Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Support for DS916+ #12

Open
mkter opened this issue Apr 1, 2020 · 1 comment
Open

Support for DS916+ #12

mkter opened this issue Apr 1, 2020 · 1 comment
Labels
report Just information

Comments

@mkter
Copy link

mkter commented Apr 1, 2020 •
edited
Loading

Description of the problem

There are several issues:

  • The adapter crashes when MTU is changed. A manual "/etc/rc.network restart" is required to revive the adapter. This comes without dmesg output or changes in lsusb/ifconfig. Afterwards the MTU settings are updated as requested.
  • As reported in Issue Support for DS716+ #7 the network connection crashes after a short data transfer. It is stable when sending data, even for large amounts of data (>50GB). It is very unstable when receiving data (SMB). Most of the time it crashes within the first 10s. Even Iperf traffic may provoke a crash in some situations. Fixed by using the front USB port.
  • Overall transfer rates are never as expected. 3650MBps(Tx)/3520MBps(Rx) is the upper limit measured with iperf. In SMB setups, Rx is always about 30% slower (3300MBps(Tx)/2100MBps(Rx))

Please have a look into the remaining two issues: Crash upon MTU change and low SMB write rates

Performance/Stability table

For Issue 2 and 3 I tried several combinations of link speed, MTU, Throttling and remote station. The interfaces are described in more detail in the following sections.

NAS interface Remote interface Link Speed MTU Throttling Used USB Port Iperf Tx Iperf Rx SMB Tx SMB Rx
QNA Intel 1G 9000 Yes 4 946 947 950 950
Synology Aquantia 1G 1500 - - 946 947 940 940
QNA Aquantia 5G 9000 Yes 4 1560 2040 1490 Crash
QNA Aquantia 2.5G 9000 Yes 4 1570 2050 2270 Crash
QNA Aquantia 5G 1500 Yes 4 1560 Crash Crash Crash
QNA Aquantia 2.5G 1500 Yes 4 1570 1720 1310 Crash
QNA Aquantia 5G 1500 No 4 1560 Crash 1530 Crash
QNA Aquantia 2.5G 1500 No 4 1530 Crash 1540 Crash
QNA Aquantia 5G 1500 Yes 2 3400 1500 1300 650
QNA Aquantia 5G 1500 No 2 2890 1780 1300 650
QNA Aquantia 2.5G 1500 Yes 2 2350 1800 1300 550
QNA Aquantia 2.5G 1500 No 2 2360 1770 1300 550
QNA Aquantia 2.5G 9000 Yes 2 2450 2470 2500 1800
QNA Aquantia 5G 9000 Yes 2 3640 3510 3300 2100
QNA Aquantia 2.5G 1500 Yes 2 2360 2140 1300 550
QNA Aquantia 5G 1500 Yes 2 3400 1740 1300 700
QNA Aquantia 2.5G 9000 Yes 2 2470 2460 2200 1800
QNA Aquantia 5G 9000 Yes 2 3650 3520 3300 2100
QNA Aquantia 2.5G 1500 Yes 3+4 1530 Crash - -
QNA Aquantia 5G 1500 Yes 3+4 1540 Crash 1300 Crash
QNA Aquantia 2.5G 9000 Yes 3+4 1580 2050 1500 1600
QNA Aquantia 5G 9000 Yes 3+4 1560 2050 1450 1600

Notes:

  • All data rates are in MBps and given relative to NAS.
  • Iperf3 was used for Iperf measurements. The remote station was always the server. Rx traffic was forced with "-R" flag. All measurements used seven parallel streams. Single streams runs are 200-300MBit/s slower. Native iperf3 was used.
  • SMB test was performed by copying an 8GiB file from NAS to remote station (Tx) and back (Rx).
  • Before Crashing on SMB transfers the transfer rates were approx. on Tx speed level.
  • The settings "USB Port 3+4" used a USB Y-cable to eradicate power issues. Port 4 was used for data and port 3 for supplementary power.

Outcome:

  • Thermal throttling does not affect the performance or stability as far as I can tell. May be this is due to the small files (<10GB) transmitted for testing.
  • The used network cable does not affect the performance or stability as far as I can tell. Tests with an old 10m Cat5e didn't yield other results than with a new 10m Cat6 cable.
  • The front USB port is much more reliable and performs much better. (Nevertheless it still has a problem with SMB RX traffic.)
  • The rear USB ports can't handle the adapter. Performance is bad and stability is even worse.

Description of your products

NAS

NAS is a DS916+ with 3 WD140PURZ and 1 WD80EFZX in RAID5.

  • uname -a : Linux DS916 3.10.105 #24922 SMP Wed Jul 3 16:34:56 CST 2019 x86_64 GNU/Linux synology_braswell_916+
  • DSM 6.2.2-24922 Update 4

Network Adapter

QNAP QNA-UC5G1T USB to 5GbE Adapter - QNAP System Inc.

  • Rev: 1.01 (as said by DSM USB device information)
  • Connected with the original USB-C USB-A cable.
  • Connected to the upper rear(4)/front(2) USB port.
  • An USB3-Y-cable was temporarily used.
  • Content of /etc/sysconfig/network-scripts/ifcfg-eth2: (touched by myself, configured via DSM)
DEVICE=eth2
BOOTPROTO=static
IPADDR=192.168.70.100
NETMASK=255.255.255.0
ONBOOT=yes
IPV6INIT=auto
IPV6_ACCEPT_RA=1

Description of your environment

NAS and remote station were connected directly. No Hubs or Switches.

  • Intel: Intel(R) I211 Gigabit Network Connection
  • Aquantia: Aquantia AQtion 10Gbit Network Adapter (AQC107)
    Both onboard on an ASRock X470 Taichi Ultimate. System is running Win10 (1809).

A 10m STP Cat.5e and a 10, SFTP Cat.6 cable were used for all test.

Output of dmesg command

There is no output after the cable was connected after 42000.
dmesg.txt
This is an exemplary output of one of the tests.

Output of lsusb command

Output when Adapter was attached at the rear port.

|__usb1          1d6b:0002:0310 09  2.00  480MBit/s 0mA 1IF  (xhci_hcd 0000:00:14.0) hub
  |__1-5         f400:f400:0100 00  2.00  480MBit/s 200mA 1IF  (Synology DiskStation 650064DA20E41658)
|__usb2          1d6b:0003:0310 09  3.00 5000MBit/s 0mA 1IF  (xhci_hcd 0000:00:14.0) hub
|__usb3          1d6b:0002:0310 09  2.00  480MBit/s 0mA 1IF  (Linux 3.10.105 etxhci_hcd-170202 Etron xHCI Host Controller 0000:04:00.0) hub
|__usb4          1d6b:0003:0310 09  3.00 5000MBit/s 0mA 1IF  (Linux 3.10.105 etxhci_hcd-170202 Etron xHCI Host Controller 0000:04:00.0) hub
  |__4-2         1c04:0015:0101 00  3.20 5000MBit/s 896mA 1IF  (QNAP QNAP QNA-UC5G1T USB to 5GbE Adapter 02I08329)

Output when Adapter was attached at the front port.

|__usb1          1d6b:0002:0310 09  2.00  480MBit/s 0mA 1IF  (xhci_hcd 0000:00:14.0) hub
  |__1-5         f400:f400:0100 00  2.00  480MBit/s 200mA 1IF  (Synology DiskStation 650064DA20E41658)
|__usb2          1d6b:0003:0310 09  3.00 5000MBit/s 0mA 1IF  (xhci_hcd 0000:00:14.0) hub
  |__2-2         1c04:0015:0101 00  3.20 5000MBit/s 896mA 1IF  (QNAP QNAP QNA-UC5G1T USB to 5GbE Adapter 02I08329)
|__usb3          1d6b:0002:0310 09  2.00  480MBit/s 0mA 1IF  (Linux 3.10.105 etxhci_hcd-170202 Etron xHCI Host Controller 0000:04:00.0) hub
|__usb4          1d6b:0003:0310 09  3.00 5000MBit/s 0mA 1IF  (Linux 3.10.105 etxhci_hcd-170202 Etron xHCI Host Controller 0000:04:00.0) hub

Output of ifconfig -a command

eth0      Link encap:Ethernet  HWaddr 00:11:32:5D:03:B9
          inet addr:192.168.60.10  Bcast:192.168.60.255  Mask:255.255.255.0
          inet6 addr: fe80::211:32ff:fe5d:3b9/64 Scope:Link
          inet6 addr: 2003:cd:770d:a300:211:32ff:fe5d:3b9/64 Scope:Global
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:21019885 errors:0 dropped:0 overruns:0 frame:0
          TX packets:13627187 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:27761844139 (25.8 GiB)  TX bytes:5625192053 (5.2 GiB)

eth1      Link encap:Ethernet  HWaddr 00:11:32:5D:03:BA
          inet addr:169.254.7.239  Bcast:169.254.255.255  Mask:255.255.0.0
          UP BROADCAST MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)

eth2      Link encap:Ethernet  HWaddr 24:5E:BE:4D:99:D9
          inet addr:192.168.70.100  Bcast:192.168.70.255  Mask:255.255.255.0
          inet6 addr: fe80::265e:beff:fe4d:99d9/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:3009123 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:3032428929 (2.8 GiB)  TX bytes:14303703448 (13.3 GiB)

lo        Link encap:Local Loopback
          inet addr:127.0.0.1  Mask:255.0.0.0
          inet6 addr: ::1/128 Scope:Host
          UP LOOPBACK RUNNING  MTU:65536  Metric:1
          RX packets:230267 errors:0 dropped:0 overruns:0 frame:0
          TX packets:230267 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:70866766 (67.5 MiB)  TX bytes:70866766 (67.5 MiB)

sit0      Link encap:IPv6-in-IPv4
          NOARP  MTU:1480  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)

Output of ethtool

The missing link mode adverstisement for 2.5G and 5G seems odd to me.

Settings for eth2:
        Supported ports: [ TP MII ]
        Supported link modes:   100baseT/Full
                                1000baseT/Full
        Supported pause frame use: No
        Supports auto-negotiation: Yes
        Advertised link modes:  100baseT/Full
                                1000baseT/Full
        Advertised pause frame use: No
        Advertised auto-negotiation: Yes
        Speed: 5000Mb/s
        Duplex: Full
        Port: MII
        PHYAD: 0
        Transceiver: internal
        Auto-negotiation: on
        Supports Wake-on: g
        Wake-on: d
        Current message level: 0x00000007 (7)
                               drv probe link
        Link detected: yes
@Balooforever
Copy link

Balooforever commented Apr 29, 2020 •
edited
Loading

Thank a lot for your idea to use the front port !
I have the same problem with the write speed (from Windows 550mb, from Linux 1,7Gb ... why ?) but i can read to 3Gb from NAS :)
With Jumbo Frame, 3,5Gb and 2Gb write 👍

@bb-qq bb-qq added the report Just information label Aug 11, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
report Just information
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@mkter @bb-qq @Balooforever