Multiple testbeds are affected, failure does not happen always, so perhaps more testbeds are susceptible, but did not fail as those tests are in coverage.
Seen on 2n-icx [0], 3n-icx [1], 3na-spr [2].
Losses on 3n-tsh [3] are smaller and affecting different tests, so not sure it is the same issue.

[0] https://s3-logs.fd.io/vex-yul-rot-jenkins-1/csit-dpdk-perf-report-coverage-2310-2n-icx/1/log.html.gz#s1-s1-s1-s3-t9-k2-k5-k16-k9
[1] https://s3-logs.fd.io/vex-yul-rot-jenkins-1/csit-dpdk-perf-report-coverage-2310-3n-icx/1/log.html.gz#s1-s1-s1-s1-t9-k2-k5-k16-k9
[2] https://s3-logs.fd.io/vex-yul-rot-jenkins-1/csit-dpdk-perf-report-coverage-2310-3na-spr/1/log.html.gz#s1-s1-s1-s1-t9-k2-k5-k16-k9
[3] https://s3-logs.fd.io/vex-yul-rot-jenkins-1/csit-dpdk-perf-report-coverage-2310-3n-tsh/1/log.html.gz#s1-s1-s1-s1-t7-k2-k5-k17-k9