Method and system for network fault monitoring with linux

An apparatus for monitoring network faults using a LINUX operating system comprising:

• a networked computer system comprising:
• a central processing unit (“CPU”) running the LINUX operating system;
• a memory;
• and
• one or more network interface cards;
• a protocol stack within the LINUX operating system, the protocol stack comprising:
• a transport layer;
• a network layer;
• and
• a data link layer, the data link layer comprising one or more input/output control requests that may directly query the status of the one or more network interface cards;
• a network fault monitoring module running on the CPU that polls and tracks the status of the one or more network interface cards;
• and
• a network fault monitoring data store residing in the memory that maintains data tracking the status of the one or more network interface cards.

A method for monitoring network faults using a LINUX operating system comprising:

• initializing data structures for tracking the status of one or more network interface cards to be monitored;
• initiating monitoring of the one or more network interface cards;
• ascertaining a configurable polling interval;
• locking a network mutex;
• determining if a shutdown condition has occurred;
• monitoring the status of the one or more network interface cards when a shutdown condition has not occurred;
• and
• clearing all resources when a shutdown condition has occurred.

The method of step 11, wherein the initializing step comprises:

• initializing a network mutex, a poll mutex, and a poll conditional variable;
• setting a shutdown flag to false;
• and
• opening a datagram socket for use in calling an input/output (“I/O”) control request to query link status of the network interface cards.

The method of step 12, wherein the initializing step further comprises executing the following steps for all network interface cards:

• creating a data structure and add it to a list of network interface cards to be monitored;
• initializing the created data structure with information describing the network interface card, where the information comprises a unique network identification number, a network interface name, a network interface index, an indicator of the type of network interface card, and the hardware path to the network interface card;
• setting a Media Independent Interface (“MII”) status indicator to up;
• selecting a correct input/output (“I/O”) control request call for the type of the network interface card;
• setting an interface status indicator to up;
• if a new Internet Protocol (“IP”) address is available for the network interface card, creating a data structure for the IP address, initializing the data structure, and adding it to a list of IP addresses assigned to the network interface card;
• and
• creating an entry for the network interface card in a status database to enable storage of status data for use by other software modules.

The method of step 11, wherein the monitoring step comprises:

• obtaining a current time in clock ticks since rebooting of the LINUX operating system;
• adding the configurable polling interval to the current time to obtain the start time of the next polling interval;
• querying the status of each network interface card;
• recalculating the current time;
• unlocking the network mutex;
• determining if the current polling interval has expired;
• locking the network mutex if the current polling interval has expired;
• locking the poll mutex if the current polling interval has not expired;
• waiting until the current polling interval has expired;
• unlocking the poll mutex;
• and
• locking the network mutex.

The method of step 14, wherein the querying step comprises:

• calling the correct I/O control request to query the status of the network interface card;
• storing the value returned by the I/O control request in the MII status indicator for the network interface card;
• determining if the interface status indicator is set to up or down;
• making no change to the status database if the interface status indicator is not set to up and the MII status indicator is not set to up;
• making no change to the status database if the interface status indicator is set to up and the MII status indicator is set to up;
• setting the interface status indicator to up and updating the status database to reflect the new status of the network interface card if the interface status indicator is not set to up but the MII status indicator is set to up;
• and
• setting the interface status indicator to down and updating the status database to reflect the new status of the network interface card if the interface status indicator is set to up and the MII status indicator is not set to up.

A computer readable medium upon which is embedded instructions for carrying out a method for monitoring network faults using a LINUX operating system, the method comprising steps for:

• initializing data structures for tracking the status of one or more network interface cards to be monitored;
• initiating monitoring of the one or more network interface cards;
• ascertaining a configurable polling interval;
• locking a network mutex;
• determining if a shutdown condition has occurred;
• monitoring the status of the one or more network interface cards when a shutdown condition has not occurred;
• and
• clearing all resources when a shutdown condition has occurred.

The computer readable medium of step 16, wherein the initializing step comprises:

• initializing a network mutex, a poll mutex, and a poll conditional variable;
• setting a shutdown flag to false;
• and
• opening a datagram socket for use in calling an input/output (“I/O”) control request to query link status of the network interface cards.

The computer readable medium of step 17, wherein the initializing step further comprises executing the following steps for all network interface cards:

• creating a data structure and add it to a list of network interface cards to be monitored;
• initializing the created data structure with information describing the network interface card, where the information comprises a unique network identification number, a network interface name, a network interface index, an indicator of the type of network interface card, and the hardware path to the network interface card;
• setting a Media Independent Interface (“MII”) status indicator to up;
• selecting a correct input/output (“I/O”) control request call for the type of the network interface card;
• setting an interface status indicator to up;
• if a new Internet Protocol (“IP”) address is available for the network interface card, creating a data structure for the IP address, initializing the data structure, and adding it to a list of IP addresses assigned to the network interface card;
• and
• creating an entry for the network interface card in a status database to enable storage of status data for use by other software modules.

The computer readable medium of step 16, wherein the monitoring step comprises:

• obtaining a current time in clock ticks since rebooting of the LINUX operating system;
• adding the configurable polling interval to the current time to obtain the start time of the next polling interval;
• querying the status of each network interface card;
• recalculating the current time;
• unlocking the network mutex;
• determining if the current polling interval has expired;
• locking the network mutex if the current polling interval has expired;
• locking the poll mutex if the current polling interval has not expired;
• waiting until the current polling interval has expired;
• unlocking the poll mutex;
• and
• locking the network mutex.

The computer readable medium of step 19, wherein the querying step comprises:

• calling the correct I/O control request to query the status of the network interface card;
• storing the value returned by the I/O control request in the MII status indicator for the network interface card;
• determining if the interface status indicator is set to up or down;
• making no change to the status database if the interface status indicator is not set to up and the MII status indicator is not set to up;
• making no change to the status database if the interface status indicator is set to up and the MII status indicator is set to up;
• setting the interface status indicator to up and updating the status database to reflect the new status of the network interface card if the interface status indicator is not set to up but the MII status indicator is set to up;
• and
• setting the interface status indicator to down and updating the status database to reflect the new status of the network interface card if the interface status indicator is set to up and the MII status indicator is not set to up.