thiagoofsaint
(usa Debian)
Enviado em 22/03/2017 - 16:26h
Há pouco recomeçou a lentidão. Não dei ainda um stop no apache2, mas dei um tcpdump -i ethX dst port 80, ao mesmo tempo que dou um ping para o site da uol. Reparei que quando o ping dá tempo esgotado, a tela do tcpdump dispara como no exemplo abaixo:
16:13:06.931637 IP server.ghb.net.br.20453 > 45.117.193.248.www: Flags [S], seq 1340437605:1340438368, win 63353, length 763
16:13:06.931674 IP server.ghb.net.br.29043 > 45.117.193.248.www: Flags [S], seq 1903378287:1903379048, win 61878, length 761
16:13:06.931705 IP server.ghb.net.br.41010 > 45.117.193.248.www: Flags [S], seq 2687650571:2687651343, win 61890, length 772
16:13:06.931734 IP server.ghb.net.br.15494 > 45.117.193.248.www: Flags [S], seq 1015469076:1015469868, win 61154, length 792
16:13:06.931741 IP server.ghb.net.br.6092 > 45.117.193.248.www: Flags [S], seq 399263326:399264113, win 62432, length 787
16:13:06.931755 IP server.ghb.net.br.9879 > 45.117.193.248.www: Flags [S], seq 647491705:647492493, win 65484, length 788
16:13:06.931768 IP server.ghb.net.br.47092 > 45.117.193.248.www: Flags [S], seq 3086259011:3086259808, win 61762, length 797
16:13:06.931775 IP server.ghb.net.br.2284 > 45.117.193.248.www: Flags [S], seq 149724492:149725258, win 65074, length 766
16:13:06.931787 IP server.ghb.net.br.1712 > 45.117.193.248.www: Flags [S], seq 112231016:112231818, win 60782, length 802
16:13:06.931802 IP server.ghb.net.br.39648 > 45.117.193.248.www: Flags [S], seq 2598415666:2598416448, win 62148, length 782
16:13:06.931822 IP server.ghb.net.br.29970 > 45.117.193.248.www: Flags [S], seq 1964166715:1964167488, win 64725, length 773
16:13:06.931829 IP server.ghb.net.br.22576 > 45.117.193.248.www: Flags [S], seq 1479557144:1479557938, win 61306, length 794
16:13:06.931840 IP server.ghb.net.br.ssh > 45.117.193.248.www: Flags [S], seq 1472289:1473053, win 60097, length 764
16:13:06.931847 IP server.ghb.net.br.32662 > 45.117.193.248.www: Flags [S], seq 2140555115:2140555878, win 65317, length 763
16:13:06.931859 IP server.ghb.net.br.12907 > 45.117.193.248.www: Flags [S], seq 845907974:845908748, win 65302, length 774
16:13:06.931873 IP server.ghb.net.br.33011 > 45.117.193.248.www: Flags [S], seq 2163417872:2163418678, win 61451, length 806
16:13:06.931892 IP server.ghb.net.br.10265 > 45.117.193.248.www: Flags [S], seq 672759660:672760428, win 60905, length 768
16:13:06.931899 IP server.ghb.net.br.52170 > 45.117.193.248.www: Flags [S], seq 3419048053:3419048828, win 62100, length 775
16:13:06.931911 IP server.ghb.net.br.62381 > 45.117.193.248.www: Flags [S], seq 4088249727:4088250518, win 60251, length 791
16:13:06.931923 IP server.ghb.net.br.15721 > 45.117.193.248.www: Flags [S], seq 1030316571:1030317378, win 64801, length 807
16:13:06.931930 IP server.ghb.net.br.26043 > 45.117.193.248.www: Flags [S], seq 1706785637:1706786428, win 63693, length 791
16:13:06.931950 IP server.ghb.net.br.23072 > 45.117.193.248.www: Flags [S], seq 1512093741:1512094523, win 62747, length 782
16:13:06.931957 IP server.ghb.net.br.15751 > 45.117.193.248.www: Flags [S], seq 1032312329:1032313118, win 61411, length 789
Aí quando o ping responde, o tcpdump volta ao normal, mas quando dá tempo esgotado de novo, recomeça esse erro acima, mas com outro IP.
Outra coisa que eu observei também foi o comando top. o processo que aparece no topo pra mim é desconhecido, um tal de fhybshcazp
top - 16:25:30 up 20:00, 5 users, load average: 3.68, 3.58, 3.05
Tasks: 168 total, 1 running, 167 sleeping, 0 stopped, 0 zombie
Cpu(s): 51.0%us, 39.4%sy, 0.0%ni, 0.0%id, 0.0%wa, 0.0%hi, 9.6%si, 0.0%st
Mem: 1885068k total, 1174008k used, 711060k free, 292964k buffers
Swap: 3447800k total, 0k used, 3447800k free, 545968k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
3132 root 20 0 43296 1176 212 S 61.6 0.1 712:09.00 fhybshcazp
4333 root 20 0 3888 1344 908 S 20.9 0.1 5:46.05 iptraf
1666 root 20 0 100m 28m 5952 S 10.3 1.5 131:20.84 Xorg
3508 server 20 0 85616 19m 14m S 6.6 1.1 85:52.98 gnome-system-mo
21 root 20 0 0 0 0 S 0.3 0.0 0:43.12 kondemand/0
1 root 20 0 2036 724 628 S 0.0 0.0 0:08.82 init
Como posso resolver isso? Pois quando isso acontece, o servidor chega até a 600.0 Mbit/s