Project

General

Profile

Actions
Copy link

Bug #1578

closed

USB-Tuner: dvb-usb: bulk message failed: -110 (5/0)

Added by Patrick Vorlicek over 12 years ago. Updated about 12 years ago.

Status:
Rejected
Priority:
Normal
Assignee:
-
Category:
DVB
Target version:
-
Start date:
2013-01-29
Due date:
% Done:

0%

Estimated time:
Found in version:
latest git
Affected Versions:

Description

Hello,

I'm not quite sure, whether this can be a bug in tvheadend, but as I don't know what to do anymore, I would you like to have a look at this issue.

I use tvheadend with three usb dvb-s2 tuners (TT-connect S2-4600). This setup works quite well, though I get thoses errors every three till five days (!). It's somehow like it gets triggered when tunening to a new channel. It looks like this:


dvb-usb: bulk message failed: -110 (5/0)
dw2102: i2c transfer failed.

and sometimes


dw2102: i2c transfer failed.
dvb-usb: bulk message failed: -110 (4/0)

in dmesg. I googled a lot, initially this happend nearly every day. I changed the usb cables and played a bit with the ports, its an x64 system with intel chipset and 4GB RAM. I even upgraded my debian to 3.7.3 kernel and removed all unnecessary drivers and modules. On the server only samba, apache, tvheadend and oscam are running. Even though it seems to happen more often if the cpu load is high, it also happend while nearly idleing. Tvheadend becomes totally unresponsive and has to be killed. Unloading and reloading dvb-modules doesn't help, so I need a cold reboot to get the tuner running again, which is quite annoying. And because it's always another tuner of those three who fails, I don't think they have a hardware problem. Temperature is no issue and they are external powered.

I use this driver as it is recommed by the manufacturer: https://bitbucket.org/liplianin/s2-liplianin-v37

The driver is also build at minimal config, only needed modules. There are two firmwares available, both seem to behave the same.

Tvheadend gets new build from git every week (+/-), last update is two hours ago. At least I found a way to provoke this failure more reliable: heavy skipping in a recording played via pvr.hts (xbmc). It only took 30 secs. and tvheadend was gone with one of the tuners.

Is it possible, that there is a timing issue? That the tuner waits under some circumstances to long for a command and stops working? Can you please have another look at the code in charge?

I'm quite happy with this config and still think this is the best thing I ever spend so much time with, but with this issue I can't get a "production" environment.

Thanks in advance...

Actions
Copy link

Also available in: Atom PDF