Description
This article describes how to check the Network Performance Metrics of the FortiGate interface running on AWS. The ENA driver provides the metrics to the instance (FortiGate). The version of the ENA driver depends on the version of FortiOS. The output of the commands below was run from the c6i.large instance.
Scope
FortiGate v7.2.x.
Solution
Run the command 'diag hardware deviceinfo nic-stats <port>'.
diag hardware deviceinfo nic-stats port1
port1 statistics:
tx_timeout: 0
wd_expired: 0
interface_up: 1
interface_down: 0
admin_q_pause: 0
bw_in_allowance_exceeded: 0
bw_out_allowance_exceeded: 0
pps_allowance_exceeded: 0
conntrack_allowance_exceeded: 0
linklocal_allowance_exceeded: 0
queue_0_tx_cnt: 11706
queue_0_tx_bytes: 10108986
queue_0_tx_queue_stop: 0
queue_0_tx_queue_wakeup: 0
queue_0_tx_dma_mapping_err: 0
queue_0_tx_linearize: 0
queue_0_tx_linearize_failed: 0
queue_0_tx_napi_comp: 8681
queue_0_tx_tx_poll: 8681
queue_0_tx_doorbells: 11706
queue_0_tx_prepare_ctx_err: 0
queue_0_tx_bad_req_id: 0
queue_0_tx_llq_buffer_copy: 0
queue_0_tx_missed_tx: 0
queue_0_tx_unmask_interrupt: 8681
queue_0_rx_cnt: 6810
queue_0_rx_bytes: 3386605
queue_0_rx_rx_copybreak_pkt: 4639
queue_0_rx_csum_good: 6585
queue_0_rx_refil_partial: 0
queue_0_rx_bad_csum: 0
queue_0_rx_page_alloc_fail: 0
queue_0_rx_skb_alloc_fail: 0
queue_0_rx_dma_mapping_err: 0
queue_0_rx_bad_desc_num: 0
queue_0_rx_bad_req_id: 0
queue_0_rx_empty_rx_ring: 0
queue_0_rx_csum_unchecked: 178
queue_0_rx_lpc_warm_up: 0
queue_0_rx_lpc_full: 0
queue_0_rx_lpc_wrong_numa: 0
queue_1_tx_cnt: 29701
queue_1_tx_bytes: 1682549
queue_1_tx_queue_stop: 0
queue_1_tx_queue_wakeup: 0
queue_1_tx_dma_mapping_err: 0
queue_1_tx_linearize: 0
queue_1_tx_linearize_failed: 0
queue_1_tx_napi_comp: 22446
queue_1_tx_tx_poll: 22446
queue_1_tx_doorbells: 29701
queue_1_tx_prepare_ctx_err: 0
queue_1_tx_bad_req_id: 0
queue_1_tx_llq_buffer_copy: 0
queue_1_tx_missed_tx: 0
queue_1_tx_unmask_interrupt: 22446
queue_1_rx_cnt: 77873
queue_1_rx_bytes: 113756811
queue_1_rx_rx_copybreak_pkt: 1074
queue_1_rx_csum_good: 77869
queue_1_rx_refil_partial: 0
queue_1_rx_bad_csum: 0
queue_1_rx_page_alloc_fail: 0
queue_1_rx_skb_alloc_fail: 0
queue_1_rx_dma_mapping_err: 0
queue_1_rx_bad_desc_num: 0
queue_1_rx_bad_req_id: 0
queue_1_rx_empty_rx_ring: 0
queue_1_rx_csum_unchecked: 4
queue_1_rx_lpc_warm_up: 0
queue_1_rx_lpc_full: 0
queue_1_rx_lpc_wrong_numa: 0
ena_admin_q_aborted_cmd: 0
ena_admin_q_submitted_cmd: 41
ena_admin_q_completed_cmd: 41
ena_admin_q_out_of_space: 0
ena_admin_q_no_completion: 0
diag hardware deviceinfo nic-stats port1 | grep exceeded
bw_in_allowance_exceeded: 0
bw_out_allowance_exceeded: 0
pps_allowance_exceeded: 0
conntrack_allowance_exceeded: 0
linklocal_allowance_exceeded: 0
get hardware nic port1
Name: port1
Driver: ena
Version: 2.6.1g
Bus: 0000:00:05.0
Under load *_exceeded counters may increase, which will lead to packet drops. This can be verified by periodically running 'diag hardware deviceinfo nic-stats <port>'.
diagnose hardware deviceinfo nic-stats port1 | grep exceed
bw_in_allowance_exceeded: 0
bw_out_allowance_exceeded: 0
pps_allowance_exceeded: 200
conntrack_allowance_exceeded: 132016
linklocal_allowance_exceeded: 0
diagnose hardware deviceinfo nic-stats port1 | grep exceed
bw_in_allowance_exceeded: 0
bw_out_allowance_exceeded: 0
pps_allowance_exceeded: 254
conntrack_allowance_exceeded: 887231
linklocal_allowance_exceeded: 0
Related documents:
