Control Plane State Freeze for Hitless Network Upgrades
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network devices without multiple routing engine modules face operational disruption during in-service software upgrades (ISSU) procedures, leading to issues like misrouting and blackholing, as they cannot perform a hitless upgrade.
Innovation Solution
The network device determines whether to reboot the control plane, stops executing applications that modify the control plane state, freezes the control plane state by saving data objects in a non-distributed data structure, performs the ISSU procedure, and then restores the control plane state and applications, allowing the data plane to continue operating.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the control plane is rebooted to perform ISSU procedure, then software update is achieved, but network device operations are disrupted causing misrouting and blackholing
Solution Approach 1:
The patent segments the control plane into multiple independent control plane instances (first control plane instance and second control plane instance) that can operate simultaneously. This allows the ISSU procedure to be performed on one instance while the other continues to handle network operations, thereby maintaining reliability during software updates.
Solution Approach 2:
The patent implements preliminary actions by freezing the control plane state before executing the ISSU procedure. The state freeze mechanism preserves the network configuration and routing information beforehand, enabling quick recovery and state restoration after the update, thus preventing misrouting and blackholing issues.
2Reliability
If the network device is taken offline for upgrade, then software update is complete, but operational performance deteriorates and networking problems occur
Solution Approach 1:
The patent ensures continuity of useful action by maintaining control plane functionality throughout the ISSU procedure. Multiple control plane instances operate in parallel, with one handling network operations while the other undergoes software updates. This continuous operation eliminates service interruptions and maintains network availability during upgrades.
Solution Approach 2:
The patent introduces a state freeze mechanism as an intermediary between the running control plane and the ISSU procedure. This intermediary preserves the operational state while allowing software updates to proceed, and then restores the state afterward, simplifying the upgrade process while maintaining network availability.
3Productivity
If applications continue executing during ISSU, then service continuity is maintained, but control plane state becomes inconsistent causing networking issues
Solution Approach 1:
The patent applies preliminary anti-action by freezing the control plane state before the ISSU procedure begins. This preemptive measure prevents applications from modifying the control plane state during the update process, avoiding inconsistencies. After the ISSU completes, the frozen state is restored, ensuring control plane consistency while minimizing service disruption.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
A network device obtains a data package associated with an ISSU procedure and determines, based on the data package, that a control plane of the network device is to be rebooted to facilitate performance of the ISSU procedure. The network device causes, based on determining that the control plane is to be rebooted, a plurality of applications of the network device to stop executing on the network device and a control plane state of the network device to be frozen. The network device then causes the ISSU procedure to be performed. The network causes, based on causing the ISSU procedure to be performed, the control plane state of the network device to be restored and the plurality of applications to resume executing on the network device.