Just-in-time identification of slow drain devices in a fibre channel network

1. A method comprising:

determining whether a first condition has been met, the first condition including that an edge port, in a fibre channel network, has experienced buffer exhaustion at least once;

determining whether a second condition has been met in response to the first condition being met, the second condition including that at least a portion of data frames received at a first switch in the fibre channel network is being buffered at the first switch, wherein the first switch is coupled to the edge port through one or more first inter-switch links;

determining whether a third condition has been met in response to the second condition being met, the third condition including that a port of a second switch in the fibre channel network that is a peer to the first switch has experienced buffer exhaustion at least once, wherein the second switch is coupled to the first switch through one or more second inter-switch links;

when the first condition, the second condition, and the third condition have been met a first predetermined number of times, characterizing a congestion level of the edge port;

when the congestion level is characterized as a level 1 slow drain port:

determining whether a fourth condition has been met, the fourth condition including that the edge port has timeout dropped at least one data packet;

when the fourth condition has been met, determining whether a fifth condition has been met, the fifth condition including that the port has hit buffer exhaustion a second predetermined number of times and has awaited receipt of ready signal from the first switch for a second predetermined time period;

when the fifth condition has been met, characterizing the edge port as a level 2 slow drain port;

when the congestion level is characterized as the level 2 slow drain port:

determining whether a sixth condition has been met, the sixth condition including that the edge port has awaited receipt of a ready signal from a target device for a third predetermined time period; and

when the sixth condition has been met, determining whether a seventh condition has been met, the seventh condition including determining whether the port has dropped a third predetermined number of packets and has awaited receipt of a ready signal from the first switch for a fourth predetermined time period; and

taking remedial action based on the congestion level of the edge port.

2. The method of claim 1 , further comprising:

taking level 1 remedial action in connection with the edge port when the congestion level is characterized as a level 1 slow drain port.

3. The method of claim 1 , further comprising:

taking level 2 remedial action in connection with the edge port when the congestion level is characterized as a level 2 slow drain port.

4. The method of claim 1 , further comprising:

taking level 3 remedial action in connection with the edge port when the congestion level is characterized as a level 3 slow drain port.

5. The method of claim 1 , further comprising:

periodically polling a first plurality of interface counters associated with the edge port and another port, the edge port connected to a target device;

periodically polling a second plurality of interface counters associated with the port of, the another port and the port;

storing counter data obtained via the polling in a time series database.

6. One or more non-transitory tangible media that includes code for execution and when executed by a processor is operable to perform operations comprising:

determining whether a first condition has been met, the first condition including that an edge port, in a fibre channel network, has experienced buffer exhaustion at least once;

determining whether a second condition has been met in response to the first condition being met, the second condition including that at least a portion of data frames received at a first switch in the fibre channel network is being buffered at the first switch, wherein the first switch is coupled to the edge port through one or more first inter-switch links;

determining whether a third condition has been met in response to the second condition being met, the third condition including that a port of a second switch in the fibre channel network that is a peer to the first switch has experienced buffer exhaustion at least once, wherein the second switch is coupled to the first switch through one or more second inter-switch links;

when the first condition, the second condition, and the third condition have been met a first predetermined number of times, characterizing a congestion level of the edge port;

when the congestion level is characterized as a level 1 slow drain port:

determining whether a fourth condition has been met, the fourth condition including that the edge port has timeout dropped at least one data packet;

when the fourth condition has been met, determining whether a fifth condition has been met, the fifth condition including that the port has hit buffer exhaustion a second predetermined number of times and has awaited receipt of ready signal from the first switch for a second predetermined time period;

when the fifth condition has been met, characterizing the edge port as a level 2 slow drain port;

when the congestion level is characterized as the level 2 slow drain port:

determining whether a sixth condition has been met, the sixth condition including that the edge port has awaited receipt of a ready signal from a target device for a third predetermined time period; and

when the sixth condition has been met, determining whether a seventh condition has been met, the seventh condition including determining whether the port has dropped a third predetermined number of packets and has awaited receipt of a ready signal from the first switch for a fourth predetermined time period; and

taking remedial action based on the congestion level of the edge port.

7. The media of claim 6 , wherein the operations include taking level 1 remedial action in connection with the edge port when the congestion level is characterized as a level 1 slow drain port.

8. The media of claim 6 , wherein the operations including taking level 2 remedial action in connection with the edge port when the congestion level is characterized as the level 2 slow drain port.

9. The media of claim 6 , wherein the operations include taking level 3 remedial action in connection with the edge port when the congestion level is characterized as a level 3 slow drain port.

10. The media of claim 6 , wherein the operations include:

periodically polling a first plurality of interface counters associated with the edge port and another port, the edge port connected to a target device;

periodically polling a second plurality of interface counters associated with the port of, the another port and the port;

storing counter data obtained via the polling in a time series database.

11. An apparatus comprising:

a memory element configured to store data; and a processor operable to execute instructions associated with the data that, when executed, cause the processor to:

determine whether a first condition has been met, the first condition including that an edge port, in a fibre channel network, has experienced buffer exhaustion at least once;

determine whether a second condition has been met in response to the first condition being met, the second condition including that at least a portion of data frames received at a first switch in the fibre channel network is being buffered at the first switch, wherein the first switch is coupled to the edge port through one or more first inter-switch links;

determine whether a third condition has been met in response to the second condition being met, the third condition including that a port of a second switch in the fibre channel network that is a peer to the first switch has experienced buffer exhaustion at least once, wherein the second switch is coupled to the first switch through one or more second inter-switch links;

when the first condition, the second condition, and the third condition have been met a first predetermined number of times, characterize a congestion level of the edge port;

when the congestion level is characterized as a level 1 slow drain port:

determine whether a fourth condition has been met, the fourth condition including that the edge port has timeout dropped at least one data packet;

when the fourth condition has been met, determine whether a fifth condition has been met, the fifth condition including that the port has hit buffer exhaustion a second predetermined number of times and has awaited receipt of ready signal from the first switch for a second predetermined time period;

when the fifth condition has been met, characterize the edge port as a level 2 slow drain port;

when the congestion level is characterized as the level 2 slow drain port:

determine whether a sixth condition has been met, the sixth condition including that the edge port has awaited receipt of a ready signal from a target device for a third predetermined time period; and

when the sixth condition has been met, determine whether a seventh condition has been met, the seventh condition including determining whether the port has dropped a third predetermined number of packets and has awaited receipt of a ready signal from the first switch for a fourth predetermined time period; and

take remedial action based on the congestion level of the edge port.

12. The apparatus of claim 11 , wherein the processor is further configured to take level 1 remedial action in connection with the edge port when the congestion level is characterized as a level 1 slow drain port.

13. The apparatus of claim 11 , wherein the processor is further configured to take level 2 remedial action in connection with the edge port when the congestion level is characterized as a level 2 slow drain port.

14. The apparatus of claim 11 , wherein the processor is further configured to take level 3 remedial action in connection with the edge port when the congestion level is characterized as a level 3 slow drain port.

15. The apparatus of claim 11 , wherein the processor is further configured to:

periodically poll a first plurality of interface counters associated with the edge port and another port, the edge port connected to a target device;

periodically poll a second plurality of interface counters associated with the port of, the another port and the port;

store counter data obtained via the polling in a time series database.