[FUG-BR] Valor aconselhavel para variável HZ

Quinta Novembro 6 11:55:38 BRST 2014

Enviado do meu smartphone Sony Xperia™

---- Fabricio Lima escreveu ----

> Edinilson, vc está certo...
> eu tb mexia mais nisso na epoca do 6, 7, 8..
> 
> Hj eles ajustaram tanta coisa (9 e 10)
> q talvez o default seja melhor q um 7 tunado...
> 
> (minha documentaçao caducou)
> 
> testar, testar, testar...
> 
> [ ]'s
> Fabricio Lima
> When your hammer is C++, everything begins to look like a thumb.
> 
> 2014-11-05 16:44 GMT-02:00 Edinilson - ATINET <edinilson em atinet.com.br>:
> 
> > ----- Original Message ----- From: "Paulo Henrique - BSDs" <
> >> paulo.rddck em bsd.com.br>
> >> To: ""Lista Brasileira de Discussão sobre FreeBSD (FUG-BR)"" <
> >> freebsd em fug.com.br>
> >> Sent: Wednesday, November 05, 2014 3:58 PM
> >> Subject: [FUG-BR] Valor aconselhavel para variável HZ
> >>
> >> Saudações,
> >>
> >> Gostaria de saber se alguém trabalha com a variável HZ com o valor
> >> superior a 2000, caso sim o ambiente fica estável, há uma melhora no
> >> desempenho do sistema e da rede ?
> >> Sei que a mesma interfere quanto ao uso da bateria em portáteis contudo a
> >> duvida é restrita a servidores.
> >> Aumentar o valor da mesma em um servidor com 12 Cores / 24 Threads com
> >> 32Gbytes de ram melhorará o desempenho.
> >> Abaixo tem os dados do sistema atualmente.
> >> O HZ do sistema está em 2000.
> >>
> >> uname -a
> >> FreeBSD xxxxx 10.0-STABLE FreeBSD 10.0-STABLE #0 r269344: Thu Jul 31
> >> 14:39:46 BRT 2014
> >> netstat -m
> >> yyyyyyy em xxxxx:/usr/obj/usr/src/sys/XXXXXXXXXXX amd64
> >> 11204/15571/26775 mbufs in use (current/cache/total)
> >> 1023/9825/10848/2036062 mbuf clusters in use (current/cache/total/max)
> >> 1023/9512 mbuf+clusters out of packet secondary zone in use
> >> (current/cache)
> >> 0/548/548/1018031 4k (page size) jumbo clusters in use
> >> (current/cache/total/max)
> >> 0/0/0/301638 9k jumbo clusters in use (current/cache/total/max)
> >> 0/0/0/169671 16k jumbo clusters in use (current/cache/total/max)
> >> 4847K/25734K/30581K bytes allocated to network (current/cache/total)
> >> 0/0/0 requests for mbufs denied (mbufs/clusters/mbuf+clusters)
> >> 0/0/0 requests for mbufs delayed (mbufs/clusters/mbuf+clusters)
> >> 0/0/0 requests for jumbo clusters delayed (4k/9k/16k)
> >> 0/0/0 requests for jumbo clusters denied (4k/9k/16k)
> >> 0 requests for sfbufs denied
> >> 0 requests for sfbufs delayed
> >> 9966280 requests for I/O initiated by sendfile
> >>
> >> uptime
> >> 15:47 up 86 days, 19:40, 1 user, load averages: 8,15 8,86 9,27
> >>
> >>
> >> Alguns tunnings que já foi feito no sistema.
> >>
> >> # $FreeBSD: stable/10/etc/sysctl.conf 112200 2003-03-13 18:43:50Z mux $
> >> #
> >> # This file is read when going to multi-user and its contents piped thru
> >> # ``sysctl'' to adjust kernel values. ``man 5 sysctl.conf'' for details.
> >> #
> >>
> >> # Uncomment this to prevent users from seeing information about processes
> >> that
> >> # are being run under another UID.
> >> #security.bsd.see_other_uids=0
> >>
> >> kern.maxfiles=1000000
> >>
> >> # Otimizacoes de rede.
> >> #kern.ipc.nmbclusters=131072 # 128Mb para buffer de rede - Ficou instavel
> >> valor definido pelo sistema.
> >> kern.ipc.maxsockbuf=33554432
> >> net.inet.tcp.sendbuf_max=33554432
> >> net.inet.tcp.recvbuf_max=33554432
> >> net.inet.tcp.sendspace=1048576 # default 65536
> >> net.inet.tcp.recvspace=1048576 # default 32768
> >> net.inet.tcp.sendbuf_inc=1048576 # 8192 default
> >> net.inet.tcp.recvbuf_inc=1048576 # 16384 default
> >> kern.ipc.somaxconn=4096 # 128 default
> >> net.inet.tcp.syncache.rexmtlimit=1
> >> net.inet.tcp.syncookies=1
> >>
> >> # COnfigura▒▒es de Seguran▒a
> >> # General Security and DoS mitigation.
> >> net.inet.ip.check_interface=1 # verify packet arrives on correct interface
> >> net.inet.ip.portrange.randomized=1 # randomize outgoing upper ports
> >> net.inet.ip.process_options=0 # IP options in the incoming packets will
> >> be ignored
> >> net.inet.ip.random_id=1 # assign a random IP_ID to each packet leaving
> >> the system
> >> net.inet.ip.redirect=0 # do not send IP redirects
> >> net.inet.ip.accept_sourceroute=0 # drop source routed packets since they
> >> can not be trusted
> >> net.inet.ip.sourceroute=0 # if source routed packets are accepted the
> >> route data is ignored
> >> #net.inet.ip.stealth=1 # do not reduce the TTL by one(1) when a packets
> >> goes through the firewall
> >> net.inet.icmp.bmcastecho=0 # do not respond to ICMP packets sent to IP
> >> broadcast addresses
> >> net.inet.icmp.maskfake=0 # do not fake reply to ICMP Address Mask Request
> >> packets
> >> net.inet.icmp.maskrepl=0 # replies are not sent for ICMP address mask
> >> requests
> >> net.inet.icmp.log_redirect=0 # do not log redirected ICMP packet attempts
> >> net.inet.icmp.drop_redirect=1 # no redirected ICMP packets
> >> #net.inet.icmp.icmplim=50 # 50 ICMP packets per second. a reasonable
> >> number for a small office.
> >> #net.inet.tcp.delayed_ack=1 # always employ delayed ack, 6 packets get 1
> >> ack to increase bandwidth
> >> net.inet.tcp.drop_synfin=1 # SYN/FIN packets get dropped on initial
> >> connection
> >> net.inet.tcp.ecn.enable=1 # explicit congestion notification (ecn)
> >> warning: some ISP routers abuse it
> >> net.inet.tcp.fast_finwait2_recycle=1 # recycle FIN/WAIT states quickly
> >> (helps against DoS, but may cause false RST)
> >> net.inet.tcp.icmp_may_rst=0 # icmp may not send RST to avoid spoofed
> >> icmp/udp floods
> >> net.inet.tcp.maxtcptw=15000 # max number of tcp time_wait states for
> >> closing connections
> >> net.inet.tcp.msl=5000 # 5 second maximum segment life waiting for an ACK
> >> in reply to a SYN-ACK or FIN-ACK
> >> net.inet.tcp.path_mtu_discovery=0 # disable MTU discovery since most
> >> ICMP packets are dropped by others
> >> net.inet.tcp.rfc3042=0 # disable the limited transmit mechanism which can
> >> slow burst transmissions
> >> #net.inet.tcp.sack.enable=1 # sack disabled? http://www.ibm.com/
> >> developerworks/linux/library/l-tcp-sack/index.html
> >> net.inet.udp.blackhole=1 # drop udp packets destined for closed sockets
> >> net.inet.tcp.blackhole=2 # drop tcp packets destined for closed ports
> >> #net.route.netisr_maxqlen=4096 # route queue length defaults 4096 (rtsock
> >> using "netstat -Q")
> >> security.bsd.see_other_uids=0 # hide processes for root from user uid's
> >>
> >> ## Configuracoes de envio
> >>
> >> net.local.stream.sendspace=164240
> >> net.local.stream.recvspace=164240
> >>
> >> ## Configuracoes contra spoof
> >>
> >> net.inet.ip.rtexpire=60 # 3600 secs
> >> net.inet.ip.rtminexpire=2 # 10 secs
> >> net.inet.ip.rtmaxcache=1024 # 128 entries
> >>
> >> Qualquer orientação é bem vinda.
> >>
> >>
> > Ate o FreeBSD 7 eu ate que me preocupava em ficar mexendo no hz. Apos
> > isto, e com estas interfaces de rede modernas da Intel, nao me preocupo
> > mais (nem com hz nem com polling) pois PARECE nao mudar muito coisa.
> >
> > Veja uma discussao de 2005 sobre o hz onde ja se discutia isto:
> > http://lists.freebsd.org/pipermail/freebsd-questions/
> > 2005-April/083482.html
> >
> > e veja tambem este link, na parte onde esta o hz o que ele diz:
> > https://calomel.org/freebsd_network_tuning.html
> >
> >
> > Edinilson
> > ------------------------------------------
> > ATINET
> > Tel Voz: (0xx11) 4412-0876
> > http://www.atinet.com.br
> >
> >
> >
> > -------------------------
> > Histórico: http://www.fug.com.br/historico/html/freebsd/
> > Sair da lista: https://www.fug.com.br/mailman/listinfo/freebsd
> >
> -------------------------
> Histórico: http://www.fug.com.br/historico/html/freebsd/
> Sair da lista: https://www.fug.com.br/mailman/listinfo/freebsd

Pessoal,  pelo que compreendi, quanto mais alto o valor do HZ, mais rápido o sistema vai responder requisições de tarefas do kernel e do hardware.
Embora como negativamente menor será o tempo disponível para limpar a sujeira de outro processo.
Diminui a latência da rede e o tempo de resposta a uma requisição. 

Nesse sábado vou fazer uns testes aumentando o hz o máximo possível e ver como será o comportamento do sistema, depois posto os resultados.

Contudo estou na dúvida de como fazer para verificar isso, o unixbench seria uma alternativa?

Att.