Project

General

Profile

Bug #4877

TVHeadEnd won't be triggered to get OTA EPG

Added by David Pollard almost 7 years ago. Updated almost 7 years ago.

Status:
Fixed
Priority:
Normal
Assignee:
Category:
EPG
Target version:
-
Start date:
2018-01-20
Due date:
% Done:

100%

Estimated time:
Found in version:
4.3-991~g9718e8e01
Affected Versions:

Description

Usually, I can click on "Trigger OTA EPG Grabber" in the EPG Grabber tab, and off it goes and subscribes to the relevant muxes to suck the data.

(I have EIT Grabber and FreeSat Grabber enabled)

All worked fine with 4.2-stable and all spare tuners would go and get the EPG.

Have just upgraded to the latest (4.3-991~g9718e8e01) (available from linuxserver/tvheadend) docker image and I can no longer trigger the EPG scan.

After the upgrade, all the EPG written to disk was not reused either.

Good News though, the memory leak issue (http://www.tvheadend.org/issues/4867) seems to no longer be present

Mystery.


Files

Screenshot from 2018-01-21 12-26-35.jpg (39.2 KB) Screenshot from 2018-01-21 12-26-35.jpg debug web options David Pollard, 2018-01-21 13:26
epg.PNG (63.8 KB) epg.PNG no epg sources in channels tab Max M, 2018-01-22 17:50
tvh.PNG (35.9 KB) tvh.PNG EIT DVB Grabber Max M, 2018-01-22 17:50

History

#1

Updated by Mark Clarkstone almost 7 years ago

4.3-991 - Can't reproduce here. Seems to be triggering OK..

2018-01-20 13:27:18.760 mpegts: 11306.5H in Astra 28.2°E - tuning on Montage Technology M88DS3103 #2 : DVB-S #0
2018-01-20 13:27:18.794 subscription: 005A: "epggrab" subscribing to mux "11306.5H", weight: 4, adapter: "Montage Technology M88DS3103 #2 : DVB-S #0", network: "Astra 28.2°E", service: "Raw PID Subscription" 
2018-01-20 13:27:18.795 mpegts: 674MHz in Sutton Coldfield - tuning on Silicon Labs Si2168 #3 : DVB-T #0
2018-01-20 13:27:21.842 subscription: 005C: "epggrab" subscribing to mux "674MHz", weight: 4, adapter: "Silicon Labs Si2168 #3 : DVB-T #0", network: "Sutton Coldfield", service: "Raw PID Subscription" 
2018-01-20 13:27:21.848 linuxdvb: Unable to provide UNC value.
2018-01-20 13:27:21.871 linuxdvb: Unable to provide signal strength value.
2018-01-20 13:27:21.871 linuxdvb: Unable to provide BER value.
2018-01-20 13:27:21.871 linuxdvb: Unable to provide SNR value.
2018-01-20 13:27:21.871 linuxdvb: Unable to provide UNC value.
#2

Updated by Jaroslav Kysela almost 7 years ago

The data should be migrated. Show some logs, there must be some errors.

#3

Updated by David Pollard almost 7 years ago

There was nothing in the logs at all, even with debug on ?!?!

Tried in an incognito window in case anything was cached, still the same.

I've reverted back to an older version which is working fine (stable 4.2.1)

The updates for the docker image are done weekly, so I'll test again in a week and see if the issue is still there.

Thanks

dp

#4

Updated by David Pollard almost 7 years ago

Tried this again this morning, again absolutely NOTHING in the logs. It's as if clicking on the button in the web interface does nothing at the backend.

I even enabled debug logging and put http and epg in the trace subsystems options....

Still nothing.....

But, when I exited TVHeadEnd to go back to 4.2.5, I did see this.....Not sure if the 3rd line from the bottom is relevant.....

2018-01-21 10:20:55.257 [ INFO] mpegts: 11508.5V in DVB-S Network (0x556b932ac100) - deleting
2018-01-21 10:20:55.258 [ INFO] mpegts: 11127V in DVB-S Network (0x556b932abcc0) - deleting
2018-01-21 10:20:55.258 [ INFO] mpegts: 10876V in DVB-S Network (0x556b932ab880) - deleting
2018-01-21 10:20:55.258 [ INFO] mpegts: 10832H in DVB-S Network (0x556b932ab440) - deleting
2018-01-21 10:20:55.258 [ INFO] mpegts: 10744H in DVB-S Network (0x556b932ab060) - deleting
2018-01-21 10:20:55.258 [ INFO] mpegts: 12480V in DVB-S Network (0x556b932aacc0) - deleting
2018-01-21 10:20:55.258 [ INFO] mpegts: 12266H in DVB-S Network (0x556b932aa920) - deleting
2018-01-21 10:20:55.258 [ INFO] mpegts: 12051V in DVB-S Network (0x556b932d9a20) - deleting
2018-01-21 10:20:55.258 [ INFO] mpegts: 12032H in DVB-S Network (0x556b932d95e0) - deleting
2018-01-21 10:20:55.258 [ INFO] mpegts: 10832.25H in DVB-S Network (0x556b932d91a0) - deleting
2018-01-21 10:20:55.258 [ INFO] mpegts: 11641.5H in DVB-S Network (0x556b9326d3a0) - deleting
2018-01-21 10:20:55.258 [ INFO] mpegts: 11817V in DVB-S Network (0x556b9326cf60) - deleting
2018-01-21 10:20:55.258 [ INFO] mpegts: 11719.5H in DVB-S Network (0x556b92cf2ca0) - deleting
2018-01-21 10:20:55.461 [ ERROR] epg: log buffer full
[s6-finish] sending all processes the TERM signal.
[s6-finish] sending all processes the KILL signal and exiting.

#5

Updated by Mark Clarkstone almost 7 years ago

I might be stating the obvious here, but you'll only get a complete log by using the debugging options.

#6

Updated by David Pollard almost 7 years ago

Hi Mark

Do you mean I need to start the process with debug parameters on the command line ? In other words, the debugging options in the attached graphic don't work ?

Thanks

dp

#8

Updated by Max M almost 7 years ago

same for me
tvh is set to force epg scan when started, even if I trigger epg scan in webui scan doesn't start.
Eit grabber is activated but I can't select epg source in channels tab.

epggrab trace https://paste.ubuntu.com/26438346/

#9

Updated by Jaroslav Kysela almost 7 years ago

Could you show me old (original) config - directory <CFG>/epggrab/otamux ?

The fix should be really simple - remove this directory or retrigger the scan for all muxes, but if you like to know why the old files are unuseable, follow my instructions.

#10

Updated by Max M almost 7 years ago

I removed the mentioned directory and made a rescan of all (active) muxes.
Epg grabber is fine again.
Forgot to copy config directory before deletion - sorry.
Is there a way to avoid this issue in future?
Thanks!

#11

Updated by Jaroslav Kysela almost 7 years ago

  • Status changed from New to Fixed
  • % Done changed from 0 to 100

Applied in changeset commit:tvheadend|181065ceb6865065fd42afe9efadd601c8809fb2.

#12

Updated by Jaroslav Kysela almost 7 years ago

It was a bug in master, it's fixed now.

Also available in: Atom PDF