How to re-enable replication in a two-node SAP performance based model

SUSE Linux Enterprise Server for SAP Applications 11 Service Pack 3
SUSE Linux Enterprise Server for SAP Applications 11 Service Pack 4
SUSE Linux Enterprise Server for SAP Applications 12 Service Pack 1
SUSE Linux Enterprise Server for SAP Applications 12 Service Pack 2

Node to Node replication of the SAP database is not working.

Enable/Re-enable SAP database synchronization:

Make sure that the cluster will not take any actions whilst you are working, by putting the cluster in to maintenance mode (see TID#7023135) or by stopping pacemaker on both nodes. Note that stopping pacemaker will attempt to shutdown the SAP database and related processes on each node. To avoid the possibility of triggering a 'take-over,' care should be taken to stop pacemaker on the 'slave' node first and allow enough time for the unload process to complete.

When re-configuring replication on an existing SAP cluster, it is imperative that the database/node with the most correct and up-to-date data is designated as the 'primary'. If this is done incorrectly, there is a risk of over-writing the most up-to-date database with an out-of-date or 'empty' database, leading to loss of data and a need to restore from any existing backup.


-----------------------------------------------------------
On the node that is to be the 'primary', enable the site for replication (Should not be needed if the node to be the primary was already the primary in a previous working configuration):

The database and services should be running.

   HDB start

   hdbnsutil -sr_enable --name=unique_site_name

Replace 'unique_site_name' with any sensible label (such as the node or server name). The server running the 'primary' database must have a unique and different site name from server running the secondary database.

   e.g. hdbnsutil -sr_enable --name=node1


-----------------------------------------------------------
On the node to be the slave database (receiving the synced data from the primary database):

Database should *not* be running (if necessary use 'HDB stop').

   hdbnsutil -sr_register --name=unique_site_name --remoteHost=name_of_Linux_server_running_primaryDB --remoteInstance=SAP_instance_number --mode=sync


   e.g. hdbnsutil -sr_register --name=node2 --remoteHost=sapn1 --remoteInstance=00 --mode=sync

   HDB start


NOTE : There are other synchronization '--mode' options in addition to 'sync'.


-----------------------------------------------------------
The configuration of the nodes can be checked by running the following on each node:

   hdbnsutil -sr_state


Below is an example of the output from 'hdbnsutil -sr_state' when run first on a primary node and then on a secondary node:

a00adm@sapn1:/hana/shared/A00/HDB00/exe/python_support> hdbnsutil -sr_state
checking for active or inactive nameserver ...

System Replication State
~~~~~~~~~~~~~~~~~~~~~~~~
online: true

mode: primary
site id: 1
site name: node1

Host Mappings:
~~~~~~~~~~~~~~

sapn1 -> [node1] sapn1
sapn1 -> [node2] sapn2


done.
a00adm@sapn1:/hana/shared/A00/HDB00/exe/python_support>

-----------------------------------------------------------

a00adm@sapn2:/hana/shared/A00/HDB00/exe/python_support> hdbnsutil -sr_state
checking for active or inactive nameserver ...

System Replication State
~~~~~~~~~~~~~~~~~~~~~~~~
online: true

mode: sync
site id: 2
site name: node2
active primary site: 1


Host Mappings:
~~~~~~~~~~~~~~

sapn2 -> [node1] sapn1
sapn2 -> [node2] sapn2

primary masters:sapn1

done.
a00adm@sapn2:/hana/shared/A00/HDB00/exe/python_support>

-----------------------------------------------------------


Before you try to manage these resources with the cluster, make sure the replication/synchronization is working between the primary and secondary databases:

See TID#7022984 - Basic health check for two-node SAP HANA performance based model

Example of the global.ini files from a working 2-node SAP cluster. These files are directly altered by some SAP commands including thoes used to configure database replication:

The file is present on each node and can be found in /hana/shared/<SITE_ID>/global/hdb/custom/config/global.ini


Master Node:

sapn1:~ # cat /hana/shared/A00/global/hdb/custom/config/global.ini
[system_information]
usage = production

[system_replication]
mode = primary
actual_mode = primary
site_id = 1
site_name = sap_site1



Slave Node:

sapn2:~ # cat /hana/shared/A00/global/hdb/custom/config/global.ini
[system_information]
usage = production

[system_replication]
site_id = 2
mode = sync
actual_mode = sync
site_name = sap_site2
operation_mode = delta_datashipping
enable_sync_via_log = false

[system_replication_site_masters]
1 = sapn1:30001


Please note that if any of the SAP software related binaries or commands do not function as expected, a support request should be logged with SAP's support organization.