Servidor entrando em swap

onovaes
(usa Debian)

Enviado em 30/12/2012 - 20:55h

Olá Pessoal,

Tenho um servidor web com debian squeeze + nginx + python + django.

Acontece que ela com uns 10 dias de uso, começa a entrar em swap e depois dá excesso de IO de disco.

Alguma dica pra eu começar a debugar o que pode estar acontecendo?

danniel-lara
(usa Fedora)

Enviado em 30/12/2012 - 21:37h

qual a configuração da maquina ?
como ficou o particionamento ?

carezia
(usa Debian)

Enviado em 30/12/2012 - 23:15h

Digite no servidor:

top

Digite M para ordenar por consumo de memória. Copie a tela e poste aqui.

onovaes
(usa Debian)

Enviado em 30/12/2012 - 23:48h

top - 20:47:01 up 2 days, 13:20, 1 user, load average: 0.00, 0.01, 0.05
Tasks: 99 total, 2 running, 97 sleeping, 0 stopped, 0 zombie
Cpu(s): 3.2%us, 0.3%sy, 0.0%ni, 96.5%id, 0.0%wa, 0.0%hi, 0.0%si, 0.1%st
Mem: 508252k total, 486476k used, 21776k free, 18760k buffers
Swap: 262140k total, 1752k used, 260388k free, 117596k cached


É um VPS linode 512.

carezia
(usa Debian)

Enviado em 31/12/2012 - 00:07h

Você não postou a tela, postou apenas o cabeçalho.

onovaes
(usa Debian)

Enviado em 31/12/2012 - 00:15h

Dei um boot na maquina agora, e liberou a swap, agora daqui uns dez dias volta e eu tenho que botar ela novamente.

Segue a tela

top - 21:14:20 up 6 min, 1 user, load average: 0.05, 0.04, 0.03
Tasks: 87 total, 1 running, 86 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.3%us, 0.1%sy, 0.0%ni, 99.6%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 508252k total, 159932k used, 348320k free, 7204k buffers
Swap: 262140k total, 0k used, 262140k free, 76708k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
2034 root 20 0 37404 19m 3216 S 1 4.0 0:11.66 uwsgi
1810 mysql 20 0 134m 18m 5640 S 1 3.7 0:01.56 mysqld
2126 root 20 0 2496 1068 824 R 0 0.2 0:00.01 top
1 root 20 0 2088 708 612 S 0 0.1 0:00.95 init
2 root 20 0 0 0 0 S 0 0.0 0:00.00 kthreadd
3 root 20 0 0 0 0 S 0 0.0 0:00.00 ksoftirqd/0
4 root 20 0 0 0 0 S 0 0.0 0:00.00 kworker/0:0
5 root 0 -20 0 0 0 S 0 0.0 0:00.00 kworker/0:0H
6 root 20 0 0 0 0 S 0 0.0 0:00.04 kworker/u:0
7 root 0 -20 0 0 0 S 0 0.0 0:00.00 kworker/u:0H
8 root RT 0 0 0 0 S 0 0.0 0:01.37 migration/0
9 root RT 0 0 0 0 S 0 0.0 0:00.00 migration/1
10 root 20 0 0 0 0 S 0 0.0 0:00.00 kworker/1:0
11 root 0 -20 0 0 0 S 0 0.0 0:00.00 kworker/1:0H
12 root 20 0 0 0 0 S 0 0.0 0:00.00 ksoftirqd/1
13 root RT 0 0 0 0 S 0 0.0 0:00.00 migration/2
14 root 20 0 0 0 0 S 0 0.0 0:00.00 kworker/2:0
15 root 0 -20 0 0 0 S 0 0.0 0:00.00 kworker/2:0H
16 root 20 0 0 0 0 S 0 0.0 0:00.00 ksoftirqd/2
17 root RT 0 0 0 0 S 0 0.0 0:00.00 migration/3
18 root 20 0 0 0 0 S 0 0.0 0:00.00 kworker/3:0
19 root 0 -20 0 0 0 S 0 0.0 0:00.00 kworker/3:0H
20 root 20 0 0 0 0 S 0 0.0 0:00.00 ksoftirqd/3
21 root 0 -20 0 0 0 S 0 0.0 0:00.00 cpuset
22 root 0 -20 0 0 0 S 0 0.0 0:00.00 khelper
23 root 20 0 0 0 0 S 0 0.0 0:00.01 kdevtmpfs
24 root 20 0 0 0 0 S 0 0.0 0:00.00 kworker/u:1
26 root 20 0 0 0 0 S 0 0.0 0:00.00 xenwatch
27 root 20 0 0 0 0 S 0 0.0 0:00.00 xenbus
172 root 20 0 0 0 0 S 0 0.0 0:00.00 bdi-default
174 root 0 -20 0 0 0 S 0 0.0 0:00.00 kblockd
182 root 20 0 0 0 0 S 0 0.0 0:00.01 kworker/3:1
186 root 0 -20 0 0 0 S 0 0.0 0:00.00 md
270 root 0 -20 0 0 0 S 0 0.0 0:00.00 rpciod
272 root 20 0 0 0 0 S 0 0.0 0:00.02 kworker/2:1
284 root 20 0 0 0 0 S 0 0.0 0:00.00 kswapd0

carezia
(usa Debian)

Enviado em 03/01/2013 - 22:39h

O processo que mais ocupa memória é o uwsgi (python). Como fica o top quando a máquina entra em swap?

andrecanhadas
(usa Debian)

Enviado em 03/01/2013 - 23:58h

Pode colocar no cron diariamente para limpar o cache da RAM :

crie um script em /usr/local/bin/memofree.sh

echo 3 > /proc/sys/vm/drop_caches

 

permissão de executar : # chmod +x /usr/local/bin/memofree.sh


agende no cron com crontab -e (como root) :

00 23 * * */usr/local/bin/memofree.sh

 

Não vai corrigir o problema mas é uma gambiarra que vai funcionar até descobrir o que esta errado

onovaes
(usa Debian)

Enviado em 10/01/2013 - 16:18h

Veja o top + m com ele em SWAP e dando I/O de disco.



top - 13:16:38 up 10 days, 16:08, 1 user, load average: 0.70, 0.69, 0.77
Tasks: 90 total, 1 running, 89 sleeping, 0 stopped, 0 zombie
Cpu(s): 1.9%us, 2.6%sy, 0.0%ni, 59.6%id, 35.9%wa, 0.0%hi, 0.0%si, 0.0%st

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
2034 root 20 0 504m 447m 384 D 14 90.2 524:40.73 uwsgi
284 root 20 0 0 0 0 D 5 0.0 5:41.83 kswapd0
975 root 0 -20 0 0 0 S 1 0.0 0:29.51 kworker/0:1H
1 root 20 0 2088 56 36 S 0 0.0 0:12.99 init
2 root 20 0 0 0 0 S 0 0.0 0:00.00 kthreadd
3 root 20 0 0 0 0 S 0 0.0 0:00.87 ksoftirqd/0
4 root 20 0 0 0 0 S 0 0.0 0:00.00 kworker/0:0
5 root 0 -20 0 0 0 S 0 0.0 0:00.00 kworker/0:0H
6 root 20 0 0 0 0 S 0 0.0 0:00.04 kworker/u:0
7 root 0 -20 0 0 0 S 0 0.0 0:00.00 kworker/u:0H
8 root RT 0 0 0 0 S 0 0.0 0:01.38 migration/0
9 root RT 0 0 0 0 S 0 0.0 0:00.03 migration/1
10 root 20 0 0 0 0 S 0 0.0 0:00.00 kworker/1:0
11 root 0 -20 0 0 0 S 0 0.0 0:00.00 kworker/1:0H
12 root 20 0 0 0 0 S 0 0.0 0:00.43 ksoftirqd/1
13 root RT 0 0 0 0 S 0 0.0 0:00.02 migration/2
14 root 20 0 0 0 0 S 0 0.0 0:00.00 kworker/2:0
15 root 0 -20 0 0 0 S 0 0.0 0:00.00 kworker/2:0H
16 root 20 0 0 0 0 S 0 0.0 0:00.44 ksoftirqd/2
17 root RT 0 0 0 0 S 0 0.0 0:00.02 migration/3
18 root 20 0 0 0 0 S 0 0.0 0:00.00 kworker/3:0
19 root 0 -20 0 0 0 S 0 0.0 0:00.00 kworker/3:0H
20 root 20 0 0 0 0 S 0 0.0 0:00.39 ksoftirqd/3
21 root 0 -20 0 0 0 S 0 0.0 0:00.00 cpuset
22 root 0 -20 0 0 0 S 0 0.0 0:00.00 khelper
23 root 20 0 0 0 0 S 0 0.0 0:00.01 kdevtmpfs
24 root 20 0 0 0 0 S 0 0.0 0:00.00 kworker/u:1
26 root 20 0 0 0 0 S 0 0.0 0:00.00 xenwatch
27 root 20 0 0 0 0 S 0 0.0 0:00.00 xenbus
172 root 20 0 0 0 0 S 0 0.0 0:00.11 bdi-default
174 root 0 -20 0 0 0 S 0 0.0 0:00.00 kblockd
182 root 20 0 0 0 0 S 0 0.0 0:28.24 kworker/3:1

carezia
(usa Debian)

Enviado em 10/01/2013 - 16:35h

Instale o iotop e rode assim alguns segundos (enquanto estiver entrando em swap):

iotop -b -o

Mande a saída.