SQUID PARANDO

cprvendas
(usa Conectiva)

Enviado em 24/01/2009 - 08:12h

Olá, a tds do VOL.
Estou com um problema meio eskisito aki na empresa. Colokei um CL10 com squid + iptables, tava funcionando belezinha. Dê uns dias pra cá, começou a travar o squid do nada, o msn funciona! mas os sites NAO!

Se eu der um "service squid restart" volta a funcionar! Deve ser algo com a cache, mas não sei onde mexer e tenho medo de dar pau em tudo!
Obs: ele volta a funcionar sozinho depois de um tempo.. Muito estranho!

Se alguem tiver alguma dica, eu agradeço...
Obrigado.

CACHE.LOG
2009/01/21 15:40:16| Starting Squid Cache version 2.5.STABLE5 for i686-pc-linux-gnu...
2009/01/21 15:40:16| Process ID 3565
2009/01/21 15:40:16| With 1024 file descriptors available
2009/01/21 15:40:16| Performing DNS Tests...
2009/01/21 15:40:16| Successful DNS name lookup tests...
2009/01/21 15:40:16| DNS Socket created at 0.0.0.0, port 32768, FD 4
2009/01/21 15:40:16| Adding nameserver 200.175.5.139 from /etc/resolv.conf
2009/01/21 15:40:16| Adding nameserver 200.175.89.139 from /etc/resolv.conf
2009/01/21 15:40:16| Unlinkd pipe opened on FD 9
2009/01/21 15:40:16| Swap maxSize 102400 KB, estimated 7876 objects
2009/01/21 15:40:16| Target number of buckets: 393
2009/01/21 15:40:16| Using 8192 Store buckets
2009/01/21 15:40:16| Max Mem size: 8192 KB
2009/01/21 15:40:16| Max Swap size: 102400 KB
2009/01/21 15:40:16| Rebuilding storage in /var/cache/squid (DIRTY)
2009/01/21 15:40:16| Using Least Load store dir selection
2009/01/21 15:40:16| Current Directory is /
2009/01/21 15:40:16| Loaded Icons.
2009/01/21 15:40:17| Accepting HTTP connections at 0.0.0.0, port 3128, FD 11.
2009/01/21 15:40:17| Accepting ICP messages at 0.0.0.0, port 3130, FD 12.
2009/01/21 15:40:17| Accepting HTCP messages on port 4827, FD 13.
2009/01/21 15:40:17| Accepting SNMP messages on port 3401, FD 14.
2009/01/21 15:40:17| WCCP Disabled.
2009/01/21 15:40:17| Pinger socket opened on FD 16
2009/01/21 15:40:17| Ready to serve requests.
2009/01/21 15:40:17| Store rebuilding is 33.1% complete
2009/01/21 15:40:17| Done reading /var/cache/squid swaplog (12362 entries)
2009/01/21 15:40:17| Finished rebuilding storage from disk.
2009/01/21 15:40:17| 11082 Entries scanned
2009/01/21 15:40:17| 0 Invalid entries.
2009/01/21 15:40:17| 0 With invalid flags.
2009/01/21 15:40:17| 11082 Objects loaded.
2009/01/21 15:40:17| 0 Objects expired.
2009/01/21 15:40:17| 0 Objects cancelled.
2009/01/21 15:40:17| 0 Duplicate URLs purged.
2009/01/21 15:40:17| 0 Swapfile clashes avoided.
2009/01/21 15:40:17| Took 0.9 seconds (12898.2 objects/sec).
2009/01/21 15:40:17| Beginning Validation Procedure
2009/01/21 15:40:17| Completed Validation Procedure
2009/01/21 15:40:17| Validated 11082 Entries
2009/01/21 15:40:17| store_swap_size = 139000k
2009/01/21 15:40:18| storeLateRelease: released 0 objects
2009/01/21 15:40:18| WARNING: Disk space over limit: 138628 KB > 102400 KB
2009/01/21 15:40:19| icmpRecv: recv: (111) Connection refused
2009/01/21 15:40:19| Closing Pinger socket on FD 16
2009/01/21 15:40:29| WARNING: Disk space over limit: 129536 KB > 102400 KB
2009/01/21 15:40:40| WARNING: Disk space over limit: 123468 KB > 102400 KB
2009/01/21 15:40:51| WARNING: Disk space over limit: 119004 KB > 102400 KB
2009/01/21 15:41:03| WARNING: Disk space over limit: 110388 KB > 102400 KB
2009/01/21 15:42:25| sslReadServer: FD 56: read failure: (104) Connection reset by peer
2009/01/21 15:42:26| sslReadServer: FD 54: read failure: (104) Connection reset by peer
2009/01/21 15:42:30| sslReadServer: FD 62: read failure: (104) Connection reset by peer
2009/01/21 15:43:25| sslReadServer: FD 27: read failure: (104) Connection reset by peer
2009/01/21 15:58:08| WARNING: 1 swapin MD5 mismatches
2009/01/21 16:07:13| urlParse: Illegal character in hostname '%25257bd30c1661-cdaf-11d0-8a3e-00c04fc9e26e%25257d'
2009/01/21 16:12:42| sslReadServer: FD 35: read failure: (104) Connection reset by peer
2009/01/21 16:12:42| sslReadServer: FD 31: read failure: (104) Connection reset by peer
2009/01/21 16:17:49| sslReadServer: FD 30: read failure: (104) Connection reset by peer
2009/01/21 16:17:49| sslReadServer: FD 35: read failure: (104) Connection reset by peer
2009/01/21 16:17:56| sslReadServer: FD 26: read failure: (104) Connection reset by peer
2009/01/21 16:17:56| sslReadServer: FD 32: read failure: (104) Connection reset by peer
2009/01/21 16:20:17| NETDB state saved; 0 entries, 0 msec
2009/01/21 17:01:57| NETDB state saved; 0 entries, 0 msec
2009/01/21 17:27:30| sslReadServer: FD 38: read failure: (104) Connection reset by peer
2009/01/21 17:27:43| sslReadServer: FD 42: read failure: (104) Connection reset by peer
2009/01/21 17:41:26| sslReadServer: FD 30: read failure: (104) Connection reset by peer
2009/01/21 17:41:34| sslReadServer: FD 29: read failure: (104) Connection reset by peer
2009/01/21 17:43:15| sslReadServer: FD 28: read failure: (104) Connection reset by peer
2009/01/21 17:43:24| sslReadServer: FD 30: read failure: (104) Connection reset by peer
2009/01/21 17:56:32| NETDB state saved; 0 entries, 0 msec
2009/01/21 18:56:03| NETDB state saved; 0 entries, 589 msec
2009/01/21 19:00:37| sslReadServer: FD 30: read failure: (104) Connection reset by peer
2009/01/21 19:00:39| sslReadServer: FD 65: read failure: (104) Connection reset by peer
2009/01/21 19:00:40| sslReadServer: FD 68: read failure: (104) Connection reset by peer
2009/01/21 19:00:41| sslReadServer: FD 67: read failure: (104) Connection reset by peer
2009/01/21 19:00:43| sslReadServer: FD 60: read failure: (104) Connection reset by peer
2009/01/21 19:00:43| sslReadServer: FD 62: read failure: (104) Connection reset by peer
2009/01/21 19:43:18| sslReadServer: FD 21: read failure: (104) Connection reset by peer
2009/01/21 19:54:14| NETDB state saved; 0 entries, 727 msec
2009/01/21 21:07:30| NETDB state saved; 0 entries, 0 msec
2009/01/21 22:10:14| NETDB state saved; 0 entries, 0 msec
2009/01/21 22:52:17| NETDB state saved; 0 entries, 0 msec
2009/01/21 23:10:14| sslReadServer: FD 20: read failure: (104) Connection reset by peer
2009/01/22 03:57:53| NETDB state saved; 0 entries, 839 msec
2009/01/22 05:02:36| NETDB state saved; 0 entries, 620 msec
2009/01/22 05:53:25| NETDB state saved; 0 entries, 0 msec
2009/01/22 06:56:33| NETDB state saved; 0 entries, 584 msec
2009/01/22 07:41:04| sslReadServer: FD 38: read failure: (104) Connection reset by peer
2009/01/22 07:41:17| sslReadServer: FD 35: read failure: (104) Connection reset by peer
2009/01/22 07:43:38| NETDB state saved; 0 entries, 0 msec
2009/01/22 07:51:08| sslReadServer: FD 34: read failure: (104) Connection reset by peer
2009/01/22 08:04:28| urlParse: Illegal character in hostname 'smartcobran%e7a'
2009/01/22 08:15:48| sslReadServer: FD 41: read failure: (104) Connection reset by peer
2009/01/22 08:28:07| NETDB state saved; 0 entries, 0 msec
2009/01/22 08:31:11| sslReadServer: FD 36: read failure: (104) Connection reset by peer
2009/01/22 08:59:22| sslReadServer: FD 34: read failure: (104) Connection reset by peer
2009/01/22 09:40:34| NETDB state saved; 0 entries, 0 msec
2009/01/22 09:51:16| sslReadServer: FD 25: read failure: (104) Connection reset by peer
2009/01/22 10:02:18| sslReadServer: FD 43: read failure: (104) Connection reset by peer
2009/01/22 10:56:44| NETDB state saved; 0 entries, 0 msec
2009/01/22 11:02:54| sslReadServer: FD 47: read failure: (104) Connection reset by peer
2009/01/22 11:41:01| sslReadServer: FD 32: read failure: (104) Connection reset by peer
2009/01/22 11:41:23| sslReadServer: FD 36: read failure: (104) Connection reset by peer
2009/01/22 12:04:11| NETDB state saved; 0 entries, 0 msec
2009/01/22 12:29:03| urlParse: Illegal character in hostname '%25257bd30c1661-cdaf-11d0-8a3e-00c04fc9e26e%25257d'
2009/01/22 12:44:12| NETDB state saved; 0 entries, 32 msec
2009/01/22 12:48:04| sslReadServer: FD 29: read failure: (104) Connection reset by peer
2009/01/22 12:51:44| sslReadServer: FD 26: read failure: (104) Connection reset by peer
2009/01/22 12:53:07| sslReadServer: FD 26: read failure: (104) Connection reset by peer
2009/01/22 13:06:01| sslReadServer: FD 39: read failure: (104) Connection reset by peer
2009/01/22 13:33:24| sslReadServer: FD 28: read failure: (104) Connection reset by peer
2009/01/22 13:41:02| sslReadServer: FD 29: read failure: (104) Connection reset by peer
2009/01/22 13:41:10| sslReadServer: FD 33: read failure: (104) Connection reset by peer
2009/01/22 13:47:40| NETDB state saved; 0 entries, 40 msec
2009/01/22 13:52:42| sslReadServer: FD 44: read failure: (104) Connection reset by peer
2009/01/22 14:28:46| NETDB state saved; 0 entries, 19 msec

Diede
(usa Debian)

Enviado em 24/01/2009 - 10:44h

Vendo este seu log, recomendo resetar o cache. Pelo que sei, no Squid 2.5, o "WARNING: Disk space over limit", está relacionado à arquivos grandes no cache e um cache pequeno para estes.
Pelo que entendi, seu cache é de apenas 100MB, e isto é de fato muito pouco para uma empresa. Pior se aí o pessoal fizer downloads de programas e músicas.
Tente extender este cache para pelo menos 2 GB de início.

E... por estas duas linhas:

2009/01/21 15:40:16| Adding nameserver 200.175.5.139 from /etc/resolv.conf
2009/01/21 15:40:16| Adding nameserver 200.175.89.139 from /etc/resolv.conf

Se o squid não estiver satisfatoriamente rápido, tente adicionar um servidor DNS para aumentar a velocidade da resolução de nomes.

matheusoveral
(usa Ubuntu)

Enviado em 24/01/2009 - 10:48h

Se for no cache, é no paramentro cache_dir:

cache_dir ufs /var/spool/squid 100 16 256

UFS: É o sistema de arquivos que será usado.
/var/spool/squid: Diretório usado pela cache.
100: Referente ao tamanho em MB a ser usado pela cache.
16: Referente aos diretórios que serão criados no primeiro nível.
256: Referente aos diretórios que serão criados no segundo nível.

Cara tenho um script de limpeza de cache se te ajudar, posso te enviar...

lordsnc
(usa Mandriva)

Enviado em 24/01/2009 - 10:54h

eu tambem estou tendo problemas com o squid...

eu instalei e configurei so q quando tento coloccar ele para rodar aparece a seguinte mensssagem de erro..

* Restarting Squid HTTP proxy squid 2009/01/24 01:24:46| aclParseIpData: WARNING: Netmask masks away part of the specified IP in '192.168.0.1/24'

PS:so não dá erro se eu trocar o numero de ip 198.168.0.1 para 198.168.1.0

só q o Ip eth0(rede local) é 198.168.0.1 e não 198.168.1.0 ..

oq faço?

matheusoveral
(usa Ubuntu)

Enviado em 24/01/2009 - 11:02h

lordsnc;

Como estam suas acls; posta ai para poder-mos analizar:

Se vc que o range? Pode deixar assim:

acl redeinterna src 192.168.0.0/24

lordsnc
(usa Mandriva)

Enviado em 24/01/2009 - 11:09h

abaixo segue o meu squid.conf


*****************************************
http_port 3129
visible_hostname servidorx-marimba

cache_mem 64 MB
maximum_object_size_in_memory 64 KB
maximum_object_size 750 MB
minimum_object_size 0 KB
cache_swap_low 90
cache_swap_high 95
cache_dir ufs /var/spool/squid 2048 16 256
cache_access_log /var/log/squid/access.log
refresh_pattern ^ftp: 15 20% 2280
refresh_pattern ^gopher: 15 0% 2280
refresh_pattern . 15 20% 2280

acl all src 0.0.0.0/0.0.0.0
acl manager proto cache_object
acl localhost src 127.0.0.1/255.255.255.255
acl SSL_ports port 443 563
acl Safe_ports port 21 80 443 563 70 210 280 488 59 777 901 1025-65535
acl purge method PURGE
acl CONNECT method CONNECT

http_access allow manager localhost
http_access deny manager
http_access allow purge localhost
http_access deny purge
http_access deny !Safe_ports
http_access deny CONNECT !SSL_ports


acl redeloca src 192.168.0.1/24
http_access allow localhost
http_access allow redelocal
http_access deny all

************************************
ai está

matheusoveral
(usa Ubuntu)

Enviado em 24/01/2009 - 11:33h

Arquivo modificado

##################################

http_port 3129
visible_hostname servidorx-marimba

cache_mem 64 MB
maximum_object_size_in_memory 64 KB
maximum_object_size 750 MB
minimum_object_size 0 KB
cache_swap_low 90
cache_swap_high 95
cache_dir ufs /var/spool/squid 2048 16 256
cache_access_log /var/log/squid/access.log
refresh_pattern ^ftp: 15 20% 2280
refresh_pattern ^gopher: 15 0% 2280
refresh_pattern . 15 20% 2280

acl all src 0.0.0.0/0.0.0.0
acl manager proto cache_object
acl localhost src 127.0.0.1/255.255.255.255
acl SSL_ports port 443 563
acl Safe_ports port 21 80 443 563 70 210 280 488 59 777 901 1025-65535
acl purge method PURGE
acl CONNECT method CONNECT

http_access allow manager localhost
http_access deny manager
http_access allow purge localhost
http_access deny purge
http_access deny !Safe_ports
http_access deny CONNECT !SSL_ports

acl redelocal src 192.168.0.0/24
http_access allow localhost
http_access allow redelocal
http_access deny all

lordsnc
(usa Mandriva)

Enviado em 24/01/2009 - 11:43h

pronto fiz as modificações q vc me indincou...

mas ainda assim não abro pagina de internet nas outras maquinas..

ja coloquei o o ip da eth0 servidor como proxy e nada....

erminhoferreira
(usa Debian)

Enviado em 30/01/2009 - 23:10h

Cara..

tive problema similar... do nada o squid parava.

No meu caso eram os arquivos de log que estavam gigantes. De uma olhada na pasta /var/log/squid se existem arquivos muito grandes. (access.log, store.log, etc..) Dê um ls -la -h e verifique o tamanho dos mesmos.
Se vc utiliza o Sarg para monitorar os acessos, gere os relatorios do SARG e depois pode apagar os arquivos de log sem dó... rsrs
Se for este o caso fique atento que caso os arquivos comecem a ficar grandes novamente, possivelmente vc está com spyware em alguma maquina da sua rede.

FLW!