HAProxy Enterprise Documentation 2.7r1

show profiling

Dump profiling details.


Specify the category of data desired: status, tasks, memory, or all.

By default, output is sorted by usage. As an alternative, you can specify the byaddr or bytime argument to sort by address or by total execution time, respectively.

Specify the aggr argument to aggregate task activity by called function instead of displaying all details.

By default, all profiling information is dumped. To limit the number of lines of output for each category, append a maximum line limit.


Show profiling settings, one per line, as well as the command needed to change them.

$ echo "show profiling status" | \
   sudo socat stdio unix-connect:/var/run/hapee-2.7/hapee-lb.sock
Per-task CPU profiling              : on            # set profiling tasks {on|auto|off}
Memory usage profiling              : off           # set profiling memory {on|off}

Report details for the top three tasks, sorted by time:

$ echo "show profiling tasks bytime 3" | \
   sudo socat stdio unix-connect:/var/run/hapee-2.7/hapee-lb.sock
Tasks activity:
function                      calls   cpu_tot   cpu_avg   lat_tot   lat_avg
process_chk                     358   17.97ms   50.20us   15.71ms   43.89us <- wake_expired_tasks@src/task.c:350 task_wakeup
process_chk                     360   3.684ms   10.23us   3.584ms   9.955us <- wake_srv_chk@src/check.c:1048 task_wakeup
task_run_applet                   5   1.421ms   284.2us   14.85us   2.970us <- sc_applet_create@src/stconn.c:491 appctx_wakeup

See also

Next up

show quic