Hello,
I have trouble understanding stats reports from our HAproxy servers, can
anyone please shed some light on this ?
1. On a backend with only one server, scur(BACKEND) > scur(server). How
can this be ?
#
pxname,svname,qcur,qmax,scur,smax,slim,stot,bin,bout,dreq,dresp,ereq,econ,eresp,wretr,wredis,status,weight,act,bck,chkfail,chkdown,lastchg,downtime,qlimit,pid,iid,sid,throttle,lbtot,tracked,type,rate,rate_lim,rate_max,check_status,check_code,check_duration,hrsp_1xx,hrsp_2xx,hrsp_3xx,hrsp_4xx,hrsp_5xx,hrsp_other,hanafail,req_rate,req_rate_max,req_tot,cli_abrt,srv_abrt,comp_in,comp_out,comp_byp,comp_rsp,lastsess,last_chk,last_agt,qtime,ctime,rtime,ttime,
backend_b,server_a,0,0,1,1,1,92433,167970344,127389583,,0,,0,0,0,0,UP,1,1,0,,,,,512,1,20,1,,6373,backend_a/server_a,2,3,,29,,,,0,92432,0,0,0,0,0,,,,0,0,,,,,0,,,12858,2,530,13826,
backend_b,BACKEND,48,142,49,143,410,92481,167970344,127389583,0,0,,0,0,0,0,UP,1,1,0,,0,88182,0,,1,20,0,,6373,,1,0,,65,,,,0,92432,0,0,0,0,,,,,0,0,0,0,0,0,183,,,12858,2,530,13826,
2. Same HAproxy instance, on another backend, smax(BACKEND) >
slim(server_a) + slim(server_b). Again, how is this possible ? It seems
like smax(BACKEND) = smax(server_a) + smax(server_b) + qmax(BACKEND) ?
#
pxname,svname,qcur,qmax,scur,smax,slim,stot,bin,bout,dreq,dresp,ereq,econ,eresp,wretr,wredis,status,weight,act,bck,chkfail,chkdown,lastchg,downtime,qlimit,pid,iid,sid,throttle,lbtot,tracked,type,rate,rate_lim,rate_max,check_status,check_code,check_duration,hrsp_1xx,hrsp_2xx,hrsp_3xx,hrsp_4xx,hrsp_5xx,hrsp_other,hanafail,req_rate,req_rate_max,req_tot,cli_abrt,srv_abrt,comp_in,comp_out,comp_byp,comp_rsp,lastsess,last_chk,last_agt,qtime,ctime,rtime,ttime,
backend_c,server_a,0,0,0,3,3,95541,232232183,241897167,,0,,0,0,0,0,UP,1,1,0,,,,,512,1,19,1,,95507,backend_a/server_a,2,0,,5,,,,0,95541,0,0,0,0,0,,,,1,0,,,,,2,,,0,1,706,2028,
backend_c,server_b,0,0,0,3,3,95587,232144919,241789273,,0,,0,0,0,0,UP,1,1,0,,,,,512,1,19,2,,95546,backend_a/server_b,2,0,,5,,,,0,95587,0,0,0,0,0,,,,0,0,,,,,3,,,0,2,675,2082,
backend_c,BACKEND,0,3,0,9,410,191128,464377102,483686440,0,0,,0,0,0,0,UP,2,2,0,,0,89074,0,,1,19,0,,191053,,1,0,,10,,,,0,191128,0,0,0,0,,,,,1,0,0,0,0,0,2,,,0,1,678,2059,
3. qmax stays at zero on each backend server, even though qmax(BACKEND)
> 0. Are queue stats not maintained for each server ?
#
pxname,svname,qcur,qmax,scur,smax,slim,stot,bin,bout,dreq,dresp,ereq,econ,eresp,wretr,wredis,status,weight,act,bck,chkfail,chkdown,lastchg,downtime,qlimit,pid,iid,sid,throttle,lbtot,tracked,type,rate,rate_lim,rate_max,check_status,check_code,check_duration,hrsp_1xx,hrsp_2xx,hrsp_3xx,hrsp_4xx,hrsp_5xx,hrsp_other,hanafail,req_rate,req_rate_max,req_tot,cli_abrt,srv_abrt,comp_in,comp_out,comp_byp,comp_rsp,lastsess,last_chk,last_agt,qtime,ctime,rtime,ttime,
backend_c,server_a,0,0,0,3,3,95541,232232183,241897167,,0,,0,0,0,0,UP,1,1,0,,,,,512,1,19,1,,95507,backend_a/server_a,2,0,,5,,,,0,95541,0,0,0,0,0,,,,1,0,,,,,2,,,0,1,706,2028,
backend_c,server_b,0,0,0,3,3,95587,232144919,241789273,,0,,0,0,0,0,UP,1,1,0,,,,,512,1,19,2,,95546,backend_a/server_b,2,0,,5,,,,0,95587,0,0,0,0,0,,,,0,0,,,,,3,,,0,2,675,2082,
backend_c,BACKEND,0,3,0,9,410,191128,464377102,483686440,0,0,,0,0,0,0,UP,2,2,0,,0,89074,0,,1,19,0,,191053,,1,0,,10,,,,0,191128,0,0,0,0,,,,,1,0,0,0,0,0,2,,,0,1,678,2059,
4. On another HAproxy instance, on a backend with two servers, we always
have scur=slim on both servers, but qcur stays at zero.
The application servers tell us their threads are ready (zero busy
thread). Is there any way to show more info about the sessions which
HAproxy thinks are in use ?
#
pxname,svname,qcur,qmax,scur,smax,slim,stot,bin,bout,dreq,dresp,ereq,econ,eresp,wretr,wredis,status,weight,act,bck,chkfail,chkdown,lastchg,downtime,qlimit,pid,iid,sid,throttle,lbtot,tracked,type,rate,rate_lim,rate_max,check_status,check_code,check_duration,hrsp_1xx,hrsp_2xx,hrsp_3xx,hrsp_4xx,hrsp_5xx,hrsp_other,hanafail,req_rate,req_rate_max,req_tot,cli_abrt,srv_abrt,comp_in,comp_out,comp_byp,comp_rsp,lastsess,last_chk,last_agt,qtime,ctime,rtime,ttime,
backend_a,server_a,0,0,12,12,12,90,114944,68086,,0,,0,0,0,0,UP,1,1,0,18,6,67872,1015,128,1,9,1,,88,,2,0,,8,L7OK,200,5,0,78,0,0,0,0,0,,,,0,0,,,,,501099,OK,,3,1,42,55,
backend_a,server_b,0,0,12,12,12,72,87204,47593,,0,,0,0,0,0,UP,1,1,0,25,6,67842,1064,128,1,9,2,,72,,2,0,,7,L7OK,200,4,0,60,0,0,0,0,0,,,,0,0,,,,,501461,OK,,0,1,28,36,
backend_a,BACKEND,0,3,24,27,410,181,223139,119707,0,0,,5,0,0,0,UP,2,2,0,,6,67872,862,,1,9,0,,160,,1,0,,15,,,,0,138,0,0,19,0,,,,,5,0,0,0,0,0,501099,,,3,1,66,86,
BTW, we are running HAproxy 1.5.9 from the vbernat Ubuntu PPA.
Thanks in advance.
Sylvain
