. > > Based on previous answers, diskd is for freebsd with 1 process only, > when > the ufs/aufs are with many processes. > Also, as you said, it seems the diskd process was modified with the > latest > builds... > I don't know about freebsd, diskd is a separate process with a light consumption Top with 3000 simultaneous users (2 x caches 250 Go full) top - 15:10:45 up 65 days, 43 min, 3 users, load average: 1,88, 1,83, 2,14 Tasks: 3194 total, 3 running, 3191 sleeping, 0 stopped, 0 zombie %Cpu(s): 11,5 us, 7,8 sy, 0,0 ni, 68,8 id, 8,2 wa, 0,0 hi, 3,7 si, 0,0 st KiB Mem: 33006984 total, 32338408 used, 668576 free, 4006796 buffers KiB Swap: 1952764 total, 15640 used, 1937124 free, 4555212 cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 1729 squid 20 0 14,7g 14g 4924 R 70,5 46,3 127:53.77 squid 21011 e2guardi 20 0 56288 18m 1180 S 14,2 0,1 18:20.66 e2guardian 14004 root 20 0 25992 4392 1180 R 3,2 0,0 0:00.41 top 1849 squid 20 0 22864 1360 1144 S 1,3 0,0 2:24.98 diskd 1850 squid 20 0 22864 1360 1144 S 1,3 0,0 2:20.31 diskd As you can see 2 caches = two process. And also 3000 e2 process With this configuration and aufs, and no change, the load is 4/5 and the CPU 80/90, global response time still the same I don't know if aufs is more fast than disk to delivers the files but enough for me _______________________________________________ squid-users mailing list squid-users@xxxxxxxxxxxxxxxxxxxxx http://lists.squid-cache.org/listinfo/squid-users
