Welcome to GOFASTMOTORSPORTS.com ... ... ... Green Flag - Green Flag - Go Go Go ... ... Brought to you by Ricks Satellite - Where the Big Dish Rules!

Ricks Satellite Wildfeed and Backhaul Forum
Register Latest Topics
 
 
 


Reply
  Author   Comment   Page 6 of 7     «   Prev   3   4   5   6   7   Next
mountaineer

Avatar / Picture

Registered:
Posts: 2,643
Reply with quote  #76 
Thank you Rempfer! Great tests. I think the minme struggles even more because of the less amount of memory it has and of course no internal HDD bay.

i do think that the problem being seen is also affecting anything plugged in to the usb port. i know the genpix tuner does not act right, the Prof doesnt act right all the time, and I believe you said that the tbs tuner also isnt exactly right either. I do believe that the usb ports / software / drivers for them are causing the issues all the way around. Hopefully Daza and others can figure it out so I (we) can reliably use the 3.92 kernel for the minime as well. 

Thanks guys for everything!

__________________
3 Edison OS mini's, 3 Openbox S10s, 2 Openbox S9's, NFusion HD, 2 Prof 7500's, Genpix Skywalker 1, 2 Optibox Anacondas, Azbox me, Azbox minime, 2 Jynxboxes, ismart m100, izzibox, TBS 6982, Vigica C60s, Optibox Raptor, azbox Bravissimo, TBS 5922SE, TBS 5980 (Broke), and TBS 6902.
 36" KU Dish Linear, 10 ft prime focus Mesh, 7.5ft prime focus mesh dish C and KU.
7 Stationary KU dishes.
rempfer

Avatar / Picture

Registered:
Posts: 878
Reply with quote  #77 
Quote:
Originally Posted by mountaineer
i do think that the problem being seen is also affecting anything plugged in to the usb port. i know the genpix tuner does not act right, the Prof doesnt act right all the time, and I believe you said that the tbs tuner also isnt exactly right either. I do believe that the usb ports / software / drivers for them are causing the issues all the way around.

Excellent observation, mountaineer!  It might almost be easier to find and fix if this hammering of the USB bus/ports is occurring at all times due to some glitch in the 3.9.2 kernel that was not present in previous kernels.  Here's hoping anyway, for a simple fix (the nerve of me to say it!) that clears up multiple issues.  [smile]
dazulrich

Registered:
Posts: 204
Reply with quote  #78 
hi.
for those committed to the cause, i've compiled a new 3.9.2 test images with a few kernel level changes pertaining to the ext2/3 drivers and USB  configs. no idea if that changes anything.
oh, and this is a ubi image - it will boot about 20sec faster (was too much reverting to get back to jffs2 image for this test.)
 
https://www.dropbox.com/s/4hisub5okw1e2kw/openatv-6.0-azboxme-20170525_usb.zip?dl=0 (usb)
https://www.dropbox.com/s/xxn8xa6uyfceo7l/openatv-6.0-azboxme-20170525_webif.zip?dl=0 (webif)

this is really for testing only. (you can switch to another partition with jffs2 image (basically ANY other known azbox image out there.. the one attached is the only public ubi image for azbox i'm aware of)
during boot when you are prompted to switch boot (press 0,1,2) , the display will then also prompt whether to change image filesystem - press (any 0,1,2) to toggle. it will remember last boot partition and last boot filesystem..

enjoy and good luck... let me know if USB acts any different.
/daz
rempfer

Avatar / Picture

Registered:
Posts: 878
Reply with quote  #79 
Downloading as we speak.  :-)


dazulrich

Registered:
Posts: 204
Reply with quote  #80 
any issues, just shout....
mountaineer

Avatar / Picture

Registered:
Posts: 2,643
Reply with quote  #81 
Great work! Thanks Daz!  Thanks Rempfer for testing! 
__________________
3 Edison OS mini's, 3 Openbox S10s, 2 Openbox S9's, NFusion HD, 2 Prof 7500's, Genpix Skywalker 1, 2 Optibox Anacondas, Azbox me, Azbox minime, 2 Jynxboxes, ismart m100, izzibox, TBS 6982, Vigica C60s, Optibox Raptor, azbox Bravissimo, TBS 5922SE, TBS 5980 (Broke), and TBS 6902.
 36" KU Dish Linear, 10 ft prime focus Mesh, 7.5ft prime focus mesh dish C and KU.
7 Stationary KU dishes.
rempfer

Avatar / Picture

Registered:
Posts: 878
Reply with quote  #82 
Quote:
Originally Posted by dazulrich

https://www.dropbox.com/s/xxn8xa6uyfceo7l/openatv-6.0-azboxme-20170525_webif.zip?dl=0 (webif)

this is really for testing only. (you can switch to another partition with jffs2 image (basically ANY other known azbox image out there.. the one attached is the only public ubi image for azbox i'm aware of)
during boot when you are prompted to switch boot (press 0,1,2) , the display will then also prompt whether to change image filesystem - press (any 0,1,2) to toggle. it will remember last boot partition and last boot filesystem..

enjoy and good luck... let me know if USB acts any different.
/daz


Daz,

Recording to that same USB drive that I've been using all along I still see a great deal of activity on the external enclosure's LEDs and the recording stops after only a half-minute or so.  So seemingly still much more port activity with this kernel than with the 3.3.1 I've been using for comparison, and still not able to complete a three-minute recording to the external USB drive.  :-(

Thanks for the opportunity to test.

/Dick

dazulrich

Registered:
Posts: 204
Reply with quote  #83 
thanks for testing rempfer. shame.
Do you have an openrsi 3.0 image with kernel 3.9.2?
I would be curious to see the comparison of the USB activity (as this would be closest to your working openrsi 3.0 3.3.1 kernel image..
testing this with the ext2 formatted usb disk would remove all the changes in E2 images that could also contribute to this.

 Ihe oldest openrsi 3.0 3.9.2 kernel image i have is OpenRSI-3.0-azboxme-20130930-USB.zip. ( I can upload to dropbox if needed)

with azbox not including usbmon or debugfs functionality in the kernel, i can't see any easy way to analyse the usb traffic.. if anyone has an idea how to, please let me know.

rempfer

Avatar / Picture

Registered:
Posts: 878
Reply with quote  #84 
OpenRSI-3.0-azboxme-20140104-WEBIF is the only version I have, and I think it exhibits the recording problem.  I would be delighted to try the older version you have on hand, daz.

EDIT (after repeats of earlier tests):
Just to re-confirm, the OpenRSI-3.0-azboxme-20140104-WEBIF (Kernel 3.9.2) version above does suffer from the hyperactive USB recording failure problem as we already knew:  
- 3 minute (successful) recording to internal SSD 883,677,456 bytes
- 3 minute (unsuccessful) recording to USB EIDE HDD 105,019,056 bytes

Also as a reminder, OpenRSI with Kernel 3.3.1 records successfully to both devices (and without Kernel 3.9.2's USB hyperactivity).
dazulrich

Registered:
Posts: 204
Reply with quote  #85 
hi rempfer
uploaded the older openrsi image.
https://www.dropbox.com/s/x6ss9h1cv2ha0vz/OpenRSI-3.0-azboxme-20130930-USB.zip?dl=0
rempfer

Avatar / Picture

Registered:
Posts: 878
Reply with quote  #86 
Downloading now!  Please see edit to my previous post.
dazulrich

Registered:
Posts: 204
Reply with quote  #87 
thanks for confirming the bad behaviour with your later openrsi version.
just trying to make sure we can rule out / reduce the variables.  (as I can't build an old openrsi image.)
rempfer

Avatar / Picture

Registered:
Posts: 878
Reply with quote  #88 
Daz,

Using OpenRSI-3.0-azboxme-20130930-USB, we still get the same behavior as with the other kernel 3.9.2 images.  However i noticed something which may be a clue.

Yes, the USB hyperactivity is present and the video still aborts after about 20 seconds, but the quality of the short video that is recorded is perfect.  

However -- at the end of the video, the recording includes the Red-X dialog box, stating "Service invalid! (Timeout reading PMT)."  Why is the error message included in the video?  This makes me wonder if the extra USB activity might be a "system dialog engine" -- intended to interact with and output messages to the display only -- also streaming (normally an empty stream i.e. nothing) to the USB port when a recording is taking place.  I'm just making this up, but it does seem unusual somehow that this error in recording message was included in the recording.

Maybe I can try to reproduce this with the 3.3.1 kernel by bumping the dish to see if I can cause the error and then to see if the error message gets recorded into the video.  I'll try that now.

EDIT:
I was not able to cause the "timeout reading PMT" error with kernel 3.3.1.  All that happened when I moved the dish off the satellite during recording was the "spinner (box busy)" animated icon displayed and the recording was unsuccessful.

Just a bit more on the USB hyperactivity.  As mountaineer observed, this hyperactivity is likely not just present during recording sessions, since  we've seen problems with tuners -- my TBS5925 keeps [re-]searching for a lock when trying to tune a transponder, and when listening to audio streams the audio stream will stop playing after about 20 seconds, the same length of time it takes for a video recording to fail! -- and so forth.  It just seems like some global setting in the 3.9.2 kernel is causing an inappropriate stream of data to be redirected, or rather "duplicately directed" to the USB bus.  If so, this problem would have nothing to do with recording, or listening to audio streams, or tuning a transponder specifically.  The symptom would just be showing up in these areas.

EDIT2:
Just an afterthought.  When successful recording is occurring with kernel 3.3.1 I would estimate the red busy light on the USB2 attached external HDD stays on about 25% of the time, with the green no-activity light present the other 75% of the time.  When recording (attempting to record) with any kernel 3.9.2 firmware, the red busy light stays on 95% of the time until the (approx) 20 second mark where it locks on solid for 5-10 seconds or so and then goes to steady green at which point disk writes appear to have stopped.  I apologize if I'm overthinking this.
dazulrich

Registered:
Posts: 204
Reply with quote  #89 
Thanks lot for that.
I know how overthinking feels. Happens to me all the time.
So comparing openrsi 3.0 with different kernels has taken out any major E2 or openembedded changes.
Leaving the actual kernel update (there are other images with 3.9.6 which don't seem to have that problem ) or the sigma related driver patches that were put in for 3.9.2. my guess is that those patches or drivers are not quite right.
Problem is how to simulate. I'll have a look if I can find a USB housing with activity LED.
dazulrich

Registered:
Posts: 204
Reply with quote  #90 
Rempfer.
Could you do another failed recording and maybe even stage the problem with the TBS5925 then post the messages and dmesg logs?
Maybe they contain some clues we haven't seen.
Edit: could you post the output of "mount" command on both the 3.3.1 and the 3.9.2 box with the USB hdd attached.
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Fellow Members, your posts are welcome here! Do not worry about posting everything perfect. Different receivers and LNB's will give you different Frequencies and Symbol Rates. Some set top boxes, PCI cards and USB receivers, Do Not Require all of the same information that others may need. It is not Required to post everything that others may need to tune in a feed. It is just most important to share the find. We can always adjust the Frequency and Symbol Rates and try the various Modulations and FEC's on our own receivers until we get a lock and then give a polite reply with what works for your receiver, as that information might help others as well. We all appreciate the efforts and energy of the Posters!

Thank You for Visiting GOFASTMOTORSPORTS.com - Keep Your Eyes on the Sky and the Track!