Change Details

The new drbd backed HA labstore cluster - with labstore1004/5 needs monitoring. Preliminary thoughts on what should be monitored: System - Load - Cpu utilization NFS - Number of concurrent clients - Network bandwidth utilization DRBD - Resources - Primary/Secondary - Connection states - Disk states - Resources on primary mounted in the right directory - Disk - Rate of read/write - Number of bytes read/written - Read/write latency - Network - Link network utilization - Link latency?

The new drbd backed HA labstore cluster - with labstore1004/5 needs monitoring. Things we should monitor explicitly * DRBD roles are assigned to nodes in cluster as expected * Status of DRBD node (connection and disk ok) * Status of DRBD service * Cluster IP is assigned to DRBD primary * NFS is served over the cluster IP * Backup jobs status

The new drbd backed HA labstore cluster - with labstore1004/5 needs monitoring. Preliminary thoughts on whatThings we should be monitored: explicitly System* DRBD roles are assigned to nodes in cluster as expected - Load* Status of DRBD node (connection and disk ok) - Cpu utilization NFS - Number of concurrent clients - Network bandwidth utilization DRBD - Resources - Primary/Secondary* Status of DRBD service - Connection states - Disk states - Resources on primary mounted in the right directory - Disk - Rate of read/write - Number of bytes read/written - Read/write latency - Network* Cluster IP is assigned to DRBD primary - Link network utilization* NFS is served over the cluster IP - Link latency?* Backup jobs status