Hi, We have two standard v5200 storage already in production. We using host based mirroring on the Solaris server for data redundancy.

We will connect in the future new ESXi servers to the SAN and will want to use hyperswap for datastore redundancy (no host based mirroring possible).

Should be not too difficult to do :

• define dedicated fiber ports on V5200 for replication traffic
• set correct portmask to dedicate those ports for replication traffic only (chsystem -localfcportmask xxxxxxxxx)
• SAN zoning between the ports listed above (use the physical WWN's) - should be in dedicated VF
• Now you should be able to change the topology to hyperswap
• Once in hyperswap, you should create "ip quorum" that should sit in a 3rd location. IP quorum is a .jar file that can run on different platforms (java)

All this should be online and existing (non-replicated) volumes will stay accessible all the time

Hi, We have two standard v5200 storage already in production. We using host based mirroring on the Solaris server for data redundancy.

We will connect in the future new ESXi servers to the SAN and will want to use hyperswap for datastore redundancy (no host based mirroring possible).

I'd like to amend this:

• FS5200_2 needs to be freed, that is the cluster config existing on that one will be entirely cleared, so both node canisters are in status CANDIDATE
• now the previous FS5002_2 can be added as additional control enclosure to the existing FS5200_1, see IBM Docs "Adding a control enclosure to an existing system"
• configure sites ids for the two control enclosures
• set up an IP quorum in a third site; it only should be set up to store metadata, if network bandwidth and/or disk throughput are sufficient
• now change topology to hyperswap

Should be not too difficult to do :

• define dedicated fiber ports on V5200 for replication traffic
• set correct portmask to dedicate those ports for replication traffic only (chsystem -localfcportmask xxxxxxxxx)
• SAN zoning between the ports listed above (use the physical WWN's) - should be in dedicated VF
• Now you should be able to change the topology to hyperswap
• Once in hyperswap, you should create "ip quorum" that should sit in a 3rd location. IP quorum is a .jar file that can run on different platforms (java)

All this should be online and existing (non-replicated) volumes will stay accessible all the time

Hi, We have two standard v5200 storage already in production. We using host based mirroring on the Solaris server for data redundancy.

We will connect in the future new ESXi servers to the SAN and will want to use hyperswap for datastore redundancy (no host based mirroring possible).

Thanks. If i understand well. there is no way to keep the volume on the second storage (FS5200_2). From the docs: "Do not start this procedure unless you are certain that you want to remove all the volume and configuration data from this system".

I will need to recreate the volumes for host based morring (Solaris server will not use hyperswap volumes) and the mirror will take some time (couple of days, depending on the size).

I'd like to amend this:

• FS5200_2 needs to be freed, that is the cluster config existing on that one will be entirely cleared, so both node canisters are in status CANDIDATE
• now the previous FS5002_2 can be added as additional control enclosure to the existing FS5200_1, see IBM Docs "Adding a control enclosure to an existing system"
• configure sites ids for the two control enclosures
• set up an IP quorum in a third site; it only should be set up to store metadata, if network bandwidth and/or disk throughput are sufficient
• now change topology to hyperswap

Should be not too difficult to do :

• define dedicated fiber ports on V5200 for replication traffic
• set correct portmask to dedicate those ports for replication traffic only (chsystem -localfcportmask xxxxxxxxx)
• SAN zoning between the ports listed above (use the physical WWN's) - should be in dedicated VF
• Now you should be able to change the topology to hyperswap
• Once in hyperswap, you should create "ip quorum" that should sit in a 3rd location. IP quorum is a .jar file that can run on different platforms (java)

All this should be online and existing (non-replicated) volumes will stay accessible all the time

Hi, We have two standard v5200 storage already in production. We using host based mirroring on the Solaris server for data redundancy.

We will connect in the future new ESXi servers to the SAN and will want to use hyperswap for datastore redundancy (no host based mirroring possible).

Is this procedure is the same as running 'leavecluster -force'  on both V5200_2 storage nodes ?

Thanks. If i understand well. there is no way to keep the volume on the second storage (FS5200_2). From the docs: "Do not start this procedure unless you are certain that you want to remove all the volume and configuration data from this system".

I will need to recreate the volumes for host based morring (Solaris server will not use hyperswap volumes) and the mirror will take some time (couple of days, depending on the size).

I'd like to amend this:

• FS5200_2 needs to be freed, that is the cluster config existing on that one will be entirely cleared, so both node canisters are in status CANDIDATE
• now the previous FS5002_2 can be added as additional control enclosure to the existing FS5200_1, see IBM Docs "Adding a control enclosure to an existing system"
• configure sites ids for the two control enclosures
• set up an IP quorum in a third site; it only should be set up to store metadata, if network bandwidth and/or disk throughput are sufficient
• now change topology to hyperswap

Should be not too difficult to do :

• define dedicated fiber ports on V5200 for replication traffic
• set correct portmask to dedicate those ports for replication traffic only (chsystem -localfcportmask xxxxxxxxx)
• SAN zoning between the ports listed above (use the physical WWN's) - should be in dedicated VF
• Now you should be able to change the topology to hyperswap
• Once in hyperswap, you should create "ip quorum" that should sit in a 3rd location. IP quorum is a .jar file that can run on different platforms (java)

All this should be online and existing (non-replicated) volumes will stay accessible all the time

Hi, We have two standard v5200 storage already in production. We using host based mirroring on the Solaris server for data redundancy.

We will connect in the future new ESXi servers to the SAN and will want to use hyperswap for datastore redundancy (no host based mirroring possible).

Should be not too difficult to do :

• define dedicated fiber ports on V5200 for replication traffic
• set correct portmask to dedicate those ports for replication traffic only (chsystem -localfcportmask xxxxxxxxx)
• SAN zoning between the ports listed above (use the physical WWN's) - should be in dedicated VF
• Now you should be able to change the topology to hyperswap
• Once in hyperswap, you should create "ip quorum" that should sit in a 3rd location. IP quorum is a .jar file that can run on different platforms (java)

All this should be online and existing (non-replicated) volumes will stay accessible all the time

Hi, We have two standard v5200 storage already in production. We using host based mirroring on the Solaris server for data redundancy.

We will connect in the future new ESXi servers to the SAN and will want to use hyperswap for datastore redundancy (no host based mirroring possible).

Hi, We have two standard v5200 storage already in production. We using host based mirroring on the Solaris server for data redundancy.

We will connect in the future new ESXi servers to the SAN and will want to use hyperswap for datastore redundancy (no host based mirroring possible).

Hi, We have two standard v5200 storage already in production. We using host based mirroring on the Solaris server for data redundancy.

We will connect in the future new ESXi servers to the SAN and will want to use hyperswap for datastore redundancy (no host based mirroring possible).