QuestionQ44

VerifiedIdentify root cause of issue/issues.

Refer to partial exhibit XX:

Question Image

The customer uses Gen5 B-Series SAN switches and is experiencing reduced throughput and latency issues on the HPE block-storage solution across a stretch-cluster solution. The bandwidth appears sufficient.

Based on the ISL port-statistics example, what is a probable cause of the described issue?

Explanation

A high tim_txcrd_z value shows that transmit buffer-to-buffer credits repeatedly reached zero while frames were queued for transmission. This credit starvation stalls transmission and produces back-pressure, reduced throughput, and latency despite sufficient physical link bandwidth. Broadcom identifies increasing BB Credit Zero counts on an ISL as an indication of downstream congestion or a credit-stalled device.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!