Jump to content


Photo

Problems with network mounts


  • Please log in to reply
34 replies to this topic

Re: Problems with network mounts #21 WanWizard

  • PLi® Core member
  • 68,312 posts

+1,719
Excellent

Posted 15 October 2021 - 14:16

I can't reproduce it, playing the recording on a Duo4K disk by a SF8008 over a CIFS mount shows the normal bandwidth usage,

Attached Files


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: Problems with network mounts #22 Beanie

  • Senior Member
  • 66 posts

0
Neutral

Posted 16 October 2021 - 07:20

Thanks WanWizard. If I get a recording to play, then the bandwidth looks normal, but if it won't play, or it won't browse the movie folder properly, then the bandwidth spikes right up.



Re: Problems with network mounts #23 WanWizard

  • PLi® Core member
  • 68,312 posts

+1,719
Excellent

Posted 16 October 2021 - 14:20

Lots of retries? What does ifconfig tell you?


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: Problems with network mounts #24 Beanie

  • Senior Member
  • 66 posts

0
Neutral

Posted 17 October 2021 - 02:36

Running ifconfig on both client and server devices. No errors on any device.

 

On bother server devices (Solo2 OP 7.3 & H9 OP 8.1) TX traffic increasing at about the same bitrate of recording that is being played. Client (Zero OP 7.3) shows RX traffic increasing at expected rate.

 

However, when a recording will not play from the OP 8.1 mount, traffic spikes to around 8MB/s (send from the server and received on the client). If you exit from trying to play, return to the movie list, and play the same movie again, it will work the second time around. I cannot confirm, but suspect it is trying to read in the entire .ts file the first time around which is causing the traffic spike. On the second attempt at the same file, it plays normally.



Re: Problems with network mounts #25 WanWizard

  • PLi® Core member
  • 68,312 posts

+1,719
Excellent

Posted 17 October 2021 - 16:13

Are the meta files missing for that recording? It could be it scans the file to determine the length if it can't find the meta data. A developer should confirm that, it's not really my thing.

 

It also means that if it is "normal" behaviour, eventually it will start to play, if you wait long enough.


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: Problems with network mounts #26 betacentauri

  • PLi® Core member
  • 7,185 posts

+323
Excellent

Posted 17 October 2021 - 16:28

Afaik if .meta file is missing or don’t include all informations, full .ts file is read.
So maybe make a new recording. After recording has finished, open the movie list in the box which has done the recording. Then try to open it from the remote box.

And afaik if a recording is running the meta file is not valid and the ts is fully read every time you open the movie list until recording finishes.
Xtrend ET-9200, ET-8000, ET-10000, OpenPliPC on Ubuntu 12.04

Re: Problems with network mounts #27 Beanie

  • Senior Member
  • 66 posts

0
Neutral

Posted 18 October 2021 - 03:47

The recordings are complete, with all meta files, .ts .eit .ap .cuts .meta .sc but I don't think they are able to be read correctly.  Just viewing the movie list takes up to 20 seconds per movie to appear on the screen (several minutes to show 20 or so movies in the movie folder), and sometimes movies will have no date etc. like the meta files aren't read. I think this is just another symptom of the underlying problem of not being able to read the share properly.

 

Downgrading the server box to OP 7.3 but retaining the same HDD with the same movie recordings resolves the problem instantly, or with the USB HDD, plugging it into the Solo2 (OP 7.3) and it works fine. I am satisfied the recordings are not at fault.

 

Interestingly if I mount an OP 8.1 samba share in Mac OS X, it works flawlessly. Looking at smb status, the only difference I see is OS X connects at SMB v3.11 whereas the E2 CIFS client connects at SMB v2.10. I can't force a higher SMB version from OpenPLi. Is 2.10 the highest the CIFS client accepts?


Edited by Beanie, 18 October 2021 - 03:48.


Re: Problems with network mounts #28 WanWizard

  • PLi® Core member
  • 68,312 posts

+1,719
Excellent

Posted 18 October 2021 - 08:46

The SMB version(s) supported depend on the kernel version. If you don't specify a version in the mount options in the mount manager, OpenPLi will try to determine it, starting with the highest possible,

 

But the version shouldn't be responsible for a huge performance difference.

 

Have you also checked the other parameters for differences on the client box, between a 7.3 and 8.1 server box? Such as rsize and wsize?


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: Problems with network mounts #29 Beanie

  • Senior Member
  • 66 posts

0
Neutral

Posted 19 October 2021 - 07:48

Haven't gone through every parameter, but both are using the same rsize & wsize. Both default to 64k, but have tried as low as 4k. Reducing r/w size speeds up response time (from 8.1), but the behaviour is still the same.



Re: Problems with network mounts #30 WanWizard

  • PLi® Core member
  • 68,312 posts

+1,719
Excellent

Posted 19 October 2021 - 11:48

Weird. I'm running out of ideas...


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: Problems with network mounts #31 Beanie

  • Senior Member
  • 66 posts

0
Neutral

Posted 20 October 2021 - 03:09

Me too! Thank you for your help though.



Re: Problems with network mounts #32 madiv

  • Senior Member
  • 25 posts

0
Neutral

Posted 20 October 2021 - 18:41

Hi Badblake

 

A 2¢ idea:

Did you check that transferring several GB files by ftp from your box to a computer shows:

- high enough transfer rate

- nearly constant transfer rate = not dropping to <1MB/s time to time

 

Background:

My Duo2 had slow reactions when moving across the record list of another box & the spinner appeared time to time while watching movies...

I noticed that FTP speed dropped time to time to ca. 500kB/s.

It turned out that a network component between the Duo2 and the other box was damaged and unfortunately still kept working most of the time, but time to time made the bandwidth shrink to ca 500kB/s. Replacing it solved the problem.

 

Above may not apply to Beanie's problem as 'Downgrading the server box to OP 7.3 but retaining the same HDD with the same movie recordings resolves the problem instantly, or with the USB HDD, plugging it into the Solo2 (OP 7.3) and it works fine. I am satisfied the recordings are not at fault.'.

 

BTW OpenPLi 8.1 improved my Duo2 'Gb' lan bandwidth vs (old) VTi: ca 40~50MB/s vs ca 30~40MB/s via FTP. Not catching a real Gb bandwidth, but thanks OpenPLi.


Edited by madiv, 20 October 2021 - 18:42.


Re: Problems with network mounts #33 BadBlake

  • Senior Member
  • 98 posts

+1
Neutral

Posted 24 October 2021 - 16:10

Hi Badblake

 

A 2¢ idea:

Did you check that transferring several GB files by ftp from your box to a computer shows:

- high enough transfer rate

- nearly constant transfer rate = not dropping to <1MB/s time to time

 

Background:

My Duo2 had slow reactions when moving across the record list of another box & the spinner appeared time to time while watching movies...

I noticed that FTP speed dropped time to time to ca. 500kB/s.

It turned out that a network component between the Duo2 and the other box was damaged and unfortunately still kept working most of the time, but time to time made the bandwidth shrink to ca 500kB/s. Replacing it solved the problem.

 

Above may not apply to Beanie's problem as 'Downgrading the server box to OP 7.3 but retaining the same HDD with the same movie recordings resolves the problem instantly, or with the USB HDD, plugging it into the Solo2 (OP 7.3) and it works fine. I am satisfied the recordings are not at fault.'.

 

BTW OpenPLi 8.1 improved my Duo2 'Gb' lan bandwidth vs (old) VTi: ca 40~50MB/s vs ca 30~40MB/s via FTP. Not catching a real Gb bandwidth, but thanks OpenPLi.

Hey! Thanks for the tip. That did it! It seems that a internet radio that's connected to my LAN created this havoc -something I noticed after pulling the plug. Now I'm running OpenPLI 8.1 on all three of my Vu+ Duo 2's and working great. Thanks again :)



Re: Problems with network mounts #34 madiv

  • Senior Member
  • 25 posts

0
Neutral

Posted 25 October 2021 - 20:18

Happy that it helped & thx for your feedback :)



Re: Problems with network mounts #35 BadBlake

  • Senior Member
  • 98 posts

+1
Neutral

Posted 2 November 2021 - 12:10

Happy that it helped & thx for your feedback :)

I think I left something out though. After I reinstalled image on all three I was given the option of using NFS instead og Cifs. When using NFS there was no problem reading from mounts on remote boxes.




1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users