Groups won't failover

Hello,

I am running Clustering on W2K3.
I just configured my cluster following instructions in the Guide to Creating
and Configuring a Server Cluster Under Windows Server 2003 (beginning at page
20). This document can be downloaded from:
http://www.microsoft.com/downloads/d...displaylang=en

Following these directions, 8 groups are configured (Cluster Group, Group 0
thru Group 8). Cluster Groups resources are: IP Addr, Network Name, and
Physical disk (Q: quorum drive). Following directions on page 35, I
attempted to move the Cluster Group. Each of the 3 resources in the group
change ownership to the 2nd node, however, the IP Address fails and the
Network Name stays offline.

The IP address failure gets logged in the system log. Other messages are
informational.

I can move the group back to the 1st node successfully.

Can anyone tell me what may be causing the IP address to fail?
Nothing relevant in the C:\windows\cluster\cluster.log file either.

All the other groups only have a single disk drive as resources. These
groups move to the 2nd node, and back successfully .

Thanks,
DetRich

Well, I disagree that there is "nothing relevant in the
C:\windows\cluster\cluster.log"....this should contain the reason why the
resource is failing. The cluster.log file is in GMT time, so make sure you
are looking at the correct time within the cluster.log. Also, you should be
reviewing the cluster.log file on the node where the resource failed. Post
the snippet of the log showing the ERR message that should show the IP
address resource fail.

Regards,
John

Visit my blog: http://msmvps.com/blogs/jtoner

"DetRich" <> wrote in message
news149D4F7-0A8A-45B4-9703-...
> Hello,
>
> I am running Clustering on W2K3.
> I just configured my cluster following instructions in the Guide to
> Creating
> and Configuring a Server Cluster Under Windows Server 2003 (beginning at
> page
> 20). This document can be downloaded from:
> http://www.microsoft.com/downloads/d...displaylang=en
>
> Following these directions, 8 groups are configured (Cluster Group, Group
> 0
> thru Group 8). Cluster Groups resources are: IP Addr, Network Name, and
> Physical disk (Q: quorum drive). Following directions on page 35, I
> attempted to move the Cluster Group. Each of the 3 resources in the group
> change ownership to the 2nd node, however, the IP Address fails and the
> Network Name stays offline.
>
> The IP address failure gets logged in the system log. Other messages are
> informational.
>
> I can move the group back to the 1st node successfully.
>
> Can anyone tell me what may be causing the IP address to fail?
> Nothing relevant in the C:\windows\cluster\cluster.log file either.
>
> All the other groups only have a single disk drive as resources. These
> groups move to the 2nd node, and back successfully .
>
> Thanks,
> DetRich

Additional info on my cluster configuraton...

This is the first cluster group:

Group Name: "Cluster Group"
This group was one of the original default groups created after evicting all
servers and restarting the service.
Resources:
IP Address: This is a valid IP Address. Responds good to pings.
Network Name: "Cluster Name". name = HUBNodeA.
This is a NetBIOS name. Resolves correctly and responds to pings. Note
that the FQDN resolves and responds to pings.
Physical Disk: Name = "Disk Q:". This is the quorum drive. It is
configured as the Q: drive on the SAN.

This group will only fail over if the server is brought down. If I attempt
to Move Group, this will not work. All resources go offline, it attempts to
give ownership to Node-B, then the IP Address fails, and then all resources
come back online on Node-A.


This is the second cluster group. It is for Node-A of the two-node cluster.
Group Name: SQLA
Resources:
IP Address: This has a confirmed valid IP Address both nodes as possible
owners, no dependencies.
Network Name: RAS040SQLA. This NetBIOS names resolves and responds to
pings correctly. This is the A-Node of the two-node cluster. Dependency on
IP Address.
Physical Disks:
Disk F: This is the F: drive on the SAN. SQL databases are here.
Disk G: This is the G: drive on the SAN. SQL backups are here.
Disk H: This is the H: drive on the SAN. SQL logs are here.
These physical disks have no dependencies.

After installing the SQL instance-1, the following resources were placed in
the group:
SQL Server (INST1)
SQL Server Agent (INST1)
SQL Server Fulltext (INST1)

When I initiate a Move Group, all resources go offline, ownership changes to
Node-B for about 1 second, the IP Address then fails, then ownership goes
back to NODE-A, and all resources come online.

Does anyone have any idea why the group won't move?

I am wondering if somehow this is a permissions problem. What permissions
does the account running the cluster service need?

I also have a group for the BizTalk 2006 Enterprise Single Sign-On service.
It has the following resources:
IP Addresss - verified good
Network Name: Verified it resolves and responds to pings.
Generic Service: This is the ENTSSO service.

This group somehow has initial owner as Node-B. Not sure how this happened.
When attempting to Move Group, I again get the IP Address failed and it
remains on NODE-B. If NODE-B is shut down, NODE-A becomes the owner, but the
IP Address fails and other resources therefore will not come online. When
NODE-B comes back online, an attempt to Move Group will then succeed.

I really need some help with this. This is a nagging problem that is
impacting my project.

thanks in advance,
DetRich


"John Toner [MVP]" wrote:

> Well, I disagree that there is "nothing relevant in the
> C:\windows\cluster\cluster.log"....this should contain the reason why the
> resource is failing. The cluster.log file is in GMT time, so make sure you
> are looking at the correct time within the cluster.log. Also, you should be
> reviewing the cluster.log file on the node where the resource failed. Post
> the snippet of the log showing the ERR message that should show the IP
> address resource fail.
>
> Regards,
> John
>
> Visit my blog: http://msmvps.com/blogs/jtoner
>
> "DetRich" <> wrote in message
> news149D4F7-0A8A-45B4-9703-...
> > Hello,
> >
> > I am running Clustering on W2K3.
> > I just configured my cluster following instructions in the Guide to
> > Creating
> > and Configuring a Server Cluster Under Windows Server 2003 (beginning at
> > page
> > 20). This document can be downloaded from:
> > http://www.microsoft.com/downloads/d...displaylang=en
> >
> > Following these directions, 8 groups are configured (Cluster Group, Group
> > 0
> > thru Group 8). Cluster Groups resources are: IP Addr, Network Name, and
> > Physical disk (Q: quorum drive). Following directions on page 35, I
> > attempted to move the Cluster Group. Each of the 3 resources in the group
> > change ownership to the 2nd node, however, the IP Address fails and the
> > Network Name stays offline.
> >
> > The IP address failure gets logged in the system log. Other messages are
> > informational.
> >
> > I can move the group back to the 1st node successfully.
> >
> > Can anyone tell me what may be causing the IP address to fail?
> > Nothing relevant in the C:\windows\cluster\cluster.log file either.
> >
> > All the other groups only have a single disk drive as resources. These
> > groups move to the 2nd node, and back successfully .
> >
> > Thanks,
> > DetRich
>
>
>

It won't move because the IP address resource is failing. This is the
default behavior of MSCS.

The answer to why the IP Address resource is failing is likely in the
cluster log. Post the cluster.log messages from the node where the IP
resource is failing and we can try to help identify what is occurring.

Regards,
John

Visit my blog: http://msmvps.com/blogs/jtoner

"DetRich" <> wrote in message
news:5C2586C2-F868-4D63-9EAD-...
> Additional info on my cluster configuraton...
>
> This is the first cluster group:
>
> Group Name: "Cluster Group"
> This group was one of the original default groups created after evicting
> all
> servers and restarting the service.
> Resources:
> IP Address: This is a valid IP Address. Responds good to pings.
> Network Name: "Cluster Name". name = HUBNodeA.
> This is a NetBIOS name. Resolves correctly and responds to pings. Note
> that the FQDN resolves and responds to pings.
> Physical Disk: Name = "Disk Q:". This is the quorum drive. It is
> configured as the Q: drive on the SAN.
>
> This group will only fail over if the server is brought down. If I
> attempt
> to Move Group, this will not work. All resources go offline, it attempts
> to
> give ownership to Node-B, then the IP Address fails, and then all
> resources
> come back online on Node-A.
>
>
> This is the second cluster group. It is for Node-A of the two-node
> cluster.
> Group Name: SQLA
> Resources:
> IP Address: This has a confirmed valid IP Address both nodes as possible
> owners, no dependencies.
> Network Name: RAS040SQLA. This NetBIOS names resolves and responds to
> pings correctly. This is the A-Node of the two-node cluster. Dependency
> on
> IP Address.
> Physical Disks:
> Disk F: This is the F: drive on the SAN. SQL databases are here.
> Disk G: This is the G: drive on the SAN. SQL backups are here.
> Disk H: This is the H: drive on the SAN. SQL logs are here.
> These physical disks have no dependencies.
>
> After installing the SQL instance-1, the following resources were placed
> in
> the group:
> SQL Server (INST1)
> SQL Server Agent (INST1)
> SQL Server Fulltext (INST1)
>
> When I initiate a Move Group, all resources go offline, ownership changes
> to
> Node-B for about 1 second, the IP Address then fails, then ownership goes
> back to NODE-A, and all resources come online.
>
> Does anyone have any idea why the group won't move?
>
> I am wondering if somehow this is a permissions problem. What permissions
> does the account running the cluster service need?
>
> I also have a group for the BizTalk 2006 Enterprise Single Sign-On
> service.
> It has the following resources:
> IP Addresss - verified good
> Network Name: Verified it resolves and responds to pings.
> Generic Service: This is the ENTSSO service.
>
> This group somehow has initial owner as Node-B. Not sure how this
> happened.
> When attempting to Move Group, I again get the IP Address failed and it
> remains on NODE-B. If NODE-B is shut down, NODE-A becomes the owner, but
> the
> IP Address fails and other resources therefore will not come online. When
> NODE-B comes back online, an attempt to Move Group will then succeed.
>
> I really need some help with this. This is a nagging problem that is
> impacting my project.
>
> thanks in advance,
> DetRich
>
>
> "John Toner [MVP]" wrote:
>
>> Well, I disagree that there is "nothing relevant in the
>> C:\windows\cluster\cluster.log"....this should contain the reason why the
>> resource is failing. The cluster.log file is in GMT time, so make sure
>> you
>> are looking at the correct time within the cluster.log. Also, you should
>> be
>> reviewing the cluster.log file on the node where the resource failed.
>> Post
>> the snippet of the log showing the ERR message that should show the IP
>> address resource fail.
>>
>> Regards,
>> John
>>
>> Visit my blog: http://msmvps.com/blogs/jtoner
>>
>> "DetRich" <> wrote in message
>> news149D4F7-0A8A-45B4-9703-...
>> > Hello,
>> >
>> > I am running Clustering on W2K3.
>> > I just configured my cluster following instructions in the Guide to
>> > Creating
>> > and Configuring a Server Cluster Under Windows Server 2003 (beginning
>> > at
>> > page
>> > 20). This document can be downloaded from:
>> > http://www.microsoft.com/downloads/d...displaylang=en
>> >
>> > Following these directions, 8 groups are configured (Cluster Group,
>> > Group
>> > 0
>> > thru Group 8). Cluster Groups resources are: IP Addr, Network Name,
>> > and
>> > Physical disk (Q: quorum drive). Following directions on page 35, I
>> > attempted to move the Cluster Group. Each of the 3 resources in the
>> > group
>> > change ownership to the 2nd node, however, the IP Address fails and the
>> > Network Name stays offline.
>> >
>> > The IP address failure gets logged in the system log. Other messages
>> > are
>> > informational.
>> >
>> > I can move the group back to the 1st node successfully.
>> >
>> > Can anyone tell me what may be causing the IP address to fail?
>> > Nothing relevant in the C:\windows\cluster\cluster.log file either.
>> >
>> > All the other groups only have a single disk drive as resources. These
>> > groups move to the 2nd node, and back successfully .
>> >
>> > Thanks,
>> > DetRich
>>
>>
>>

Hello All,

I was able to get this figured out. It turns out that the IP Config on
Node-B had an incorrect subnet mask. I noticed in Cluster Admin that after
attempting to move the group, the subnet mask for the IP Address resource was
not correct. I changed it to the correct value (from 255.255.255.0 to
255.255.254.0) and the IP Address resource came online. I think checked the
IP config for the server and saw that the subnet mask was wrong. When
corrected, everything worked fine.

Lesson Learned:
If Cluster Groups are not functioning correctly, verify the IP config on the
host server.

Issue Resolved!!!

"John Toner [MVP]" wrote:

> It won't move because the IP address resource is failing. This is the
> default behavior of MSCS.
>
> The answer to why the IP Address resource is failing is likely in the
> cluster log. Post the cluster.log messages from the node where the IP
> resource is failing and we can try to help identify what is occurring.
>
> Regards,
> John
>
> Visit my blog: http://msmvps.com/blogs/jtoner
>
> "DetRich" <> wrote in message
> news:5C2586C2-F868-4D63-9EAD-...
> > Additional info on my cluster configuraton...
> >
> > This is the first cluster group:
> >
> > Group Name: "Cluster Group"
> > This group was one of the original default groups created after evicting
> > all
> > servers and restarting the service.
> > Resources:
> > IP Address: This is a valid IP Address. Responds good to pings.
> > Network Name: "Cluster Name". name = HUBNodeA.
> > This is a NetBIOS name. Resolves correctly and responds to pings. Note
> > that the FQDN resolves and responds to pings.
> > Physical Disk: Name = "Disk Q:". This is the quorum drive. It is
> > configured as the Q: drive on the SAN.
> >
> > This group will only fail over if the server is brought down. If I
> > attempt
> > to Move Group, this will not work. All resources go offline, it attempts
> > to
> > give ownership to Node-B, then the IP Address fails, and then all
> > resources
> > come back online on Node-A.
> >
> >
> > This is the second cluster group. It is for Node-A of the two-node
> > cluster.
> > Group Name: SQLA
> > Resources:
> > IP Address: This has a confirmed valid IP Address both nodes as possible
> > owners, no dependencies.
> > Network Name: RAS040SQLA. This NetBIOS names resolves and responds to
> > pings correctly. This is the A-Node of the two-node cluster. Dependency
> > on
> > IP Address.
> > Physical Disks:
> > Disk F: This is the F: drive on the SAN. SQL databases are here.
> > Disk G: This is the G: drive on the SAN. SQL backups are here.
> > Disk H: This is the H: drive on the SAN. SQL logs are here.
> > These physical disks have no dependencies.
> >
> > After installing the SQL instance-1, the following resources were placed
> > in
> > the group:
> > SQL Server (INST1)
> > SQL Server Agent (INST1)
> > SQL Server Fulltext (INST1)
> >
> > When I initiate a Move Group, all resources go offline, ownership changes
> > to
> > Node-B for about 1 second, the IP Address then fails, then ownership goes
> > back to NODE-A, and all resources come online.
> >
> > Does anyone have any idea why the group won't move?
> >
> > I am wondering if somehow this is a permissions problem. What permissions
> > does the account running the cluster service need?
> >
> > I also have a group for the BizTalk 2006 Enterprise Single Sign-On
> > service.
> > It has the following resources:
> > IP Addresss - verified good
> > Network Name: Verified it resolves and responds to pings.
> > Generic Service: This is the ENTSSO service.
> >
> > This group somehow has initial owner as Node-B. Not sure how this
> > happened.
> > When attempting to Move Group, I again get the IP Address failed and it
> > remains on NODE-B. If NODE-B is shut down, NODE-A becomes the owner, but
> > the
> > IP Address fails and other resources therefore will not come online. When
> > NODE-B comes back online, an attempt to Move Group will then succeed.
> >
> > I really need some help with this. This is a nagging problem that is
> > impacting my project.
> >
> > thanks in advance,
> > DetRich
> >
> >
> > "John Toner [MVP]" wrote:
> >
> >> Well, I disagree that there is "nothing relevant in the
> >> C:\windows\cluster\cluster.log"....this should contain the reason why the
> >> resource is failing. The cluster.log file is in GMT time, so make sure
> >> you
> >> are looking at the correct time within the cluster.log. Also, you should
> >> be
> >> reviewing the cluster.log file on the node where the resource failed.
> >> Post
> >> the snippet of the log showing the ERR message that should show the IP
> >> address resource fail.
> >>
> >> Regards,
> >> John
> >>
> >> Visit my blog: http://msmvps.com/blogs/jtoner
> >>
> >> "DetRich" <> wrote in message
> >> news149D4F7-0A8A-45B4-9703-...
> >> > Hello,
> >> >
> >> > I am running Clustering on W2K3.
> >> > I just configured my cluster following instructions in the Guide to
> >> > Creating
> >> > and Configuring a Server Cluster Under Windows Server 2003 (beginning
> >> > at
> >> > page
> >> > 20). This document can be downloaded from:
> >> > http://www.microsoft.com/downloads/d...displaylang=en
> >> >
> >> > Following these directions, 8 groups are configured (Cluster Group,
> >> > Group
> >> > 0
> >> > thru Group 8). Cluster Groups resources are: IP Addr, Network Name,
> >> > and
> >> > Physical disk (Q: quorum drive). Following directions on page 35, I
> >> > attempted to move the Cluster Group. Each of the 3 resources in the
> >> > group
> >> > change ownership to the 2nd node, however, the IP Address fails and the
> >> > Network Name stays offline.
> >> >
> >> > The IP address failure gets logged in the system log. Other messages
> >> > are
> >> > informational.
> >> >
> >> > I can move the group back to the 1st node successfully.
> >> >
> >> > Can anyone tell me what may be causing the IP address to fail?
> >> > Nothing relevant in the C:\windows\cluster\cluster.log file either.
> >> >
> >> > All the other groups only have a single disk drive as resources. These
> >> > groups move to the 2nd node, and back successfully .
> >> >
> >> > Thanks,
> >> > DetRich
> >>
> >>
> >>
>
>
>