Follow these steps to reset replication for an entire dataservice. The current Primary will remain the Primary. Use the switch after completion to change the Primary.
The procedures in this section are designed for the Multi-Site/Active-Active topology ONLY. Do NOT use these procedures with the Composite Active/Active topology.
For Composite Active/Active Clustering, please refer to Deploying Composite Active/Active Clustering.
See Section 6.11.7, “Resetting an Entire Dataservice from Filesystem Snapshots” if you would like to restore a file system snapshot to every server as part of this process.
Put the dataservice into
MAINTENANCE mode. This ensures
that Tungsten Cluster will not attempt to automatically recover the
service.
cctrl> set policy maintenanceEnable force mode:
cctrl> set force trueShun each datasource:
cctrl>datasource Primary shuncctrl>datasource Replica1 shuncctrl>datasource Replica2 shun
Put each Tungsten Connector offline:
cctrl> router * offlineOn each datasource, reset the service:
shell>trepctl -service east offlineshell>trepctl -service east reset -all -y
Reconfigure the replicator and datasource configuration on each host, starting with the Primary:
cctrl>set force truecctrl>replicator new-Primary mastercctrl>replicator new-Primary onlinecctrl>datasource new-Primary mastercctrl>datasource new-Primary onlinecctrl>replicator Replica1 slave new-Primarycctrl>replicator Replica1 onlinecctrl>datasource Replica1 slavecctrl>datasource Replica1 onlinecctrl>replicator Replica2 slave new-Primarycctrl>replicator Replica2 onlinecctrl>datasource Replica2 slavecctrl>datasource Replica2 online
The connector can now be re-enabled and the cluster returned to operational state:
cctrl>router * onlinecctrl>set policy automaticcctrl>cluster heartbeat
Any servers not matching the Primary must be reprovisioned. Use the tungsten_provision_slave tool to reprovision from the Primary or valid Replica server.