13 replies to this topic

[Bust3D]

Hello there,

 

i just bought a Vu+ Uno4K SE and put OpenPLi because of my favorite image

 

Firmware version: OpenPLi 7.2-rc (2019-11-20-rc)
Kernel / Drivers: 4.1.20 / 4.1.20-20190424.r0.1

 

I use also an SSD Drive witch i used in previous box just mounted in the new one.

Now what i saw after some testings that the LAN 1Gbit port running well ... if i Copy from box to my PC i am able to copy with 110MByte/sec this is cool.

If i upload to box from my PC i reach 90Mbyte/sec witch is also cool.

The problem i am facing with is when i Copy to box i got packers in error(s) and this only if i Copy to it if i am Copying from the errors: stays at 0.

 

root@vuuno4kse:~# ifconfig
eth0      Link encap:Ethernet  HWaddr xx:xD:Ex:x4:x6:xx
          inet addr:192.168.88.204  Bcast:192.168.88.255  Mask:255.255.255.0
          inet6 addr: fe80::21d:ecff:fe14:b634%lo/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:12715029 errors:2050 dropped:2608 overruns:0 frame:0
          TX packets:5496586 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:326568296 (311.4 MiB)  TX bytes:3723057894 (3.4 GiB)

 

I have a Miktorik Router and monitored the tx and rx packets and there is now errors showing up.

Cable issue couldn't be if there no error when download ... (using cat7 cable).

 

Maybe to mention that i have also a conax cam stick in the CI, i don't know if that can causing problems like this.

 

Maybe someone else can also test.

 

King Regards.

[WanWizard]

I think it's more a matter of the interface not being able to keep up. An STB doesn't have separate hardware, everything is integrated in the SoC.

[betacentauri]

Is it really a problem? Yes, errors in general are bad, but you have high transfer rates and I guess all transferred files don’t show any errors. Or are the transferred files corrupt?

[Pr2]

You have a problem on the network side.
You should know that autodetect feature is a real ethernet protocol, so both sides must be configured the same way.
So do not force the port speed and duplex on your switch since STB are by default in auto-negotiate mode, if you have auto-negotiate on one side and fixed speed on the other side then you will have such high error rate problem.

I have cisco switches connected to my STB and I have 0 errors.

[Bust3D]

@WanWizard I think you have right, cause the " RX bytes:326568296 (311.4 MiB)  TX bytes:3723057894 (3.4 GiB)" doesn't reflect the reality, in certain point i think the interface make reset.

 

@betacentauri For you is not a problem of course  but all my other devices on same network and my previous STB all show me 0 errors and this should be like that. Regarding to Files all the files are not be corrupted after the transfer so this is not a problem ... just he errors ... why they appears ? And why the other my devices are good ?

 

@Pr2 My network is fine ... I never touched the autodetect feature and never forced the speed or duplex ... and all my other devices as well same settings everything is auto and no errors at all non even dropped packet , only on mentioned STB.

 

Is any of you did tested to transfer files from a PC to STB ? or just guessing ?

 

Kind Regards.

[betacentauri]

Do you have other data transfers running between other devices in your LAN. It might be that the router is not able to handle more than x Gb/s and so need to drop some packets.

Edited by betacentauri, 1 December 2019 - 11:29.

[Bust3D]

@betacentauri I have a RB962UiGS-5HacT2HnT Mikrotik Router witch able to handle x Gb/s devices same time. During the test there was no other LAN activity witch has passed more than 1-2 mbit / sec. And as i told the my old box Mut@nt HD51 has never had errors on same network same cable. And also wrote that i tested via Winbox (mikrotik router app ) and monitored the traffic on LAN3 Port where the VU+ is and there was 0 errors but on VU+ raised till 2050 ( RX packets:12715029 errors:2050 dropped:2608 overruns:0 frame:0).

[Pr2]

If you have a manageable router do you see also errors on the router port where the VU+ is connected too?
Did you enable jumbo frame on it? No every devices support them.

[Bust3D]

@Pr2 Sure i monitored the errors on router where the VU+ connected ... but there was 0 errors ... but the same time on VU+ was raised till 2050 errors and some Dropped packets.

Jumbo frame never heard about it and my router has no options like that. 

All settings are default Auto negotiation speed and duplex. Interesting why the Mut@nt HD51 on same cable same image OpenPLi has no errors ...

[Pr2]

Then it sounds like a Vu+ driver bug.

[WanWizard]

I don't think it is a bug, as such.

 

The STB simply doesn't have the hardware to process 100MB/s+ incoming (the SATA interface in the SoC can't process that either). If you try to push it in, you'll cause overruns and data drops... Assuming it's TCP, it is not relevant, it will take care of any retransmissions needed.

 

You can't compare the SoC with an Intel southbridge chip, or dedicated ethernet chipsets.

[Bust3D]

Somebody else can test like me ?

Copy form PC to STB via samba client ... (Windows Share) ... one big file to HDD

and also Copy from STB to PC same samba client ...

and report here Speed and errors from ifconfig.

 

PS. On other STB's not just like VU+ also ... to see if it's my STB hardware is faulty or what's wrong with this?

Any who can make this test please reply here. Thanks.

 

King Regards.

[WanWizard]

iperf3 is available in the feed, for test purposes.

 

From my Duo 4K to my NAS (over two gigabit switches):

root@vuduo4k:~# iperf3 -c 172.16.1.31
Connecting to host 172.16.1.31, port 5201
[  4] local 172.16.1.56 port 46368 connected to 172.16.1.31 port 5201
[ ID] Interval           Transfer     Bandwidth       Retr  Cwnd
[  4]   0.00-1.00   sec  60.2 MBytes   505 Mbits/sec  360   31.1 KBytes       
[  4]   1.00-2.00   sec  75.9 MBytes   637 Mbits/sec  572   29.7 KBytes       
[  4]   2.00-3.00   sec  75.8 MBytes   636 Mbits/sec  577   29.7 KBytes       
[  4]   3.00-4.00   sec  75.8 MBytes   635 Mbits/sec  605   24.0 KBytes       
[  4]   4.00-5.00   sec  76.1 MBytes   638 Mbits/sec  585   32.5 KBytes       
[  4]   5.00-6.00   sec  75.6 MBytes   634 Mbits/sec  595   49.5 KBytes       
[  4]   6.00-7.00   sec  75.8 MBytes   636 Mbits/sec  558   32.5 KBytes       
[  4]   7.00-8.00   sec  75.6 MBytes   634 Mbits/sec  506   31.1 KBytes       
[  4]   8.00-9.00   sec  75.6 MBytes   634 Mbits/sec  684   35.4 KBytes       
[  4]   9.00-10.00  sec  75.3 MBytes   631 Mbits/sec  710   33.9 KBytes       
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth       Retr
[  4]   0.00-10.00  sec   742 MBytes   622 Mbits/sec  5752             sender
[  4]   0.00-10.00  sec   741 MBytes   622 Mbits/sec                  receiver

iperf Done.

As you can see here, also iperf3 causes packet corruption and retries on the side of the box.

 

I have already explained to you twice why.

[Bust3D]

@WanWizard you hare right ... you just wrote here twice what is the problem ... and also your answer was accepted form my side as the best Answer.

I wrote to others if someone else also can test it with different or same VU+ Model or Other STB like Mut@nt or Dreambox or other witch has Gigabit LAN.

Me just want to have more tests on this, but your answer is accepted as the best and logical answer.

 

Kind Regards.