To configure Highly Available production servers:
-
On the primary production server, the secondary production server and the recovery server, install the Assure MIMIX for AIX software.
-
On the primary production server, use PowerHA to add Resource Group MIMIX_COMMON_RG to the PowerHA configuration with the primary and secondary Production Servers as its participating nodes with the following Policies:
-
Startup Policy : Online On Home Node Only
-
Fallover Policy : Fallover To Next Priority Node In The List
-
Fallback Policy : Never Fallback
This Resource Group will be used control the configuration required by Assure MIMIX for AIX.
-
-
On the primary production server, use PowerHA C-SPOC to create a shared enhanced concurrent mode volume group MIMIX_VG on both production servers.
-
On the primary production server, use PowerHA C-SPOC to create file system /usr/scrt/run/common with an inline log on volume group MIMIX_VG. The filesystem must be at least 128 MB in size.
-
On the primary production server, use PowerHA C-SPOC to create file systems /usr/scrt/run/c<Primary Context ID> and /usr/scrt/run/c<FailOver Context ID> with inline logs on one of the shared enhanced concurrent mode volume groups that contains the production server logical volumes you want protected. The filesystem must be at least 128 MB in size.
-
On the primary production server, use PowerHA to add a Service IP Label/Address resource to the PowerHA configuration. Also add the Service IP Address and associated IP label to the /etc/hosts file on the primary and secondary production servers, the recovery server and the Windows PC that will be used to create the Assure MIMIX for AIX Replication Group configuration.
-
On the primary production server, use PowerHA to add the Service IP Label/Address, Volume Group MIMIX_VG and Filesystem /usr/scrt/run/common resources to Resource Group MIMIX_COMMON_RG.
-
On the primary production server, use PowerHA to verify and synchronize the cluster configuration.
-
On the primary production server, bring Resource Group MIMIX_COMMON_RG online.
-
On the primary production server, mount the filesystems associated with the Assure MIMIX for AIX configuration.
-
mount /usr/scrt/run/c1<Primary Context ID>
-
mount /usr/scrt/run/c17<Failover ContextID>
As in this example, 1 and 17 are the configured Primary and Failover Context IDs, respectively.
-
-
On the recovery server, create volume group MIMIX_VG, which holds the logical volumes for the Assure MIMIX for AIX containers and the volume group(s) that will contain the replicated data. Please specify the AUTO ON option for the volume groups so that they are varied on automatically at system startup.
-
Use the Replication Group wizard to create the Assure MIMIX for AIX replication group configuration.
Refer to the section Configuring Replication Groups. Make sure to add a portal connection, using the Service IP Address as the Host or IP Address for the production server. This will allow the Assure UI portal to connect to either the primary or secondary production server, depending on which server is online. While configuring the replication IP addresses, select the option to use server IP addresses for replication. This will set the Service IP address as the replication IP address.
Also, while configuring the production server logical volumes you want protected, do not select the logical volumes for file systems /usr/scrt/run/c<primary context id>, /usr/scrt/run/c<failover context id> and /usr/scrt/run/common.
After the creation of the Assure MIMIX for AIX replication group configuration, perform the following steps.
-
On the primary production server, verify that filesystem /usr/scrt/run/common is mounted, then execute es_ha_production_setup, as shown below, to link the Assure MIMIX for AIX ODMs /etc/objrepos/SCCuAttr, /etc/objrepos/SCCuObj and /etc/objrepos/SCCuRel to /usr/scrt/run/common/ and create the required subsystem resource controller (SRC) entries:
-
es_ha_production_setup <Primary Context ID> <Failover Context ID>
For example:
-
/usr/scrt/bin/es_ha_production_setup 1 17
-
-
On the secondary production server, execute es_ha_production_setup, as shown below, to link the Assure MIMIX for AIX ODMs /etc/objrepos/SCCuAttr, /etc/objrepos/SCCuObj and /etc/objrepos/SCCuRel to /usr/scrt/run/common/ and create the required subsystem resource controller (SRC) entries:
-
es_ha_production_setup <Primary Context ID> <Failover Context ID>
For example:
-
/usr/scrt/bin/es_ha_production_setup 1 17
-
-
On the primary production server, execute the following command to get the Instance value for the Primary Context ID.
-
rtattr -C<Primary Context ID> -o production -a Instance
The ObjectAttributeValue shown in the output is the Instance number.
For example when you execute:
/usr/scrt/bin/rtattr –C1 -o production -a Instance SCCuAttr: ObjectName = "production" ConfigObjectSerial = 15 ObjectType = "SCRT/info/host" ObjectAttributeName = "Instance" ObjectAttributeValue = "4" ObjectAttributeType = "uint" SerialNumber = 150011 ObjectNlsIndex = 0 SC_reserved = 0 ContextID = 1In the above case, the Instance number for Context ID 1 is 4.
-
-
On the primary production server, execute the following command to obtain the device major number used by Primary ContextID:
-
es_ha_config <Primary Context ID>
For example:
-
/usr/scrt/bin/es_ha_config 1Current major number: 73
-
-
On the secondary production server, use lvlstmajor to obtain a list of unused device major numbers:
Note: The Assure MIMIX for AIX major numbers must match on both the primary and secondary production servers. In case of a change in the major number on one of the production servers, the other must also be updated with the same number to match. Otherwise, Assure MIMIX for AIX will not function properly.For example:
-
/usr/sbin/lvlstmajor 46,50,54,57,64,67,70,73..75,82..93,95...
Note: If the device major number used by Primary Context ID 1 on the primary production server is displayed in the list of unused device major numbers on the secondary production server, move Resource Group MIMIX_COMMON_RG to the secondary production server.If the device major number used by the Primary Context ID on the primary production server is not displayed in the list of unused device major numbers on the secondary production server, use the lvlstmajor AIX command on both production servers to find a matching unused device major number that is unused on both production servers. Then, update the configuration on the primary production server by running the following command before moving Resource Group MIMIX_COMMON_RG to the secondary production server.
For example:
-
/usr/scrt/bin/es_update_major <Primary Context ID> <Instance> <New Major Number>
-
-
On the secondary production server, verify that resource group MIMIX_COMMON_RG is online and filesystem /usr/scrt/run/common is mounted, then execute the following command to update the HostId attribute in the RecoverNow configuration.
-
es_ha_update_hostid <Primary Context ID> < Failover Context ID >
For example:
-
/usr/scrt/bin/es_ha_update_hostid 1 17
-
-
On the secondary production server, verify that resource group MIMIX_COMMON_RG is online and filesystem /usr/scrt/run/common is mounted, then execute the following command to update the HostId attribute in the RecoverNow configuration.
-
es_update_major <Primary Context ID> <Instance> <Major Number>
For example:
-
/usr/scrt/bin/es_update_major 1 4 73
Note that the instance in the above command was obtained from step 15 from the primary production server.
-
-
On the secondary production server, add the replication group port numbers to /etc/services.
The port numbers to add can be found in the /etc/services file on the primary production server.
For example, these port numbers are for primary context id 1 and failover context id 17:
sc1aba_dchannel 5779/tcp sc1aba_dchannel 5783/tcp sc1lca_channel 5780/tcp sc1lca_dchannel 5784/tcp sc1aa_channel 5778/tcp sc1aa_achannel 5782/tcp sc1ra_channel 5785/tcp sc1ca_channel 5781/tcp sc17aba_channel 5787/tcp sc17aba_dchannel 5791/tcp sc17lca_channel 5788/tcp sc17lca_dchannel 5792/tcp sc17aa_channel 5786/tcp sc17aa_achannel 5790/tcp sc17ra_channel 5793/tcp sc17ca_channel 5789/tcp -
On the secondary production server, use PowerHA to move resource group MIMIX_COMMON_RG to the primary production server.
-
On the primary production server, unmount the filesystems associated with the Assure MIMIX for AIX configuration.
-
unmount /usr/scrt/run/c1 <Primary Context ID>
-
unmount /usr/scrt/run/c17 <Failover ContextID>
-
-
On the primary production server, use PowerHA to add the filesystem resources /usr/scrt/run/c1 and /usr/scrt/run/c17 to the resource group APPLICATION_RG, which contains the volume group and the logical volumes you want protected.
Note: The PowerHA Resource Group attribute Filesystems (empty is ALL for volume groups specified) cannot be empty, because the file systems protected by Assure MIMIX for AIX must be mounted by Assure MIMIX for AIX.If there are filesystems on the volume groups specified in the PowerHA Resource Group that are not being protected by Assure MIMIX for AIX, those filesystems will not be mounted by Assure MIMIX for AIX; add the mount points of those filesystems which should be mounted when the resource group is acquired.
-
On the primary production server, use PowerHA to add or change Application Controller Scripts. For your convenience, sample start and stop scripts are provided in the /usr/scrt/samples directory.
Copy the sample Assure MIMIX for AIX start and stop scripts to the location of your choice and change the values for CONTEXT_ID and FAILOVER_CONTEXT_ID and insert your application start and stop scripts where indicated.
For example:
Application Controller Name: App_Controller_C1Start Script: [/scripts/es_ha_start_script_C1.ksh]Stop Script: [/scripts/es_ha_stop_script_C1.ksh]Application Monitor Name(s)Application startup mode: [background]
See the sample Assure MIMIX for AIX start and stop scripts below.
-
On the primary production server, use PowerHA to Add Parent/Child Dependency between resource groups, as follows: MIMIX_COMMON_RG as the Parent Resource Group and the APPLICATION_RG as the Child Resource Group.
-
On the primary production server, use PowerHA to add Online on the Same Node Dependency between resource groups MIMIX_COMMON_RG, APPLICATION_RG and any other resource groups associated with Assure MIMIX for AIX.
-
On the primary production server, use PowerHA to verify and synchronize the cluster configuration.
-
On the recovery server use rtstart –C<Primary Context ID> to manually start the Assure MIMIX for AIX primary context ID.
For example:
-
rtstart –C1
-
-
On the primary production server, bring resource group APPLICATION_RG online.