[Cake] Need help troubleshooting cake
Gustavo Lapido Loureiro
gustavo.lapido at gmail.com
Sat Mar 6 12:56:13 EST 2021
Ok, I decided to step backwards, and start from the beginning.
My first goal will be to make sure the hardware is ok, mainly, the network
interfaces.
I reinstalled Debian 10, turned it into a barebones router, and ran the
*speedtest-netperf.sh* script from Openwrt
<https://github.com/openwrt/packages/blob/openwrt-19.07/net/speedtest-netperf/files/speedtest-netperf.sh>,
either from within the router, as well as from my 100/10 notebook.
The results seem to indicate everything is fine:
-
From the server, the test maxes out the Internet link;
-
Fromt the notebook, the test kinda maxes out its interface roof: close
to 90 Mbps in all four tests (I will assume this is because of poor
hardware quality)
I also switched WAN and LAN interfaces, just in case, and repeated the
test, which gave similar results as before.
PARTIAL CONCLUSION: So far so good.
Attached, both results.
*Atenciosamente,*
*Gustavo Lapido Loureiro*
Em qui., 4 de mar. de 2021 às 17:44, Gustavo Lapido Loureiro <
gustavo.lapido at gmail.com> escreveu:
> Hello, new member here.
> First, thank you all involved in making this great piece of software,
> which I have already experienced without issues during a couple of months.
> Back then, it was on Openwrt, in a TP-LINK Archer C-7 router, bandwidth
> capped to 60 Mbps (on the device's limits, but still making Internet usage
> much smoother to all users).
> Then I moved the router to a debian 10 Celeron J1900 mini-ITX, 4 GB of
> RAM, two entry level Gigabit cards.
> So far I couldn't get the stable A's in dslreports test I was getting with
> the previous setup.
> The point is that my Internet has been upgraded from 60 Mbps to 240/20
> Mbps. SInce most, if not all, devices are still 100 Mbps, I'm currently
> capping cake at 85/18.
> I basically "transported" the cake settings from Openwrt to debian,
> basically adjusting the speed limits.
> I would like to know where I start troubleshooting this, to help identify
> what is causing bufferbloat on this new setup.
>
> Follow cake settings, then tc, then network hardware info
>
> *CAKE*
>
> # Uplink and Downlink values are in kbps
> UPLINK=17500
> #DOWNLINK=216000
> DOWNLINK=90000
>
> # SQM recipe to use. For more information, see /usr/lib/sqm/*.help
> SCRIPT=piece_of_cake.qos
>
> # Optional/advanced config
>
> ENABLED=1
> QDISC=cake
>
> #LLAM=tc_stab
> LINKLAYER=ethernet
> OVERHEAD=22
> #STAB_MTU=2047
> #STAB_TSIZE=512
> #STAB_MPU=0
>
> #ILIMIT=
> #ELIMIT=
> #ITARGET=
> #ETARGET=
>
> # ECN ingress resp. egress. Values are ECN or NOECN.
> IECN=ECN
> EECN=NOECN
>
> # Extra qdisc options ingress resp. egress
> IQDISC_OPTS="nat dual-dsthost"
> EQDISC_OPTS="nat dual-srchost ack-filter"
>
> # CoDel target
> #TARGET=5ms
>
> #ZERO_DSCP_INGRESS=1
> #IGNORE_DSCP_INGRESS=1
>
> *TC SETTINGS:*
>
> qdisc noqueue 0: dev lo root refcnt 2
> qdisc pfifo_fast 0: dev enp2s0 root refcnt 2 bands 3 priomap 1 2 2 2 1 2
> 0 0 1 1 1 1 1 1 1 1
> qdisc cake 801b: dev enp3s0 root refcnt 2 bandwidth 17500Kbit besteffort
> dual-srchost nat nowash ack-filter split-gso rtt 100.0ms noatm overhead 22
> qdisc ingress ffff: dev enp3s0 parent ffff:fff1 ----------------
> qdisc cake 801c: dev ifb4enp3s0 root refcnt 2 bandwidth 90Mbit besteffort
> dual-dsthost nat wash no-ack-filter split-gso rtt 100.0ms noatm overhead 22
>
> *NETWORK HARDWARE:*
>
> SCSI
> *-network
> description: Ethernet interface
> product: RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller
> vendor: Realtek Semiconductor Co., Ltd.
> physical id: 0
> bus info: pci at 0000:02:00.0
> logical name: enp2s0
> version: 06
> serial: c4:e9:84:00:1d:7b
> size: 1Gbit/s
> capacity: 1Gbit/s
> width: 64 bits
> clock: 33MHz
> capabilities: pm msi pciexpress msix vpd bus_master cap_list
> ethernet physical tp mii 10bt 10bt-fd 100bt 100bt-fd 1000bt 1000bt-fd
> autonegotiation
> configuration: autonegotiation=on broadcast=yes driver=r8169
> duplex=full firmware=rtl_nic/rtl8168e-2.fw ip=192.168.1.1 latency=0
> link=yes multicast=yes port=MII speed=1Gbit/s
> resources: irq:17 ioport:e000(size=256) memory:d0704000-d0704fff
> memory:d0700000-d0703fff
> *-network
> description: Ethernet interface
> product: RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller
> vendor: Realtek Semiconductor Co., Ltd.
> physical id: 0
> bus info: pci at 0000:03:00.0
> logical name: enp3s0
> version: 0c
> serial: 44:8a:5b:94:96:37
> size: 1Gbit/s
> capacity: 1Gbit/s
> width: 64 bits
> clock: 33MHz
> capabilities: pm msi pciexpress msix vpd bus_master cap_list
> ethernet physical tp aui bnc mii fibre 10bt 10bt-fd 100bt 100bt-fd
> 1000bt-fd autonegotiation
> configuration: autonegotiation=on broadcast=yes driver=r8169
> duplex=full firmware=rtl8168g-2_0.0.1 02/06/13 ip=179.210.1.62 latency=0
> link=yes multicast=yes port=MII speed=1Gbit/s
> resources: irq:18 ioport:d000(size=256) memory:d0604000-d0604fff
> memory:d0600000-d0603fff
> *-network:0
> description: Ethernet interface
> physical id: 1
> logical name: ifb4enp3s0
> serial: 76:6b:66:3c:6d:e9
> capabilities: ethernet physical
> configuration: broadcast=yes
> *-network:1 DISABLED
> description: Ethernet interface
> physical id: 2
> logical name: ifb1
> serial: 02:9e:3d:12:6e:16
> capabilities: ethernet physical
> configuration: broadcast=yes
> *-network:2 DISABLED
> description: Ethernet interface
> physical id: 3
> logical name: ifb0
> serial: 32:fe:93:2e:3a:1d
> capabilities: ethernet physical
> configuration: broadcast=yes
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.bufferbloat.net/pipermail/cake/attachments/20210306/424cbccb/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: netperf before sqm from notebook
Type: application/octet-stream
Size: 5666 bytes
Desc: not available
URL: <https://lists.bufferbloat.net/pipermail/cake/attachments/20210306/424cbccb/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: netperf before sqm from router
Type: application/octet-stream
Size: 5676 bytes
Desc: not available
URL: <https://lists.bufferbloat.net/pipermail/cake/attachments/20210306/424cbccb/attachment-0003.obj>
More information about the Cake
mailing list