Testing and error recovery across multiple switching fabrics
A packet-based traffic switching system with error detection and correction without taking the system offline. The system tests offline paths without interfering with other online paths. Also, the system tests online paths even while no data cell traffic is sent over the paths. The system responds to the addition or removal of paths or path components without interrupting cell traffic. The system detects and selectively flushes defective paths without impacting paths that are working properly. The system initializes new switching fabrics automatically without using software to set values. Thus, the system tests online paths and corrects errors without going offline.
1. A method for performing offline fabric diagnostics, comprising:
removing a switching fabric from a switching fabric configuration so that the removed switching fabric is disabled from receiving data cells from an ingress, wherein the switching fabric configuration identifies active switching fabrics which are enabled to receive the data cells from the ingress;
generating a diagnostic cell;
simultaneously transmitting:
the diagnostic cell from the ingress across the disabled switching fabric; and
the data cells across the active switching fabrics identified in the switching fabric configuration;
receiving the diagnostic cell at an egress;
determining that the diagnostic cell was correctly received; and
adding the disabled switching fabric to the switching fabric configuration so that the added switching fabric is enabled to receive subsequent data cells from the ingress.
2. The method of claim 1 , further comprising:
determining the switching fabric comprises a stuck path, wherein a packet fails to reach a destination over the stuck path of the switching fabric;
generating a flush cell;
selectively flushing the stuck path of the switching fabric; and
enabling switching fabrics according to the switching fabric configuration.
3. The method of claim 1 , wherein said removing further comprises:
selecting a switching fabric for offline diagnostics; and
altering an active fabric mask associated with the switching fabric configuration, wherein the active fabric mask comprises a bit to indicate whether the selected switching fabric is off-line and disabled for data transfer, or whether the selected switching fabric is online and enabled for data transfer.
4. The method of claim 1 , wherein the diagnostic cell includes test data, said determining further comprising:
determining that a previous diagnostic cell was not skipped; and
determining that test data associated with the diagnostic cell was received correctly.
5. A traffic switching system configured to perform offline fabric diagnostics, comprising:
a plurality of switching fabrics;
an ingress module, coupled to the plurality of switching fabrics, including:
an ingress diagnostics module configured to:
remove a switching fabric of the plurality of switching fabrics from a switching fabric configuration so that the removed switching fabric is disabled from receiving data cells from the ingress module, wherein the switching fabric configuration identifies active switching fabrics which are enabled to receive the data cells from the ingress; and
generate a diagnostic cell;
a transmission module configured to simultaneously transmit:
the diagnostic cell from the ingress module across the disabled switching fabric; and
the data cells across the active switching fabrics identified in the switching fabric configuration;
an egress module, coupled to the switching fabrics, including:
a receive module configured to receive cells, including the diagnostic cell and the data cells, from the switching fabrics;
a packet constructor configured to reassemble packets from cells received across the switching fabrics according to the switching fabric configuration; and
an egress diagnostics module configured to:
determine that the diagnostic cell was correctly received; and
add the disabled switching fabric to the switching fabric configuration so that the added switching fabric is enabled to receive subsequent data cells from the ingress.
6. The system of claim 5 , wherein the ingress diagnostics module is further configured to determine the plurality of switching fabrics comprises a stuck path, wherein a packet fails to reach a destination over the stuck path of the plurality of switching fabrics, and to generate a flush cell to selectively flush the stuck path of the plurality of switching fabrics.
7. The system of claim 5 , wherein the diagnostic cell is a second diagnostic cell of a diagnostic cell sequence, and wherein the egress diagnostics module is further configured to:
determine that a first diagnostic cell of the diagnostic cell sequence was received; and
determine that test data associated with the second diagnostic cell was received correctly.