19 October 2019, 19:35:24, by "Michael Tuexen" <michael.tuexen at
lurchi.franken.de>:
> > On 19. Oct 2019, at 18:09, Paul <devgs at ukr.net> wrote:
> >
> > Hi Michael,
> >
> > Thank you, for taking your time!
> >
> > We use physical machines. We don not have any special `pf` rules.
> > Both sides ran `pfctl -d` before testing.
> Hi Paul,
>
> OK. How are the physical machines connected to each other?
We have tested different connections. The old, copper ethernet, cable,
as well as optics connection with an identical outcome. Machines are
connected through Juniper QFX5100.
>
> What happens when you don't use a lagg interface, but the physical
ones?
>
> (Trying to localise the problem...)
Same thing, lagg does not change anything. Originally, the problem was
observed on a regular interface.
We have tested a on different hardware. Results are consistently
stable on 11.2-STABLE and consistently unstable on 12.1-STABLE.
The only unchanged thing is the network card vendor, it's Intel.
>
> Best regards
> Michael
> >
> >
> > `nginx` config is primitive, no secrets there:
> >
> > -------------------------------------------------------------------
> > user www;
> > worker_processes auto;
> >
> > error_log /var/log/nginx/error.log warn;
> >
> > events {
> > worker_connections 81920;
> > kqueue_changes 4096;
> > use kqueue;
> > }
> >
> > http {
> > include mime.types;
> > default_type application/octet-stream;
> >
> > sendfile off;
> > keepalive_timeout 65;
> > tcp_nopush on;
> > tcp_nodelay on;
> >
> > # Logging
> > log_format main '$remote_addr - $remote_user
[$time_local] "$request" '
> > '$status $request_length
$body_bytes_sent "$http_referer" '
> > '"$http_user_agent"
"$http_x_real_ip" "$realip_remote_addr"
"$request_completion" "$request_time" '
> > '"$request_body"';
> >
> > access_log /var/log/nginx/access.log main;
> >
> > server {
> > listen 80 default;
> >
> > server_name localhost _;
> >
> > location / {
> > return 404;
> > }
> > }
> > }
> > -------------------------------------------------------------------
> >
> >
> > `wrk` is compiled with a default configuration. We test like this:
> >
> > `wrk -c 10 --header "Connection: close" -d 10 -t 1 --latency
http://10.10.10.92:80/missing`
> >
> >
> > Also, it seems that our issue, and the one described in this thread,
are identical:
> >
> >
https://lists.freebsd.org/pipermail/freebsd-net/2019-June/053667.html
> >
> > We both have the Intel network cards, BTW. Our network cards are
these:
> >
> > em0 at pci0:10:0:0: class=0x020000 card=0x000015d9
chip=0x10d38086 rev=0x00 hdr=0x00
> > vendor = 'Intel Corporation'
> > device = '82574L Gigabit Network Connection'
> >
> > ixl0 at pci0:4:0:0: class=0x020000 card=0x00078086
chip=0x15728086 rev=0x01 hdr=0x00
> > vendor = 'Intel Corporation'
> > device = 'Ethernet Controller X710 for 10GbE SFP+'
> >
> >
> > =============================> >
> > Additional info:
> >
> > During the tests, we have bonded two interfaces into a lagg:
> >
> > ixl0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric
0 mtu 1500
> >
options=c500b8<VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,VLAN_HWFILTER,VLAN_HWTSO,TXCSUM_IPV6>
> > ether 3c:fd:fe:aa:60:20
> > media: Ethernet autoselect (10Gbase-SR <full-duplex>)
> > status: active
> > nd6 options=29<PERFORMNUD,IFDISABLED,AUTO_LINKLOCAL>
> > ixl1: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric
0 mtu 1500
> >
options=c500b8<VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,VLAN_HWFILTER,VLAN_HWTSO,TXCSUM_IPV6>
> > ether 3c:fd:fe:aa:60:20
> > hwaddr 3c:fd:fe:aa:60:21
> > media: Ethernet autoselect (10Gbase-SR <full-duplex>)
> > status: active
> > nd6 options=29<PERFORMNUD,IFDISABLED,AUTO_LINKLOCAL>
> >
> >
> > lagg0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric
0 mtu 1500
> >
options=c500b8<VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,VLAN_HWFILTER,VLAN_HWTSO,TXCSUM_IPV6>
> > ether 3c:fd:fe:aa:60:20
> > inet 10.10.10.92 netmask 0xffff0000 broadcast 10.10.255.255
> > laggproto failover lagghash l2,l3,l4
> > laggport: ixl0 flags=5<MASTER,ACTIVE>
> > laggport: ixl1 flags=0<>
> > groups: lagg
> > media: Ethernet autoselect
> > status: active
> > nd6 options=29<PERFORMNUD,IFDISABLED,AUTO_LINKLOCAL>
> >
> > using this config:
> >
> > ifconfig_ixl0="up -lro -tso -rxcsum -txcsum" (tried
different options - got the same outcome)
> > ifconfig_ixl1="up -lro -tso -rxcsum -txcsum"
> > ifconfig_lagg0="laggproto failover laggport ixl0 laggport
ixl1 10.10.10.92/24"
> >
> >
> > We have randomly picked `ixl0` and restricted number of RX/TX queues
to 1:
> > /boot/loader.conf :
> > dev.ixl.0.iflib.override_ntxqs=1
> > dev.ixl.0.iflib.override_nrxqs=1
> >
> > leaving `ixl1` with a default number, matching number of cores (6).
> >
> >
> > ixl0: <Intel(R) Ethernet Controller X710 for 10GbE SFP+ -
2.1.0-k> mem 0xf8800000-0xf8ffffff,0xf9808000-0xf980ffff irq 40 at device 0.0
on pci4
> > ixl0: fw 5.0.40043 api 1.5 nvm 5.05 etid 80002927 oem 1.261.0
> > ixl0: PF-ID[0]: VFs 64, MSI-X 129, VF MSI-X 5, QPs 768, I2C
> > ixl0: Using 1024 TX descriptors and 1024 RX descriptors
> > ixl0: Using 1 RX queues 1 TX queues
> > ixl0: Using MSI-X interrupts with 2 vectors
> > ixl0: Ethernet address: 3c:fd:fe:aa:60:20
> > ixl0: Allocating 1 queues for PF LAN VSI; 1 queues active
> > ixl0: PCI Express Bus: Speed 8.0GT/s Width x4
> > ixl0: SR-IOV ready
> > ixl0: netmap queues/slots: TX 1/1024, RX 1/1024
> > ixl1: <Intel(R) Ethernet Controller X710 for 10GbE SFP+ -
2.1.0-k> mem 0xf8000000-0xf87fffff,0xf9800000-0xf9807fff irq 40 at device 0.1
on pci4
> > ixl1: fw 5.0.40043 api 1.5 nvm 5.05 etid 80002927 oem 1.261.0
> > ixl1: PF-ID[1]: VFs 64, MSI-X 129, VF MSI-X 5, QPs 768, I2C
> > ixl1: Using 1024 TX descriptors and 1024 RX descriptors
> > ixl1: Using 6 RX queues 6 TX queues
> > ixl1: Using MSI-X interrupts with 7 vectors
> > ixl1: Ethernet address: 3c:fd:fe:aa:60:21
> > ixl1: Allocating 8 queues for PF LAN VSI; 6 queues active
> > ixl1: PCI Express Bus: Speed 8.0GT/s Width x4
> > ixl1: SR-IOV ready
> > ixl1: netmap queues/slots: TX 6/1024, RX 6/1024
> >
> >
> > This allowed us easy switch between different configurations without
> > the need to reboot, by simply shutting down one interface or the
other:
> >
> > `ifconfig XXX down`
> >
> > When testing `ixl0` that runs only a single queue:
> > ixl0: Using 1 RX queues 1 TX queues
> > ixl0: netmap queues/slots: TX 1/1024, RX 1/1024
> >
> > we've got these results:
> >
> > `wrk -c 10 --header "Connection: close" -d 10 -t 1 --latency
http://10.10.10.92:80/missing`
> > Running 10s test @ http://10.10.10.92:80/missing
> > 1 threads and 10 connections
> > Thread Stats Avg Stdev Max +/- Stdev
> > Latency 281.31us 297.74us 22.66ms 99.70%
> > Req/Sec 19.91k 2.79k 21.25k 97.59%
> > Latency Distribution
> > 50% 266.00us
> > 75% 309.00us
> > 90% 374.00us
> > 99% 490.00us
> > 164440 requests in 10.02s, 47.52MB read
> > Socket errors: read 0, write 0, timeout 0
> > Non-2xx or 3xx responses: 164440
> > Requests/sec: 16412.09
> > Transfer/sec: 4.74MB
> >
> >
> > When testing `ixl1` that runs 6 queues:
> > ixl1: Using 6 RX queues 6 TX queues
> > ixl1: netmap queues/slots: TX 6/1024, RX 6/1024
> >
> > we've got these results:
> >
> > `wrk -c 10 --header "Connection: close" -d 10 -t 1 --latency
http://10.10.10.92:80/missing`
> > Running 10s test @ http://10.10.10.92:80/missing
> > 1 threads and 10 connections
> > Thread Stats Avg Stdev Max +/- Stdev
> > Latency 216.16us 71.97us 511.00us 47.56%
> > Req/Sec 4.34k 2.76k 15.44k 83.17%
> > Latency Distribution
> > 50% 216.00us
> > 75% 276.00us
> > 90% 312.00us
> > 99% 365.00us
> > 43616 requests in 10.10s, 12.60MB read
> > Socket errors: connect 0, read 24, write 8, timeout 0
> > Non-2xx or 3xx responses: 43616
> > Requests/sec: 4318.26
> > Transfer/sec: 1.25MB
> >
> > Do note, that, not only multiple queues cause issues they also
dramatically
> > decrease the performance of the network.
> >
> > Using `sysctl -w net.inet.tcp.ts_offset_per_conn=0` didn't help at
all.
> >
> > Best regards,
> > -Paul
> >
> >
>
>