Document driven network configuration updater
Systems and methods are disclosed for updating network configuration documentation. In an example, a user can upload network configuration documentation with updates to a network to a server. The server can create an update topology corresponding to the documentation by identifying symbols that represent network components. The server can identify changes by comparing the update topology to a configuration data of an existing network. For example, the address of a gateway or the connections to the gateway can change. The server can cause the changes to be presented to a user, such as by highlighting the changes in a diagram. The user can confirm the changes, such as with a conversational workflow, and the server can save the changes to a database. The system can also send commands to the applicable network components to effectuate the confirmed changes.
1. A method for document driven network configuration, comprising:
receiving network documentation that includes updates to a network configuration, wherein the network configuration corresponds to at least a portion of a deployed network;
creating an update topology based on the network documentation, including identifying a plurality of network components in the network documentation using at least one of image recognition and natural language processing;
comparing the update topology to the network configuration to identify a difference between the update topology and the network configuration;
identifying a missing property value in the identified difference, wherein the property value corresponds to a required field for a network component in the update topology;
inputting the identified difference into a machine learning model, wherein the machine learning model outputs a suggested value for the property value;
prompting a user to confirm the difference and the suggested property value; and
in an instance where the user confirms the difference and the suggested property value, updating a database that stores the network configuration with the difference and the suggested property value, wherein the database is used by a configuration process to change the network configuration to implement the difference.
2. The method of claim 1 , further comprising:
determining that a first network component of the plurality of network components in the update topology is missing the required property value;
prompting the user to verify the suggested property value;
receiving user input that differs from the suggested property value;
storing the user input as the property value; and
training another version of the machine learning model based on the user input.
3. The method of claim 1 , wherein, in an instance where the user does not confirm the difference, the method further comprises:
receiving user input corresponding to a correction in the update topology; and
updating the database with the correction, wherein the correction is used in improving the creation of subsequent update topologies.
4. The method of claim 1 , wherein the suggested property value is based on contextual data for the property, the contextual data including attributes of nearby network components in the network topology.
5. The method of claim 4 , wherein the attributes of nearby network components includes at least one of an internet protocol (“IP”) address, a gateway address, and a configuration setting.
6. The method of claim 5 , further comprising:
determining that a portion of the update topology includes an invalid configuration based on properties associated with the plurality of network components;
generating an alternative topology of the update topology that corrects the invalid configuration;
causing an alternative diagram corresponding to the alternative topology to be displayed on a display; and
prompting the user to confirm or reject the alternative topology.
7. The method of claim 1 , further comprising updating at least one network device based on the difference and the suggested property value.
8. A non-transitory, computer-readable medium containing instructions that, when executed by a hardware-based processor, causes the processor to perform stages for document driven network configuration, the stages comprising:
receiving network documentation that includes updates to a network configuration, wherein the network configuration corresponds to at least a portion of a deployed network;
creating an update topology based on the network documentation, including identifying a plurality of network components in the network documentation using at least one of image recognition and natural language processing;
comparing the update topology to the network configuration to identify a difference between the update topology and the network configuration;
identifying a missing property value in the identified difference, wherein the property value corresponds to a required field for a network component in the update topology;
inputting the identified difference into a machine learning model, wherein the machine learning model outputs a suggested value for the property value;
prompting a user to confirm the difference and the suggested property value; and
in an instance where the user confirms the difference and the suggested property value, updating a database that stores the network configuration with the difference and the suggested property value, wherein the database is used by a configuration process to change the network configuration to implement the difference.
9. The non-transitory, computer-readable medium of claim 8 , the stages further comprising:
determining that a first network component of the plurality of network components in the update topology is missing the required property value;
prompting the user to verify the suggested property value;
receiving user input that differs from the suggested property value;
storing the user input as the property value; and
training another version of the machine learning model based on the user input.
10. The non-transitory, computer-readable medium of claim 8 , wherein, in an instance where the user does not confirm the difference, the stages further comprise:
receiving user input corresponding to a correction in the update topology; and
updating the database with the correction, wherein the correction is used to improve the creation of subsequent update topologies.
11. The non-transitory, computer-readable medium of claim 8 , wherein the suggested property value is based on contextual data for the property, the contextual data including attributes of nearby network components in the network topology.
12. The non-transitory, computer-readable medium of claim 11 , wherein the attributes of nearby network components includes at least one of an internet protocol (“IP”) address, a gateway address, and a configuration setting.
13. The non-transitory, computer-readable medium of claim 8 , the stages further comprising:
determining that a portion of the update topology includes an invalid configuration based on properties associated with the plurality of network components;
generating an alternative topology of the update topology that corrects the invalid configuration;
causing an alternative diagram corresponding to the alternative topology to be displayed on a display; and
prompting the user to confirm or reject the alternative topology.
14. The non-transitory, computer-readable medium of claim 13 , further comprising updating at least one network device based on the difference and the suggested property value.
15. A system for document driven network configuration, comprising:
a memory storage including a non-transitory, computer-readable medium comprising instructions; and
a hardware-based processor that executes the instructions to carry out stages comprising:
receiving network documentation that includes updates to a network configuration, wherein the network configuration corresponds to at least a portion of a deployed network;
creating an update topology based on the network documentation, including identifying a plurality of network components in the network documentation using at least one of image recognition and natural language processing;
comparing the update topology to the network configuration to identify a difference between the update topology and the network configuration;
identifying a missing property value in the identified difference, wherein the property value corresponds to a required field for a network component in the update topology;
inputting the identified difference into a machine learning model, wherein the machine learning model outputs a suggested value for the property value;
prompting a user to confirm the difference and the suggested property value; and
in an instance where the user confirms the difference and the suggested property value, updating a database that stores the network configuration with the difference and the suggested property value, wherein the database is used by a configuration process to change the network configuration to implement the difference.
16. The system of claim 15 , the stages further comprising:
determining that a first network component of the plurality of network components in the update topology is missing the required property value;
prompting the user to verify the suggested property value;
receiving user input that differs from the suggested property value;
storing the user input as the property value; and
training another version of the machine learning model based on the user input.
17. The system of claim 15 , wherein, in an instance where the user does not confirm the difference, the stages further comprise:
receiving user input corresponding to a correction in the update topology; and
updating the database with the correction, wherein the correction is used to improve the creation of subsequent update topologies.
18. The system of claim 15 , wherein the suggested property value is based on contextual data for the property, the contextual data including attributes of nearby network components in the network topology.
19. The system of claim 18 , wherein the attributes of nearby network components includes at least one of an internet protocol (“IP”) address, a gateway address, and a configuration setting.
20. The system of claim 15 , the stages further comprising:
determining that a portion of the update topology includes an invalid configuration based on properties associated with the plurality of network components;
generating an alternative topology of the update topology that corrects the invalid configuration;
causing an alternative diagram corresponding to the alternative topology to be displayed on a display; and
prompting the user to confirm or reject the alternative topology.