|
Do you mean 50,000 TPS? That is the kind of maximum performance I would
expect on a gigabit network.
We can do 110,000+ on a 10G network (both tests with a single Xeon Quad
core chip + intel NICs)
On 14 August 2014 08:23, yige2008123 <yige2008123@xxxxxxx> wrote:
>
>
> hi everyone:
> I do test throughput of lvs server, the mode is fullnat. Num of
> director is only one.The testing tool is webbentch, when throughput more
> than 50,000， It's not increase, I do not know what is the reason, I
> optimization params of kernal, include limits of open file, tcp connection.
> etc, thank you
>
> my ipvs version:
> [root@xxxxxxxxxxxxxxx ~]# ipvsadm
> IP Virtual Server version 1.2.1 (size=4194304)
>
> kernel:
> Linux lvs-test 2.6.32 #1 SMP Wed May 21 14:31:51 CST 2014 x86_64 x86_64
> x86_64 GNU/Linux
>
> optimization of kernel:
> net.nf_conntrack_max = 25000000
> net.netfilter.nf_conntrack_max = 25000000
> net.netfilter.nf_conntrack_tcp_timeout_established = 1500
> net.ipv4.tcp_max_tw_buckets = 6000
> net.ipv4.tcp_sack = 1
> net.ipv4.tcp_window_scaling = 1
> net.ipv4.tcp_rmem = 4096 87380 4194304
> net.ipv4.tcp_wmem = 4096 16384 4194304
> net.core.wmem_default = 8388608
> net.core.rmem_default = 8388608
> net.core.rmem_max = 16777216
> net.core.wmem_max = 16777216
> net.core.netdev_max_backlog = 262144
> net.core.somaxconn = 262144
> net.ipv4.tcp_max_orphans = 3276800
> net.ipv4.tcp_max_syn_backlog = 262144
> net.ipv4.tcp_timestamps = 0
> net.ipv4.tcp_synack_retries = 1
> net.ipv4.tcp_syn_retries = 1
> net.ipv4.tcp_tw_recycle = 1
> net.ipv4.tcp_tw_reuse = 1
> net.ipv4.tcp_mem = 94500000 915000000 927000000
> net.ipv4.tcp_fin_timeout = 100
> net.ipv4.tcp_keepalive_time = 30
> net.ipv4.ip_local_port_range = 1024 65000
> net.ipv4.route.flush = 1
> net.ipv4.conf.default.rp_filter = 1
> net.ipv4.conf.all.rp_filter = 1
> net.ipv4.tcp_syncookies = 0
> net.ipv6.conf.all.disable_ipv6 = 1
> net.ipv4.tcp_max_syn_backlog = 819200
> net.ipv4.tcp_synack_retries = 1
> net.ipv4.tcp_max_tw_buckets = 819200
> net.ipv4.tcp_tw_reuse = 1
> net.ipv4.tcp_tw_recycle = 1
> net.ipv4.conf.all.arp_ignore = 1
> net.ipv4.conf.all.arp_announce = 2
> net.core.netdev_max_backlog = 500000
> net.ipv4.ip_forward=1
> net.ipv4.ip_conntrack_max=1024000
>
>
> [root@xxxxxxxxxxxxxxx ~]# ulimit -n
> 655350
>
> log:
> Aug 14 15:06:07 10 Keepalived_healthcheckers: TCP connection to
> [10.153.75.152]:80 success.
> Aug 14 15:06:07 10 Keepalived_healthcheckers: Enabling service
> [10.153.75.152]:80 to VS [test]:0
> Aug 14 15:06:15 10 Keepalived_healthcheckers: TCP connection to
> [10.153.74.85]:80 failed !!!
> Aug 14 15:06:15 10 Keepalived_healthcheckers: Disabling service
> [10.153.74.85]:80 from VS [test]:0
> Aug 14 15:06:18 10 Keepalived_healthcheckers: TCP connection to
> [10.153.74.139]:80 failed !!!
> Aug 14 15:06:18 10 Keepalived_healthcheckers: Disabling service
> [10.153.74.139]:80 from VS [test]:0
> Aug 14 15:06:22 10 Keepalived_healthcheckers: TCP connection to
> [10.153.74.85]:80 success.
> Aug 14 15:06:22 10 Keepalived_healthcheckers: Enabling service
> [10.153.74.85]:80 to VS [test]:0
> Aug 14 15:06:25 10 Keepalived_healthcheckers: TCP connection to
> [10.153.74.139]:80 success.
> Aug 14 15:06:25 10 Keepalived_healthcheckers: Enabling service
> [10.153.74.139]:80 to VS [test]:0
> Aug 14 15:06:43 10 Keepalived_healthcheckers: TCP connection to
> [10.153.75.153]:80 failed !!!
> Aug 14 15:06:43 10 Keepalived_healthcheckers: Disabling service
> [10.153.75.153]:80 from VS [test]:0
> Aug 14 15:06:50 10 Keepalived_healthcheckers: TCP connection to
> [10.153.75.153]:80 success.
> Aug 14 15:06:50 10 Keepalived_healthcheckers: Enabling service
> [10.153.75.153]:80 to VS [test]:0
> Aug 14 15:06:52 10 Keepalived_healthcheckers: TCP connection to
> [10.153.74.140]:80 failed !!!
> Aug 14 15:06:52 10 Keepalived_healthcheckers: Disabling service
> [10.153.74.140]:80 from VS [test]:0
> Aug 14 15:06:59 10 Keepalived_healthcheckers: TCP connection to
> [10.153.74.140]:80 success.
> Aug 14 15:06:59 10 Keepalived_healthcheckers: Enabling service
> [10.153.74.140]:80 to VS [test]:0
> Aug 14 15:07:12 10 Keepalived_healthcheckers: TCP connection to
> [10.153.75.56]:80 failed !!!
> Aug 14 15:07:12 10 Keepalived_healthcheckers: Disabling service
> [10.153.75.56]:80 from VS [test]:0
> Aug 14 15:07:19 10 Keepalived_healthcheckers: TCP connection to
> [10.153.75.56]:80 success.
> Aug 14 15:07:19 10 Keepalived_healthcheckers: Enabling service
> [10.153.75.56]:80 to VS [test]:0
> Aug 14 15:07:24 10 Keepalived_healthcheckers: TCP connection to
> [10.153.74.85]:80 failed !!!
> Aug 14 15:07:24 10 Keepalived_healthcheckers: Disabling service
> [10.153.74.85]:80 from VS [test]:0
> Aug 14 15:07:28 10 Keepalived_healthcheckers: TCP connection to
> [10.153.75.153]:80 failed !!!
> Aug 14 15:07:28 10 Keepalived_healthcheckers: Disabling service
> [10.153.75.153]:80 from VS [test]:0
> Aug 14 15:07:31 10 Keepalived_healthcheckers: TCP connection to
> [10.153.74.85]:80 success.
> Aug 14 15:07:31 10 Keepalived_healthcheckers: Enabling service
> [10.153.74.85]:80 to VS [test]:0
> Aug 14 15:07:35 10 Keepalived_healthcheckers: TCP connection to
> [10.153.75.153]:80 success.
> Aug 14 15:07:35 10 Keepalived_healthcheckers: Enabling service
> [10.153.75.153]:80 to VS [test]:0
>
>
> iptables: iptables is stopped
>
> route table:
> [root@xxxxxxxxxxx ~]# route -n
> Kernel IP routing table
> Destination Gateway Genmask Flags Metric Ref Use
> Iface
> 172.27.0.0 172.27.0.9 255.255.255.252 UG 101 0 0
> eth1
> 172.27.0.4 172.27.0.9 255.255.255.252 UG 101 0 0
> eth1
> 172.27.0.8 0.0.0.0 255.255.255.252 U 0 0 0
> eth1
> 172.27.0.16 172.27.0.9 255.255.255.252 UG 101 0 0
> eth1
> 172.27.0.20 172.27.0.9 255.255.255.252 UG 101 0 0
> eth1
> 10.153.72.0 0.0.0.0 255.255.248.0 U 0 0 0
> eth0
> 169.254.0.0 0.0.0.0 255.255.0.0 U 1002 0 0
> eth0
> 169.254.0.0 0.0.0.0 255.255.0.0 U 1003 0 0
> eth1
> 10.0.0.0 10.153.79.254 255.0.0.0 UG 0 0 0
> eth0
> 0.0.0.0 172.27.0.9 0.0.0.0 UG 0 0 0
> eth1
>
>
> keepalived conf：
> the num of realserver is 14:
>
> gobal_defs {
> notification_email {
> acassen@xxxxxxxxxxxx
> failover@xxxxxxxxxxxx
> sysadmin@xxxxxxxxxxxx
> }
> notification_email_from Alexandre.Cassen@xxxxxxxxxxxx
> smtp_connect_timeout 30
> router_id LVS_DEVEL
> }
>
> local_address_group laddr_g1 {
> 10.153.72.2
> }
>
> vrrp_instance VI_1 {
> state MASTER
> interface eth1
> virtual_router_id 200
> priority 150
> advert_int 1
> authentication {
> auth_type PASS
> auth_pass 123456
> }
> virtual_ipaddress {
> xxx.xxx.xxx.xxx/32
> }
> }
> virtual_server_group test {
> xxx.xxx.xxx.xxx 80 //vip1
> }
>
>
> virtual_server group test {
> delay_loop 7
> lb_algo rr
> lb_kind FNAT
> laddr_group_name laddr_g1
> protocol TCP
> # syn_proxy
> # persistence_timeout 50
> omega
> quorum 1
> alpha
> hysteresis 0
>
> real_server 10.153.75.118 80 {
> weight 1
> inhibit_on_failure
> TCP_CHECK {
> connect_timeout 3
> nb_get_retry 2
> delay_before_retry 5
> }
> }
>
> real_server 10.153.74.139 80 {
> weight 1
> inhibit_on_failure
> TCP_CHECK {
> connect_timeout 3
> nb_get_retry 2
> delay_before_retry 5
> }
> }
>
> real_server 10.153.74.140 80 {
> weight 1
> inhibit_on_failure
> TCP_CHECK {
> connect_timeout 3
> #nb_get_retry 2
> #delay_before_retry 5
> }
> }
> .
> .
> .
>
> }
>
>
>
>
> _______________________________________________
> Please read the documentation before posting - it's available at:
> http://www.linuxvirtualserver.org/
>
> LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
> Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
> or go to http://lists.graemef.net/mailman/listinfo/lvs-users
--
Regards,
Malcolm Turnbull.
Loadbalancer.org Ltd.
Phone: +44 (0)330 1604540
http://www.loadbalancer.org/
_______________________________________________
Please read the documentation before posting - it's available at:
http://www.linuxvirtualserver.org/
LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
or go to http://lists.graemef.net/mailman/listinfo/lvs-users
|
