| Author |
Message |
Isaac Story
Guest
|
 Posted:
Fri Oct 21, 2005 12:50 am Post subject:
When adding 2nd node, unable to see shared drives |
|
|
I am fairly new to clustering and there is an issue I am having when adding
subsequent nodes after the first one while the first node is still up and
running. Due to the "shared nothing" model that restricts shared storage
access to one host, when I bring the second node up, the drives I need on the
SAN show up as unreadable. As a result, adding the second node fails because
it cannot see the shared quorum drive nor the shared data drive. I have seen
several people say this is normal behavior, yet it is preventing me from
setting up multi-node clustering. Am I missing something here???
My hardware setup is as follows:
IBM HS20 Blades connected to a Hitachi 9570V through a Brocade fabric. All
hosts are running Windows Server 2003 Enterprise. |
|
| Back to top |
|
 |
Isaac Story
Guest
|
| Posted:
Fri Oct 21, 2005 4:50 pm Post subject:
Re: When adding 2nd node, unable to see shared drives |
|
|
Is there a way to disable the "shared-nothing" disk protection until I have
clustering up and running? I don't any other way to get this to work.
Isaac
"John Toner [MVP]" wrote:
| Quote: | Have you seen this KB article?
http://support.microsoft.com/kb/331801
Regards,
John
"Isaac Story" <IsaacStory@discussions.microsoft.com> wrote in message
news:D33A2DA4-E1F0-434C-8E52-A7C3164A9D59@microsoft.com...
I am fairly new to clustering and there is an issue I am having when
adding
subsequent nodes after the first one while the first node is still up and
running. Due to the "shared nothing" model that restricts shared storage
access to one host, when I bring the second node up, the drives I need on
the
SAN show up as unreadable. As a result, adding the second node fails
because
it cannot see the shared quorum drive nor the shared data drive. I have
seen
several people say this is normal behavior, yet it is preventing me from
setting up multi-node clustering. Am I missing something here???
My hardware setup is as follows:
IBM HS20 Blades connected to a Hitachi 9570V through a Brocade fabric. All
hosts are running Windows Server 2003 Enterprise.
|
|
|
| Back to top |
|
|
Isaac Story
Guest
|
| Posted:
Fri Oct 21, 2005 4:50 pm Post subject:
Re: When adding 2nd node, unable to see shared drives |
|
|
I actually had seen that. I tried the work around (and applied SP1) and I was
able to create the cluster, but as they warned in the KB article, if the
drives are accessing properly the cluster will not failover. And sure enough
it does not. I did check to make sure that the drives are set up the same on
both servers (Same bus, Port, TID, and LUN numbers on each server). I have
inserted what results in the cluster srv log. It is still not seeing valid
partitions on one of the hosts and cannot see the quorum. It states that it
had to create a local "dummy" quorum. Am I still missing something??
Found SCSI disk '\\.\PHYSICALDRIVE0' on Bus '0' and Port '0'; at TID '0' and
LUN '2' (hr=00000000, {05AA0768-5F49-49CD-AFDC-96F9D51802D4},
{00000000-0000-0000-0000-000000000000}, 1, 1, 1), (null)
Physical disk PHYSICALDRIVE0 has signature 0. (hr=00000000,
{05AA0768-5F49-49CD-AFDC-96F9D51802D4},
{00000000-0000-0000-0000-000000000000}, 1, 1, 1), (null)
The physical disk '\\.\PHYSICALDRIVE0' does not have any partitions and will
not be managed (hr=0x000001, {05AA0768-5F49-49CD-AFDC-96F9D51802D4},
{00000000-0000-0000-0000-000000000000}, 1, 1, 1), (null)
"John Toner [MVP]" wrote:
| Quote: | Have you seen this KB article?
http://support.microsoft.com/kb/331801
Regards,
John
"Isaac Story" <IsaacStory@discussions.microsoft.com> wrote in message
news:D33A2DA4-E1F0-434C-8E52-A7C3164A9D59@microsoft.com...
I am fairly new to clustering and there is an issue I am having when
adding
subsequent nodes after the first one while the first node is still up and
running. Due to the "shared nothing" model that restricts shared storage
access to one host, when I bring the second node up, the drives I need on
the
SAN show up as unreadable. As a result, adding the second node fails
because
it cannot see the shared quorum drive nor the shared data drive. I have
seen
several people say this is normal behavior, yet it is preventing me from
setting up multi-node clustering. Am I missing something here???
My hardware setup is as follows:
IBM HS20 Blades connected to a Hitachi 9570V through a Brocade fabric. All
hosts are running Windows Server 2003 Enterprise.
|
|
|
| Back to top |
|
|
John Toner [MVP]
Guest
|
| Posted:
Fri Oct 21, 2005 4:50 pm Post subject:
Re: When adding 2nd node, unable to see shared drives |
|
|
Have you seen this KB article?
http://support.microsoft.com/kb/331801
Regards,
John
"Isaac Story" <IsaacStory@discussions.microsoft.com> wrote in message
news:D33A2DA4-E1F0-434C-8E52-A7C3164A9D59@microsoft.com...
| Quote: | I am fairly new to clustering and there is an issue I am having when
adding
subsequent nodes after the first one while the first node is still up and
running. Due to the "shared nothing" model that restricts shared storage
access to one host, when I bring the second node up, the drives I need on
the
SAN show up as unreadable. As a result, adding the second node fails
because
it cannot see the shared quorum drive nor the shared data drive. I have
seen
several people say this is normal behavior, yet it is preventing me from
setting up multi-node clustering. Am I missing something here???
My hardware setup is as follows:
IBM HS20 Blades connected to a Hitachi 9570V through a Brocade fabric. All
hosts are running Windows Server 2003 Enterprise. |
|
|
| Back to top |
|
|
Ryan Sokolowski [MVP]
Guest
|
| Posted:
Sat Oct 22, 2005 12:50 am Post subject:
Re: When adding 2nd node, unable to see shared drives |
|
|
Nope, MSCS is (and remains currently) a "shared-nothing" model...could this
change in the future? Who knows, but for now, you need to make this work as
is...
--
Ryan Sokolowski
MVP - Windows Server - Clustering
MCSE, CCNA, CCDA, BCFP
"A troubleshooter's best tool is the Event Viewer and understanding the
events and messages contained therein."
This posting is provided "AS IS" with no warranties, and confers no rights.
"Isaac Story" <IsaacStory@discussions.microsoft.com> wrote in message
news:11E5B3F2-BE79-4F66-91EA-29961CBD0C2E@microsoft.com...
| Quote: | Is there a way to disable the "shared-nothing" disk protection until I
have
clustering up and running? I don't any other way to get this to work.
Isaac
"John Toner [MVP]" wrote:
Have you seen this KB article?
http://support.microsoft.com/kb/331801
Regards,
John
"Isaac Story" <IsaacStory@discussions.microsoft.com> wrote in message
news:D33A2DA4-E1F0-434C-8E52-A7C3164A9D59@microsoft.com...
I am fairly new to clustering and there is an issue I am having when
adding
subsequent nodes after the first one while the first node is still up
and
running. Due to the "shared nothing" model that restricts shared
storage
access to one host, when I bring the second node up, the drives I need
on
the
SAN show up as unreadable. As a result, adding the second node fails
because
it cannot see the shared quorum drive nor the shared data drive. I have
seen
several people say this is normal behavior, yet it is preventing me
from
setting up multi-node clustering. Am I missing something here???
My hardware setup is as follows:
IBM HS20 Blades connected to a Hitachi 9570V through a Brocade fabric.
All
hosts are running Windows Server 2003 Enterprise.
|
|
|
| Back to top |
|
|
John Toner [MVP]
Guest
|
| Posted:
Mon Oct 24, 2005 4:50 pm Post subject:
Re: When adding 2nd node, unable to see shared drives |
|
|
Are you booting off the SAN or do you have your boot/system disk on the same
bus as your shared storage?
I would troubleshoot this by uninstalling cluster completely on all nodes
and verify that each cluster node can access the shared storage. Power down
Node2 and leave Node1 up and running. With Node1 up, make sure you have
access to the shared storage and create a partition (if necessary) and copy
some data to the disk. Next, power down Node1 and power up Node2 and see if
you can access the data on this partition (you might need to mount it in
order to access the data). If you do not see the partition or the data,
something is screwed up in your SAN (zoning/masking). If you can access the
data without issue, cluster should be able to detect your shared storage
properly.
Regards,
John
"Isaac Story" <IsaacStory@discussions.microsoft.com> wrote in message
news:5F7C84F5-9554-4091-A0F6-BDE2A79F1CF7@microsoft.com...
| Quote: | I actually had seen that. I tried the work around (and applied SP1) and I
was
able to create the cluster, but as they warned in the KB article, if the
drives are accessing properly the cluster will not failover. And sure
enough
it does not. I did check to make sure that the drives are set up the same
on
both servers (Same bus, Port, TID, and LUN numbers on each server). I have
inserted what results in the cluster srv log. It is still not seeing valid
partitions on one of the hosts and cannot see the quorum. It states that
it
had to create a local "dummy" quorum. Am I still missing something??
Found SCSI disk '\\.\PHYSICALDRIVE0' on Bus '0' and Port '0'; at TID '0'
and
LUN '2' (hr=00000000, {05AA0768-5F49-49CD-AFDC-96F9D51802D4},
{00000000-0000-0000-0000-000000000000}, 1, 1, 1), (null)
Physical disk PHYSICALDRIVE0 has signature 0. (hr=00000000,
{05AA0768-5F49-49CD-AFDC-96F9D51802D4},
{00000000-0000-0000-0000-000000000000}, 1, 1, 1), (null)
The physical disk '\\.\PHYSICALDRIVE0' does not have any partitions and
will
not be managed (hr=0x000001, {05AA0768-5F49-49CD-AFDC-96F9D51802D4},
{00000000-0000-0000-0000-000000000000}, 1, 1, 1), (null)
"John Toner [MVP]" wrote:
Have you seen this KB article?
http://support.microsoft.com/kb/331801
Regards,
John
"Isaac Story" <IsaacStory@discussions.microsoft.com> wrote in message
news:D33A2DA4-E1F0-434C-8E52-A7C3164A9D59@microsoft.com...
I am fairly new to clustering and there is an issue I am having when
adding
subsequent nodes after the first one while the first node is still up
and
running. Due to the "shared nothing" model that restricts shared
storage
access to one host, when I bring the second node up, the drives I need
on
the
SAN show up as unreadable. As a result, adding the second node fails
because
it cannot see the shared quorum drive nor the shared data drive. I
have
seen
several people say this is normal behavior, yet it is preventing me
from
setting up multi-node clustering. Am I missing something here???
My hardware setup is as follows:
IBM HS20 Blades connected to a Hitachi 9570V through a Brocade fabric.
All
hosts are running Windows Server 2003 Enterprise.
|
|
|
| Back to top |
|
|
|
|
|
|
FAQ
Memberlist
Register
Profile
Log in to check your private messages
Log in