Configuration of node port for maintenance without load balancer coordination
A computer implemented method providing a network policy configured to prevent transaction data processing from being interrupted during maintenance performed on nodes in a cluster network. The network policy reduces maintenance period times and the amount of personnel needed to perform node maintenance by simulating node failure. The simulated node failure prevents new transaction data from being sent the node that is undergoing maintenance.
1 . A method comprising:
receiving, by a network management server, a network policy corresponding to a node in a cluster pool comprising a plurality of nodes, wherein the node comprises a first port configured to receive transaction data and a second port configured to receive periodic health check messages from a load balancer;
implementing, by the network management server, the network policy to configure a port restriction of the second port of the node, wherein the port restriction causes the load balancer to stop transmitting the transaction data to the first port;
determining, by the network management server, that the transaction data received at the first port is processed by the node after implementing the network policy in preparing the node for maintenance; and
revoking, by the network management server, the network policy to remove the port restriction at the second port after completion of the maintenance at the node based on a network policy status indicator.
2 . The method of claim 1 , wherein the port restriction causes a simulated error at the node that prevents the node from responding to the periodic health check messages within a predetermined period of time, and wherein the simulated error causes the load balancer to determine that the node is unable to process the transaction data.
3 . The method of claim 1 , wherein the network policy status indicator is a second network policy corresponding to the second port of the node, and wherein the second network policy removes the port restriction at the second port of the node.
4 . The method of claim 1 , wherein the network policy status indicator is a TTL (time-to-live) value, and wherein the network policy is removed upon expiration of the TTL value.
5 . The method of claim 1 , further comprising monitoring, by the network management server, that the node responds to at least one of the periodic health check messages within a predetermined period after the revoking of the network policy.
6 . The method of claim 5 , further comprising monitoring, by the network management server, that the node receives new transaction data at the first port after the revoking of the network policy.
7 . The method of claim 1 , further comprising:
provisioning, by the network management server, a back-up node to the cluster pool; and
initializing, by the network management server, the back-up node to receive the transaction data.
8 . The method of claim 1 , wherein the maintenance comprises at least one of installing a software patch on the node or repairing a hardware failure at the node.
9 . A system comprising:
a plurality of nodes;
a cluster pool comprising the plurality of nodes;
a load balancer server;
a network management server communicably coupled to the plurality of nodes, wherein the network management server is configured to:
receive a network policy corresponding to a node in a cluster pool, wherein the cluster pool comprises a plurality of nodes, and wherein the node in the cluster pool comprises a first port configured to receive transaction data and a second port configured to receive periodic health check messages from a load balancer;
implement the network policy to configure a port restriction of the second port of the node, wherein the port restriction causes the load balancer to stop transmitting the transaction data to the first port;
determine that transaction data received at the first port is processed by the node after implementing the network policy to prepare the node for maintenance; and
revoke the network policy to remove the port restriction at the second port after completion of the maintenance at the node based on a network policy status indicator.
10 . The system of claim 9 , wherein the port restriction causes a simulated error at the node that prevents the node from responding to the periodic health check messages within a predetermined period of time, and wherein the simulated error causes the load balancer determines that the node us unable to process the transaction data.
11 . The system of claim 9 , wherein the network policy status indicator is a second network policy corresponding to the second port of the node, and wherein the second network policy removes the port restriction at the second port of the node.
12 . The system of claim 9 , wherein the network policy status indicator is a TTL (time-to-live) value, and wherein the network policy is removed upon expiration of the TTL value.
13 . The system of claim 9 , wherein the network management server is further configured to monitor that the node responds to at least one of the periodic health check messages within a predetermined period after revoking of the network policy.
14 . The system of claim 13 , wherein the network management server is further configured to monitor that the node receives new transaction data at the first port after revoking of the network policy.
15 . The system of claim 9 , wherein the network management server is further configured to:
provision a back-up node to the cluster pool; and
initialize the back-up node to receive the transaction data.
16 . The system of claim 9 , wherein the maintenance comprises at least one of installing a software patch on the node or repairing a hardware failure at the node.
17 . A non-transitory computer readable medium, with instructions stored thereon, when executed by one or more processors, cause the one or more processors to perform operations comprising:
receiving a network policy corresponding to a node in a cluster pool comprising a plurality of nodes, wherein the node comprises a first port configured to receive transaction data and a second port configured to receive periodic health check messages from a load balancer;
provisioning a back-up node to the cluster pool;
initializing the back-up node to receive the transaction data;
implementing the network policy to configure a port restriction of the second port of the node, wherein the port restriction causes the load balancer to stop transmitting the transaction data to the first port;
determining that transaction data received at the first port is processed by the node after implementing the network policy in preparing the node for maintenance; and
revoking the network policy to remove the port restriction at the second port after completion of the maintenance at the node based on a network policy status indicator.
18 . The non-transitory computer readable medium of claim 17 , wherein the port restriction causes a simulated error at the node that prevents the node from responding to the periodic health check messages within a predetermined period of time, and wherein the simulated error causes the load balancer determines that the node us unable to process the transaction data.
19 . The non-transitory computer readable medium of claim 17 , when executed by one or more processors is further configured to perform operations comprising:
receiving a second network policy corresponding to the node in the cluster pool, wherein the second network policy removes the port restriction for the second port.
20 . The non-transitory computer readable medium of claim 17 , wherein preparing the node for maintenance comprises draining pod data from the node to the back-up node.