot your current problem.
Please let me know about the results.
Regards
Ramazan
"Ehren" <> wrote in message
news:85BCC621-7EDA-4251-B755-...
> Yes those two protocols are enabled on the heartbeat and external network
> adapters on both nodes.
>
> I have 4 nics in the cluster, two of which are san nics and cluster
> communication is disabled. The other is the external nic which cluster
> communication and management connections are enabled on, the 4th is the
> heartbeat which cluster communication is enabled on but the management
> connections box is not checked since this network is not able to be
> connected
> to from anywhere other than these 2 nodes.
>
> Haha, i know everyone says it is supposed to work flawless and great but
> mine isnt and everything seems to be configured right, however if that was
> the case it would probably work 
>
> Any ideas?
>
> Thanks!
>
> "RCan" wrote:
>
>> Can you post the section with the warning please ?
>> I mean the protocols "client for microsoft networks" and "file and
>> printer
>> sharing" on "heartbeat" adapter, are they enabled now ?
>>
>> How many NICs do you have in your cluster and on which adapter do you
>> have
>> intra cluster communication enabled ?
>>
>> If your cluster is correctly configured the CSV access from non-owning
>> nodes
>> are really flawless
>>
>> Regards
>> Ramazan
>>
>> "Ehren" <> wrote in message
>> news:B8416DB9-86BE-44F3-8007-...
>> > Thanks for your replies guys.
>> >
>> > Yes I did follow the instructions and I can validate the cluster also.
>> > At
>> > first I was thinking it came down to SMB as the second poster drew out,
>> > and I
>> > do have those services enabled on all network adapters that are enabled
>> > for
>> > cluster communication. I actually did not have them enabled at first
>> > so I
>> > was hopeful enabling that would fix the issue, so I enabled those
>> > services
>> > on
>> > both the external and heartbeat network adapters and then actually
>> > destroyed
>> > the cluster and rebuilt it.
>> >
>> > I am able to validate the cluster configuration as well with only one
>> > warning which was a warning saying that one node could not access one
>> > of
>> > the
>> > NON CSV volumes I have that was not owned by that node (no surprise).
>> >
>> > So I am still lost as to why this is happening. THe CSV is NOT in
>> > redirected access or maintenance mode so there is no issue there. I
>> > can
>> > transfer ownership back and forth between nodes and access it on the
>> > node
>> > that owns it so there should be no problem with actual access to the
>> > volume.
>> >
>> > I notice that sometimes on the non-owning node if I go to the C drive
>> > and
>> > click on ClusterStorage and let it sit there for a few minutes it will
>> > have
>> > the hour glass on it for 5-7 minutes or so and then in the area where
>> > folders
>> > would be shown it says "This folder is empty." (where it should show
>> > "Volume1").
>> >
>> > Any further ideas, I am at a loss.
>> >
>> > Thanks!
>> >
>> > "RCan" wrote:
>> >
>> >> Additional interesting link around "Understanding redirected I/O mode
>> >> in
>> >> CSV
>> >> communication"
>> >> http://technet.microsoft.com/en-us/l...KMK_redirected
>> >>
>> >> Regards
>> >> Ramazan
>> >>
>> >> "Ehren" <> wrote in message
>> >> news:355A8BD7-D0BA-45FD-B197-...
>> >> > Hi,
>> >> >
>> >> > I have a question about a CSV I just set up to host hyper-v vm's for
>> >> > my
>> >> > two
>> >> > node Server 2008 R2 Datacenter Hyper-V cluster.
>> >> >
>> >> > I have the cluster all set up and onlined one of our SAN luns (via
>> >> > disk
>> >> > management) and formated it as NTFS without a drive letter or
>> >> > directory
>> >> > mapping and then set it back to offline. I then went into failover
>> >> > cluster
>> >> > manager on one of the nodes and added the volume to cluster storage,
>> >> > then
>> >> > went into the clustered shared volumes area and added the
>> >> > aforementioned
>> >> > storage as a CSV.
>> >> >
>> >> > My understanding is that both nodes should be able to access it
>> >> > simultaneously, however that does not seem to be the case. If I
>> >> > open
>> >> > failover cluster manager and go into cluster shared volumes and look
>> >> > under
>> >> > current owner for the disk in question, if for example NodeA is
>> >> > listed
>> >> > as
>> >> > the
>> >> > owner, that is the only one that can access it. If I go to
>> >> > C:\ClusterStorage\Volume1 on NodeA (the current owner) I can browse
>> >> > the
>> >> > volume just fine, however if I go to C:\ClusterStorage on Node2 It
>> >> > just
>> >> > brings up an hour glass and explorer stops responding.
>> >> >
>> >> > If I right click on the CSV Volume in Failover cluster manager and
>> >> > move
>> >> > the
>> >> > owner to Node2 then Node2 is able to access and browse it, and Node1
>> >> > then
>> >> > can
>> >> > not (brings up hour glass and explorer stops responding.)
>> >> >
>> >> > Did I do something wrong? Thanks for your insight!
>> >>
>> >> .
>> >>
>> .
>>