Tvheadend

Well I tried the new versions today, the channel editor appears to be working great now. For future reference (when you run out of other things to work on) it might be nice to have a couple of toggle buttons that would enable/disable all HD, or all SD channels, and also maybe an "invert selection" button that would invert the checked status of all channels. But those are very minor things because once you've gone though and selected the channels you want or don't want, you probably won't need to do it again any time soon.

The bigger issue I am seeing with the new version is that when I send the channel list to VLC, for some reason VLC skips over a lot of channels rather than playing them. I don't know if this is a VLC issue or an issue in the software but I do know that every time it happens I see a couple of lines like this:

2021-05-15 17:59:18,323-WARNING:web_tuner Unknown channel id ###
2021-05-15 17:59:18,323-WARNING:web_handler [ ip address ] "GET /Locast/watch/### HTTP/1.1" 501 -

Where ### is the channel ID. I don't recall seeing this with previous versions. I am testing with three instances now and the problem strikes random channels on all instances. My belief is that something is not waiting long enough for a channel to "lock in" because when it skips a channel it does so very quickly, and often will quickly skip the next three or four channels in the list (give or take) before finding one it likes. Sometimes if you go back and click on a channel again it will play it, or it may take several tries, or it may not work at all. I have only tested this in VLC so far, it is quite possible that this might not be an issue in Tvheadend but I need to make some changes to the backend configuration before I can test that. Have you ever encountered this in VLC?

Other than that, using three instances doesn't appear to be a problem now, and I was able to get a xmltv.xml file with only my selected channels without crashing the system, so either that is fixed, or just reducing the number of channels causes it to not use so much memory.

Quick question, when I upgraded to the 0.8.4a version this afternoon I lost my edited channels list. Normally what I do with a new version (in Linux) is unzip the source file into a new directory, then copy the config file from the current directory to the new one, then rename the current and new directories so that the new one has the same name as the previous (tvheadend-locast). This has always worked in the past but today it lost the editing I had done on the channels list. I suspect what I should have done is copy the data directory to the new version's directory but didn't know if doing that would mess anything up since you had told us to delete the contents of that directory before installing 0.8.4. But if I had copied or moved the data directory over would that have preserved my channels list without causing any other issues?

Thank you again for all you are doing on this!

I was not saying I was able to edit the channels list in the past, I was saying that I have always used the same method of upgrading in the past, creating a new directory and copying my settings from the old one and then renaming both. When I did both upgrades today I did remove the [loggers] section. All I was asking is, when doing an upgrade how do I not lose any channel edits I have made? Of course this was not a problem in the past because you could not do channel edits. It was only a problem when I went from 0.8.4 to 0.8.4a.

As for those 501 errors, why they are so confusing to me is that I am clicking on the link that goes to http://ip:6077/channels.m3u (the one that reads "channels.m3u: M3U formatted channel list") and my web browser asks me if I want to download it or open with VLC (it downloads in either case). I say open with VLC and that is where VLC is getting the channel list from. If I open http://ip:6077/Locast/lineup.json it is giving me the same list, just in a different format. I can look at the errors that VLC is throwing and see that they contain the same URLs that are in the lineup.json file. If I cut it back to a single instance the problem goes away. You may not be seeing it because you are testing two same instances that only have different enable and disable settings. I was testing using three different instances with different logins and also specifying different overrides in each instance. The channel lineups were all mixed together in the channels lists (sorted by channel number rather than by instance) so it was trying to quickly switch from one lineup to another (because when VLC fails it immediately tries the next item in the list; I REALLY wish there were a way to shut that off, it is one thing I hate about VLC!) but still that should not have been an issue because each instance had its own login. But as I say if I went back a single instance then it seems to work fine and not skip channels (by the way it also lost all my channel edits again when I did that, but that's not my main concern). So I am not sure what is going on.

EDIT: I guess that for now you really have to go in and delete the files in the data directory if you are removing instances, otherwise some data from the removed instances remains and it tends to confuse the program (for example you still see tabs for the removed instances).

I happened to look and saw you had released 0.8.4b, and it fixes the issue I mentioned in my previous posts. All the channels in all instances are working reliably now, did not see a single skip. Great job! And thanks much!!

EDIT: I had this page open in my browser so did not see your two posts above before I posted this, but thanks also for that information, glad you were able to track it down. I'm not too worried about the issue with deleting instances since once I get it configured the way I want, I doubt I will ever change it or remove an instance. It definitely won't be a frequent occurrence.

I have a couple of questions/comments about your top post.

You say, "Make sure to set the number of tuners to 1. Locast does not like the speed of channel changes during initial scanning. Doing a force scan will create the mux and service values." I assume by that you mean the value of the " Maximum # input streams " setting. By default it is 0 and you are correct, if you leave it there most of the created muxes will fail until you manually force scan them. Setting that value to 1 limits Tvheadend to scanning one at a time. But my question is, after you have done the initial scan, should you set it back to 0? Or whatever the limit is for the number of simultaneous tuners Locast will support, and if so do you happen to know what that number is? I am pretty sure that the "2" in your screenshot is not right because if I recall correctly Locast allows you to use four or five tuners simultaneously per account, but I don't know which is correct. Also just a suggestion, in the Advanced section of the Network settings I would set the " Re-fetch period (mins) " to something other than zero, because if you don't it will re-fetch your channel list every minute, which is quite unnecessary once everything is configured. I set it to 720 so it will re-fetch every 12 hours, but might go to something even higher once all the configuration is finished.

Also, you say, "Next you can setup the grabber. I use a URL grabber which is an extremely small file" but you don't mention what the name of the small file is or where to get it, and the screenshot doesn't have enough resolution to see it. You may want to mention where that small file can be obtained, if it is still available on the web. Otherwise there is always the option to use wget or curl to get the xml file where you need it, and then use tv_grab_file or a socket-based grabber or whatever you would normally use to import the file into Tvheadend, and those actions could be combined in a small shell script.

The only thing I am not sure about is where you say: Also, I set the Network discovery in the table to "Disable".

In the version of TVH that I have there is no "Network discovery" option in the "Edit Network" dialogue. There is the "Re-fetch period (mins)", which taking your advice I have set to 43200 which is 30 days, and while in there I set the "Maximum # input streams" to 4 -- thank you for the explanations on those. But I do not see anything called "Network discovery" or anything similar that would take a non-numeric value such as Disable. The only thing I found that even came close was in the screen that came up when I clicked on Help...

IPTV Automatic Network - Don't Probe Services

Tvheadend will by default probe each playlist entry for service information. Some service providers do not allow such probing & will deny (or rate limit) access, leading to scan failures.

To create services without probing, Service ID must be set (usually to 1) and the Scan after creation check box un-ticked.

Note, the above two settings are only visible with the view level set to Expert.

But I think that is only for when the network is first created. I believe I am running a build of TVH from 2019, it is version 4.2.something, so it is possible that setting may only be in newer versions. The reason I have not updated is that I have been burned by updates in the past that caused things to stop working, so I usually only update TVH right after updating Ubuntu, and right now I'm still on Ubuntu 18.04. I figure that as long as it is working reliably, there is no good reason to update.

(...several hours later...) I know someone that is running a 4.3 version of TVH and I checked with them and they can't see a "Network discovery" option either, except in their ATSC network, and I can see it there in mine as well in the "Advanced Settings" section. But it just does not appear for the Locast network, which was created as an "IPTV Automatic Network" as shown in the top post.

Oh, I thought it was supposed to be in popup you see when you add or edit a network. I see it now, and apparently it was set to Disabled by default, and I did not change it. Thanks for explaining that.

Tvheadend is funny, it is the only software I have ever used that doesn't suggest that regular users stay with the stable version, but instead suggests they use the beta or whatever they call it. The only reason I am reluctant to do that is that several years ago I had it set to auto update every time a new release came out (in other words I had not commented out their repository in my apt sources list as I do now), and there was an update that caused Tvheadend to stop working! And when you have people who are using the software as their primary means of watching and recording TV, well let's just say I heard about it immediately and with great emphasis! So now as long as I am not having issues I don't let it update anymore unless I am also updating the operating system, and the only reason I do it then is because typically I can swap out a hard drive and put the new OS version on a totally different drive, and that way if I have major problems I can go back to the previous version simply by swapping the old hard drive back.

I would probably be a whole lot less reluctant if Ubuntu didn't make it so difficult to go back to a previous version of a program, especially with programs installed using apt. They just assume that no one would ever have a reason to go back a version so they don't give you a way to save a package in the same way that MacOS or Windows does. But the other issue is that as far as I can tell, Tvheadend itself offers no way to backup and restore a configuration from within the GUI. So even if you could drop back a version, it might not work if any databases have been updated to a different format. I know there is a way you can save a directory or directories prior to an upgrade and then restore them to transfer your old settings to a new version of Tvheadend because I have the procedure to do that written down somewhere. But they don't make it easy, especially if Linux isn't your primary operating system. And going in the reverse direction might not work if any settings or databases were changed by the new version.

By way of comparison, I usually don't fear upgrading software on a Mac because if something goes wrong, or even if I just hate the new version of a program for some reason, I can easily restore the old one and its configuration from my Time Machine backup. But Linux is a whole different animal, and not particularly friendly toward those who don't use it regularly and know all its secrets. And it doesn't even have anything exactly like Time Machine, particularly if you are running a server only with no Gnome desktop or whatever. So with Linux my general philosophy is, "if it ain't broke don't fix it!" - if it weren't for the necessity of getting critical security fixes I would not even do apt updates/upgrades!

Now that you have explained the issue I know that if I do run into the problem of having more channels recording than there are available tuners, upgrading Tvheadend to the latest 4.3 version might be the way to fix it. I just hope that if I do wind up doing that it doesn't break anything! But Ubuntu 18.04 is still supported until April 2023 so I am not in any big hurry.

So far, everything appears to be working great!!! Question about instances: If I add another instance that is the same as my first instance so as to have additional tuners available, in my config.ini do I just duplicate the entire [locast_city] section and change only the username and password? I assume I'd need to tweak the instance name slightly also, for example [locast_city_2]? And if I do that will it use the channel selections I have already defined for that city's lineup? In other words what I am asking is some assurance that it won't re-enable channels that I have disabled or add duplicates of channels.

You had said it does load balancing in this case. So consider two possible scenarios:

First, you have set Tvheadend to record two programs from the SAME channel, back to back. Normally Tvheadend overlaps the recordings by padding the start and end times a little, but at least in the case of ATSC it usually uses the same tuner. I am assuming that in the case of using your program it would be the same and that Tvheadend would use the same mux and the same incoming stream for both recordings, correct? So in that case, if I am correct, only one tuner and stream would be used for both recordings, even though they overlap.

Second, you have set Tvheadend to record two programs from DIFFERENT channels from the same city, back to back. Again there will be overlap but in this case I am assuming that it will obviously need to use two tuners and two streams. But if you have two instances for the same city, I am assuming that due to load balancing it would request the first program using the login for the first instance and the second program using the login for the second instance, correct?

Well first of all I am sorry to hear about your hard drive but it was good that you had a backup!

I installed 0.8.4c and added a second instance for the city by putting a 2 after the name (so, I have [locast_city] and [locast_city2]) with different login credentials, and upped the Client -> concurrent_listeners and locast providers -> tuner counts. And what I found was what when requesting the channel list it did work as expected, in other words there were only the channels I wanted and no duplicates. But in the Channel Editor it does show all the channels twice, however for the second instance instead of the first column being red or green it is yellow, and all of those channels are checked as enabled including the ones I had previously disabled in the original instance. So, I went through and unchecked the ones I did not want in the second instance, and now they show as pink rather than red. So I assume that if the first column is yellow or pink then data for that channel is not included in the channel listings, either because it's a duplicate (yellow) or it's a duplicate and it is disabled (pink).

The sorting by column does work and was actually quite helpful in checking that I had disabled all the duplicate channels that I didn't want, but also seeing all the channels that were SD vs. HD by sorting the metadata column. It all seems to be working quite well, so thanks again!

Well I am sorry to say that things did not go well with 0.8.4e!

First of all it would no longer start after a system reboot so I had to launch it manually. But when I attempted to view the xmltv.xml file it threw an error that said:

XML Parsing Error: no root element found
Location: http://ipaddress:6077/xmltv.xml
Line Number 1, Column 1:

And in the terminal output:

$ python3 tvh_main.py
2021-05-22 04:08:27,158-WARNING:main #########################################
2021-05-22 04:08:27,159-WARNING:main MIT License, Copyright (C) 2021 ROCKY4546
Traceback (most recent call last):
  File "tvh_main.py", line 15, in <module>
    main.main(script_dir)
  File "/home/hts/locast/tvheadend-locast/lib/main.py", line 88, in main
    plugins.refresh_channels()
  File "/home/hts/locast/tvheadend-locast/lib/plugins/plugin_handler.py", line 90, in refresh_channels
    self.call_function(plugin, 'refresh_channels')
  File "/home/hts/locast/tvheadend-locast/lib/plugins/plugin_handler.py", line 104, in call_function
    call_f(*args)
  File "/home/hts/locast/tvheadend-locast/plugins/provider_video_locast/lib/locast.py", line 41, in refresh_channels
    instance.refresh_channels()
  File "/home/hts/locast/tvheadend-locast/plugins/provider_video_locast/lib/locast_instance.py", line 64, in refresh_channels
    self.channels.refresh_channels()
  File "/home/hts/locast/tvheadend-locast/plugins/provider_video_locast/lib/channels.py", line 64, in refresh_channels
    ch_dict = self.get_locast_channels()
  File "/home/hts/locast/tvheadend-locast/lib/common/decorators.py", line 44, in wrapper_func
    return f(self, *args, **kwargs)
  File "/home/hts/locast/tvheadend-locast/lib/common/decorators.py", line 29, in wrapper_func
    return f(self, *args, **kwargs)
  File "/home/hts/locast/tvheadend-locast/plugins/provider_video_locast/lib/channels.py", line 101, in get_locast_channels
    if 'videoProperties' in locast_channel['listings'][0]:
IndexError: list index out of range
hts@server:~/locast/tvheadend-locast$ ----------------------------------------
Exception happened during processing of request from ('127.0.0.1', 48964)
Traceback (most recent call last):
  File "/usr/lib/python3.7/socketserver.py", line 316, in _handle_request_noblock
    self.process_request(request, client_address)
  File "/usr/lib/python3.7/socketserver.py", line 347, in process_request
    self.finish_request(request, client_address)
  File "/usr/lib/python3.7/socketserver.py", line 360, in finish_request
    self.RequestHandlerClass(request, client_address, self)
  File "/home/hts/locast/tvheadend-locast/lib/clients/web_admin.py", line 189, in __init__
    super(CustomWebAdminHttpHandler, self).__init__(*args, **kwargs)
  File "/home/hts/locast/tvheadend-locast/lib/clients/web_admin.py", line 77, in __init__
    super().__init__(*args)
  File "/usr/lib/python3.7/socketserver.py", line 720, in __init__
    self.handle()
  File "/usr/lib/python3.7/http/server.py", line 426, in handle
    self.handle_one_request()
  File "/usr/lib/python3.7/http/server.py", line 414, in handle_one_request
    method()
  File "/home/hts/locast/tvheadend-locast/lib/clients/web_admin.py", line 89, in do_GET
    if getrequest.call_url(self, self.content_path):
  File "/home/hts/locast/tvheadend-locast/lib/common/decorators.py", line 75, in call_url
    self.url2func[_name](_tuner, *args, **kwargs)
  File "/home/hts/locast/tvheadend-locast/lib/clients/channels/channels_html.py", line 27, in get_channels_html
    html = channels_html.get()
  File "/home/hts/locast/tvheadend-locast/lib/clients/channels/channels_html.py", line 40, in get
    return ''.join([self.header, self.body])
  File "/home/hts/locast/tvheadend-locast/lib/clients/channels/channels_html.py", line 57, in header
    list(self.tab_names.keys())[0], '")</script>' ])
IndexError: list index out of range
----------------------------------------
----------------------------------------
Exception happened during processing of request from ('127.0.0.1', 48968)
Traceback (most recent call last):
  File "/usr/lib/python3.7/socketserver.py", line 316, in _handle_request_noblock
    self.process_request(request, client_address)
  File "/usr/lib/python3.7/socketserver.py", line 347, in process_request
    self.finish_request(request, client_address)
  File "/usr/lib/python3.7/socketserver.py", line 360, in finish_request
    self.RequestHandlerClass(request, client_address, self)
  File "/home/hts/locast/tvheadend-locast/lib/clients/web_admin.py", line 189, in __init__
    super(CustomWebAdminHttpHandler, self).__init__(*args, **kwargs)
  File "/home/hts/locast/tvheadend-locast/lib/clients/web_admin.py", line 77, in __init__
    super().__init__(*args)
  File "/usr/lib/python3.7/socketserver.py", line 720, in __init__
    self.handle()
  File "/usr/lib/python3.7/http/server.py", line 426, in handle
    self.handle_one_request()
  File "/usr/lib/python3.7/http/server.py", line 414, in handle_one_request
    method()
  File "/home/hts/locast/tvheadend-locast/lib/clients/web_admin.py", line 89, in do_GET
    if getrequest.call_url(self, self.content_path):
  File "/home/hts/locast/tvheadend-locast/lib/common/decorators.py", line 75, in call_url
    self.url2func[_name](_tuner, *args, **kwargs)
  File "/home/hts/locast/tvheadend-locast/lib/clients/channels/channels_html.py", line 27, in get_channels_html
    html = channels_html.get()
  File "/home/hts/locast/tvheadend-locast/lib/clients/channels/channels_html.py", line 40, in get
    return ''.join([self.header, self.body])
  File "/home/hts/locast/tvheadend-locast/lib/clients/channels/channels_html.py", line 57, in header
    list(self.tab_names.keys())[0], '")</script>' ])
IndexError: list index out of range
----------------------------------------
2021-05-22 04:09:41,890-WARNING:epg EPG Days to collect is too high.  2021-06-01 has no data
----------------------------------------
Exception happened during processing of request from ('127.0.0.1', 48998)
Traceback (most recent call last):
  File "/usr/lib/python3.7/socketserver.py", line 316, in _handle_request_noblock
    self.process_request(request, client_address)
  File "/usr/lib/python3.7/socketserver.py", line 347, in process_request
    self.finish_request(request, client_address)
  File "/usr/lib/python3.7/socketserver.py", line 360, in finish_request
    self.RequestHandlerClass(request, client_address, self)
  File "/home/hts/locast/tvheadend-locast/lib/clients/web_admin.py", line 189, in __init__
    super(CustomWebAdminHttpHandler, self).__init__(*args, **kwargs)
  File "/home/hts/locast/tvheadend-locast/lib/clients/web_admin.py", line 77, in __init__
    super().__init__(*args)
  File "/usr/lib/python3.7/socketserver.py", line 720, in __init__
    self.handle()
  File "/usr/lib/python3.7/http/server.py", line 426, in handle
    self.handle_one_request()
  File "/usr/lib/python3.7/http/server.py", line 414, in handle_one_request
    method()
  File "/home/hts/locast/tvheadend-locast/lib/clients/web_admin.py", line 89, in do_GET
    if getrequest.call_url(self, self.content_path):
  File "/home/hts/locast/tvheadend-locast/lib/common/decorators.py", line 75, in call_url
    self.url2func[_name](_tuner, *args, **kwargs)
  File "/home/hts/locast/tvheadend-locast/lib/clients/epg2xml.py", line 38, in xmltv_xml
    reply_str = epg.get_epg_xml(_tuner)
  File "/home/hts/locast/tvheadend-locast/lib/clients/epg2xml.py", line 65, in get_epg_xml
    self.plugins.refresh_epg(self.namespace, self.instance)
  File "/home/hts/locast/tvheadend-locast/lib/plugins/plugin_handler.py", line 98, in refresh_epg
    self.call_function(plugin, 'refresh_epg')
  File "/home/hts/locast/tvheadend-locast/lib/plugins/plugin_handler.py", line 104, in call_function
    call_f(*args)
  File "/home/hts/locast/tvheadend-locast/plugins/provider_video_locast/lib/locast.py", line 48, in refresh_epg
    instance.refresh_epg()
  File "/home/hts/locast/tvheadend-locast/plugins/provider_video_locast/lib/locast_instance.py", line 74, in refresh_epg
    self.epg.refresh_epg()
  File "/home/hts/locast/tvheadend-locast/plugins/provider_video_locast/lib/epg.py", line 56, in refresh_epg
    self.refresh_programs(epg_day, True)
  File "/home/hts/locast/tvheadend-locast/plugins/provider_video_locast/lib/epg.py", line 111, in refresh_programs
    for ch_data in json_data:
TypeError: 'NoneType' object is not iterable
----------------------------------------

One other thing I noticed was the lineup.xml file was not right, it was all run together like all the line breaks were missing or something.

Weird thing is that after going back to the 0.8.4c version, it started at bootup and everything seems to work except now I can't see the XML guide data there anymore. in the browser it just sits there for a few seconds and then says:

The connection was reset

The connection to the server was reset while the page was loading.

Can't explain why the problems all of a sudden but just wanted to let you know.

Well I tried the 0.8.4f version, specifically I expanded the zip file, renamed the resulting folder to tvheadend-locast after copying in my config.ini and the data directory from the previous version, although I then removed all the db files except channels.db as you suggested. I then attempted to do

sudo systemctl restart locast.service

python3 tvh_main.py

$ python3 tvh_main.py
2021-05-23 00:51:20,508-WARNING:main #########################################
2021-05-23 00:51:20,509-WARNING:main MIT License, Copyright (C) 2021 ROCKY4546
2021-05-23 00:51:33,077-WARNING:epg EPG Days to collect is too high.  2021-06-01 has no data
2021-05-23 00:51:33,855-WARNING:epg EPG Days to collect is too high.  2021-06-02 has no data
2021-05-23 00:51:39,667-WARNING:epg EPG Days to collect is too high.  2021-06-01 has no data
2021-05-23 00:51:39,781-WARNING:epg EPG Days to collect is too high.  2021-06-02 has no data
2021-05-23 00:51:45,938-WARNING:epg EPG Days to collect is too high.  2021-06-01 has no data
2021-05-23 00:51:46,588-WARNING:epg EPG Days to collect is too high.  2021-06-02 has no data
2021-05-23 00:51:53,498-WARNING:epg EPG Days to collect is too high.  2021-06-01 has no data
2021-05-23 00:51:53,599-WARNING:epg EPG Days to collect is too high.  2021-06-02 has no data
2021-05-23 00:52:04,675-WARNING:epg EPG Days to collect is too high.  2021-06-02 has no data
Traceback (most recent call last):
  File "tvh_main.py", line 15, in <module>
    main.main(script_dir)
  File "/home/hts/locast/tvheadend-locast/lib/main.py", line 111, in main
    plugins.refresh_epg()
  File "/home/hts/locast/tvheadend-locast/lib/plugins/plugin_handler.py", line 98, in refresh_epg
    self.call_function(plugin, 'refresh_epg')
  File "/home/hts/locast/tvheadend-locast/lib/plugins/plugin_handler.py", line 104, in call_function
    call_f(*args)
  File "/home/hts/locast/tvheadend-locast/plugins/provider_video_locast/lib/locast.py", line 48, in refresh_epg
    instance.refresh_epg()
  File "/home/hts/locast/tvheadend-locast/plugins/provider_video_locast/lib/locast_instance.py", line 74, in refresh_epg
    self.epg.refresh_epg()
  File "/home/hts/locast/tvheadend-locast/plugins/provider_video_locast/lib/epg.py", line 56, in refresh_epg
    self.refresh_programs(epg_day, True)
  File "/home/hts/locast/tvheadend-locast/plugins/provider_video_locast/lib/epg.py", line 118, in refresh_programs
    self.db.save_program_list(self.locast_instance.locast.name, self.instance, _day, program_list)
  File "/home/hts/locast/tvheadend-locast/lib/db/db_epg.py", line 82, in save_program_list
    json.dumps(_prog_list)))
  File "/home/hts/locast/tvheadend-locast/lib/db/db.py", line 64, in add
    DB.conn[self.db_name][threading.get_ident()].commit()
sqlite3.OperationalError: database is locked

BUT even so I could once again access the xmltv.xml file in my browser, so that is working again. And it recreated the missing databases, so I used control-C to exit and then ran

sudo systemctl restart locast.service

sudo systemctl restart locast.service

May 23 01:33:36 server systemd[1]: locast.service: Service RestartSec=54s expired, scheduling restart.
May 23 01:33:36 server systemd[1]: locast.service: Scheduled restart job, restart counter is at 50.
May 23 01:33:36 server systemd[1]: Stopped Locast Service.
May 23 01:33:36 server systemd[1]: Started Locast Service.
May 23 01:33:37 server locast[26630]: ERROR: Config file missing None, Exiting...
May 23 01:33:37 server systemd[1]: locast.service: Main process exited, code=exited, status=1/FAILURE
May 23 01:33:37 server systemd[1]: locast.service: Failed with result 'exit-code'.

This confuses me because there is a config.ini file in the same directory as tvh_main.py and locast certainly appears to be finding in and using it, so why is it filling this log file with these messages, and how do I make it stop? Also these message appear to be duplicated to /var/log/syslog. After a reboot I see a similar issue, it says this:

May 23 01:44:11 server locast[350]: ERROR: Config file missing None, Exiting...
May 23 01:44:11 server systemd[1]: locast.service: Main process exited, code=exited, status=1/FAILURE
May 23 01:44:11 server systemd[1]: locast.service: Failed with result 'exit-code'.

If I start over from scratch (fresh install of 0.8.4f with the ONLY thing copied over being the config.ini file) then it still fails after a reboot but the log messages are slightly different:

May 23 02:09:54 server systemd[1]: locast.service: Service RestartSec=54s expired, scheduling restart.
May 23 02:09:54 server systemd[1]: locast.service: Scheduled restart job, restart counter is at 2.
May 23 02:09:54 server systemd[1]: Stopped Locast Service.
May 23 02:09:54 server systemd[1]: Started Locast Service.
May 23 02:09:55 server locast[1400]: ERROR: Config file missing None, Exiting...
May 23 02:09:55 server systemd[1]: locast.service: Main process exited, code=exited, status=1/FAILURE
May 23 02:09:55 server systemd[1]: locast.service: Failed with result 'exit-code'.

If I then go into the tvheadend-locast directory and run

python3 tvh_main.py

Basically all these issues seem to come from trying to run it as a service. I finally got fed up and did

sudo systemctl disable locast.service

sudo systemctl stop locast.service

@reboot cd /home/hts/locast/tvheadend-locast/ && /usr/bin/python3 /home/hts/locast/tvheadend-locast/tvh_main.py &

Note that if I don't add that cd /home/hts/locast/tvheadend-locast/ at the beginning then it gives me the same error about the config file missing, so that may be a clue as to why that error keeps happening! I should mentioned that I had copied the config file to /home/hts/locast/ at one point to see if maybe it was looking for it there, but that didn't help. Anyway I hope this is not a tough nut to crack, but anyway it is running, just as a background task started at bootup and not as a service, which I suppose to real Linux devotees is some kind of sin but for now at least it works!

Thanks again! I am not sure if your first comment was intended to be a compliment or a polite way of saying I'm being a pest, but for whatever reason I have sometimes been that one guy that would quite accidentally uncover a bug that seems to affect no one else in the universe. Many years ago I was testing an online communications service and I found a bug that turned out to be a hash collision, whatever that is, but I was told that there was a less than one in a million chance that someone would trigger it, and yet I did.

I did have the number of days set to collect to 11 but changed it to 10 and still get errors when I start it manually:

$ python3 tvh_main.py
2021-05-24 07:35:16,529-WARNING:main #########################################
2021-05-24 07:35:16,530-WARNING:main MIT License, Copyright (C) 2021 ROCKY4546
2021-05-24 07:35:19,845-WARNING:epg EPG Days to collect is too high.  2021-06-01 has no data
2021-05-24 07:35:19,965-WARNING:epg EPG Days to collect is too high.  2021-06-02 has no data

Still no luck in starting it as a service but the errors in the log are different:

May 24 07:32:31 server systemd[1]: locast.service: Service RestartSec=54s expired, scheduling restart.
May 24 07:32:31 server systemd[1]: locast.service: Scheduled restart job, restart counter is at 1.
May 24 07:32:31 server systemd[1]: Stopped Locast Service.
May 24 07:32:31 server systemd[1]: Started Locast Service.
May 24 07:32:32 server locast[1305]: Traceback (most recent call last):
May 24 07:32:32 server locast[1305]:   File "/home/hts/locast/tvheadend-locast/tvh_main.py", line 21, in <module>
May 24 07:32:32 server locast[1305]:     main.main(script_dir)
May 24 07:32:32 server locast[1305]:   File "/home/hts/locast/tvheadend-locast/lib/main.py", line 75, in main
May 24 07:32:32 server locast[1305]:     config_obj = user_config.get_config(script_dir, opersystem, args)
May 24 07:32:32 server locast[1305]:   File "/home/hts/locast/tvheadend-locast/lib/config/user_config.py", line 36, in get_config
May 24 07:32:32 server locast[1305]:     return TVHUserConfig(script_dir, opersystem, args)
May 24 07:32:32 server locast[1305]:   File "/home/hts/locast/tvheadend-locast/lib/config/user_config.py", line 62, in __init__
May 24 07:32:32 server locast[1305]:     self.import_config(config_file)
May 24 07:32:32 server locast[1305]:   File "/home/hts/locast/tvheadend-locast/lib/config/user_config.py", line 100, in import_config
May 24 07:32:32 server locast[1305]:     utils.logging_setup(self.data['paths'])
May 24 07:32:32 server locast[1305]:   File "/home/hts/locast/tvheadend-locast/lib/common/utils.py", line 51, in logging_setup
May 24 07:32:32 server locast[1305]:     os.makedirs('data/logs')
May 24 07:32:32 server locast[1305]:   File "/usr/lib/python3.7/os.py", line 211, in makedirs
May 24 07:32:32 server locast[1305]:     makedirs(head, exist_ok=exist_ok)
May 24 07:32:32 server locast[1305]:   File "/usr/lib/python3.7/os.py", line 221, in makedirs
May 24 07:32:32 server locast[1305]:     mkdir(name, mode)
May 24 07:32:32 server locast[1305]: PermissionError: [Errno 13] Permission denied: 'data'
May 24 07:32:32 server systemd[1]: locast.service: Main process exited, code=exited, status=1/FAILURE
May 24 07:32:32 server systemd[1]: locast.service: Failed with result 'exit-code'.

But, still can start it using the crontab entry. So I hope that means something to you. I don't see anything weird about the permission settings on the data directory but maybe that is not the data being referred to in that error?

One more issue, for some reason today the channel editor insists on re-enabling many channels that I had disabled!!

G Kazaroth wrote:

OK, I was right. It needed to create a data/logs folder and could not due to the permissions on the data folder. The data folder in general needs to be write accessible by the user of the app. Either change the owner/group or change the permissions and that should fix the issue.

Okay, this is what I have tried. I completely blew away my previous tvheadend-locast folder and then expanded the 0.8.4g.zip file to create the folder inside, which I then renamed to tvheadend-locast. I copied over the config-ini file but removed everything other than the configurations for the instances, so it would regenerate everything including the data directory. I had to redo my channel edits, but then I enabled it as a service again and after a reboot it still would not start:

May 24 10:12:16 server locast[354]: Traceback (most recent call last):
May 24 10:12:16 server locast[354]:   File "/home/hts/locast/tvheadend-locast/tvh_main.py", line 21, in <module>
May 24 10:12:16 server locast[354]:     main.main(script_dir)
May 24 10:12:16 server locast[354]:   File "/home/hts/locast/tvheadend-locast/lib/main.py", line 75, in main
May 24 10:12:16 server locast[354]:     config_obj = user_config.get_config(script_dir, opersystem, args)
May 24 10:12:16 server locast[354]:   File "/home/hts/locast/tvheadend-locast/lib/config/user_config.py", line 36, in get_config
May 24 10:12:16 server locast[354]:     return TVHUserConfig(script_dir, opersystem, args)
May 24 10:12:16 server locast[354]:   File "/home/hts/locast/tvheadend-locast/lib/config/user_config.py", line 62, in __init__
May 24 10:12:16 server locast[354]:     self.import_config(config_file)
May 24 10:12:16 server locast[354]:   File "/home/hts/locast/tvheadend-locast/lib/config/user_config.py", line 100, in import_config
May 24 10:12:16 server locast[354]:     utils.logging_setup(self.data['paths'])
May 24 10:12:16 server locast[354]:   File "/home/hts/locast/tvheadend-locast/lib/common/utils.py", line 51, in logging_setup
May 24 10:12:16 server locast[354]:     os.makedirs('data/logs')
May 24 10:12:16 server locast[354]:   File "/usr/lib/python3.7/os.py", line 211, in makedirs
May 24 10:12:16 server locast[354]:     makedirs(head, exist_ok=exist_ok)
May 24 10:12:16 server locast[354]:   File "/usr/lib/python3.7/os.py", line 221, in makedirs
May 24 10:12:16 server locast[354]:     mkdir(name, mode)
May 24 10:12:16 server locast[354]: PermissionError: [Errno 13] Permission denied: 'data'
May 24 10:12:16 server systemd[1]: locast.service: Main process exited, code=exited, status=1/FAILURE
May 24 10:12:16 server systemd[1]: locast.service: Failed with result 'exit-code'.

But nevertheless it keeps retrying and fills up the log with these:

May 24 10:14:58 server systemd[1]: locast.service: Service RestartSec=54s expired, scheduling restart.
May 24 10:14:58 server systemd[1]: locast.service: Scheduled restart job, restart counter is at 3.
May 24 10:14:58 server systemd[1]: Stopped Locast Service.
May 24 10:14:58 server systemd[1]: Started Locast Service.
May 24 10:14:59 server locast[1438]: Traceback (most recent call last):
May 24 10:14:59 server locast[1438]:   File "/home/hts/locast/tvheadend-locast/tvh_main.py", line 21, in <module>
May 24 10:14:59 server locast[1438]:     main.main(script_dir)
May 24 10:14:59 server locast[1438]:   File "/home/hts/locast/tvheadend-locast/lib/main.py", line 75, in main
May 24 10:14:59 server locast[1438]:     config_obj = user_config.get_config(script_dir, opersystem, args)
May 24 10:14:59 server locast[1438]:   File "/home/hts/locast/tvheadend-locast/lib/config/user_config.py", line 36, in get_config
May 24 10:14:59 server locast[1438]:     return TVHUserConfig(script_dir, opersystem, args)
May 24 10:14:59 server locast[1438]:   File "/home/hts/locast/tvheadend-locast/lib/config/user_config.py", line 62, in __init__
May 24 10:14:59 server locast[1438]:     self.import_config(config_file)
May 24 10:14:59 server locast[1438]:   File "/home/hts/locast/tvheadend-locast/lib/config/user_config.py", line 100, in import_config
May 24 10:14:59 server locast[1438]:     utils.logging_setup(self.data['paths'])
May 24 10:14:59 server locast[1438]:   File "/home/hts/locast/tvheadend-locast/lib/common/utils.py", line 51, in logging_setup
May 24 10:14:59 server locast[1438]:     os.makedirs('data/logs')
May 24 10:14:59 server locast[1438]:   File "/usr/lib/python3.7/os.py", line 211, in makedirs
May 24 10:14:59 server locast[1438]:     makedirs(head, exist_ok=exist_ok)
May 24 10:14:59 server locast[1438]:   File "/usr/lib/python3.7/os.py", line 221, in makedirs
May 24 10:14:59 server locast[1438]:     mkdir(name, mode)
May 24 10:14:59 server locast[1438]: PermissionError: [Errno 13] Permission denied: 'data'
May 24 10:14:59 server systemd[1]: locast.service: Main process exited, code=exited, status=1/FAILURE
May 24 10:14:59 server systemd[1]: locast.service: Failed with result 'exit-code'.

I have no idea what directory may be causing the issue, especially since the data directory was completely regenerated. Starting it from a cron job still works and then it did keep my channel edits after a reboot so no idea why I had problems with that. If I do ls -l you can see there is nothing weird about the permissions or ownership:

$ ls -l /home/hts/locast/tvheadend-locast
total 88
drwxr-xr-x  4 hts hts 4096 May 23 09:50 build
-rwxr-xr-x  1 hts hts 4146 May 23 09:50 CHANGELOG.md
-rwxr-xr-x  1 hts hts 3485 May 24 10:19 config.ini
-rwxr-xr-x  1 hts hts 1554 May 23 09:50 CONTRIBUTING.md
drwxr-xr-x  5 hts hts 4096 May 24 09:15 data
-rwxr-xr-x  1 hts hts  182 May 23 09:50 docker-compose.yml
-rwxr-xr-x  1 hts hts  321 May 23 09:50 Dockerfile
-rwxr-xr-x  1 hts hts  304 May 23 09:50 Dockerfile_l2p
-rwxr-xr-x  1 hts hts  304 May 23 09:50 Dockerfile_tvh
-rwxr-xr-x  1 hts hts  518 May 23 09:50 Dockerfile_tvh_crypt.alpine
-rwxr-xr-x  1 hts hts  849 May 23 09:50 Dockerfile_tvh_crypt.slim-buster
drwxr-xr-x 14 hts hts 4096 May 24 09:15 lib
-rwxr-xr-x  1 hts hts 1095 May 23 09:50 LICENSE
drwxr-xr-x  4 hts hts 4096 May 23 09:50 misc
drwxr-xr-x  4 hts hts 4096 May 24 09:15 plugins
-rwxr-xr-x  1 hts hts 9982 May 23 09:50 README.md
-rw-r--r--  1 hts hts   22 May 23 09:50 requirements.txt
-rwxr-xr-x  1 hts hts 2996 May 23 09:50 TVHEADEND.md
-rwxr-xr-x  1 hts hts  824 May 23 09:50 tvh_main.py

$ ls -l /home/hts/locast/tvheadend-locast/data
total 12
drwxr-xr-x 3 hts hts 4096 May 24 09:15 cache
drwxr-xr-x 2 hts hts 4096 May 24 10:19 db
drwxr-xr-x 2 hts hts 4096 May 24 09:15 logs

$ ls -l /home/hts/locast/tvheadend-locast/data/logs
total 0
-rw-r--r-- 1 hts hts 0 May 24 09:15 cabernet.log

I wish I knew what "data" it is unable to access when trying to run as a service. Would it make any difference what the permissions are on the service file itself? Right now the owner is set to hts but it is world executable:

$ ls -l /lib/systemd/system/locast*
-rwxr-xr-x 1 hts hts 954 May 23 09:50 /lib/systemd/system/locast.service

I know previously I had set the user/group to root/root but I copied it back over again today as part of trying to refresh everything, and leaving it as hts/hts doesn't seem to make any difference.

I am totally stumped, does any of this make sense to you?

Version 0.8.4j appears to be working great, starts as a service and does not leave error messages in the log file! I very much appreciate all your hard work on this, thanks again!

I was thinking about giving this a try but I have some questions.

My Tvheadend server runs on Ugoos X3 Cube running CoreELEC 19. Will this be usable in CoreELEC?

I currently use Tvheadend for my local OTA channels. But I live in a fringe reception area and sometimes the signal strength drops too low. OTA would be my preferred method of watching/recording with Locast as a backup.

How would the redundant channel numbers be dealt with in the guide?

Interesting!
I'll have to read up on how to install it on CoreELEC. I may be able to stumble through it.
Thanks

Any suggestion on which Entware Python3 package I should install? I'd like to keep the installation light but I don't want cause problems by leaving something out that is needed. I was thinking the Python3-light package.

OK, so I really don't know what I'm doing but I'm trying.

I installed Entware and Python3-light.
I downloaded the source file tvheadend-locast-0.8.5.zip and unzipped it to /storage/tvheadend-locast.
I edited config_example.ini with my Locast username and password and saved it as config.ini in the /storage/tvheadend-locast folder.

When I issue the "python3 tvh_main.py" command (I've SSH'd in using Putty) from the tvheadend-locast folder I get the following error:

CoreELEC:~/tvheadend-locast # python3 tvh_main.py
Traceback (most recent call last):
File "/storage/tvheadend-locast/tvh_main.py", line 10, in <module>
from lib import main
File "/storage/tvheadend-locast/lib/main.py", line 20, in <module>
import logging
ModuleNotFoundError: No module named 'logging'

Did I do the folder extraction and location placement correctly?

I may have got it installed correctly or I may not have. If I had to do it over again I would have installed the complete Python3 version, there were a lot of missing modules.
I was getting errors when I tried to install the '_ssl' module so I tried it without the '_'. It seemed like it tried to reinstall Python3. However, I no longer got any missing module errors when I ran the 'python3 tvh_main.py' command.

But I did get the errors below:
CoreELEC:~/tvheadend-locast # python3 tvh_main.py
Unable to load pip module to install upgrades
Unable to load cryptography module, will not encrypt passwords
2021-06-05 07:02:02,177-WARNING:main #########################################
2021-06-05 07:02:02,177-WARNING:main MIT License, Copyright (C) 2021 ROCKY4546
2021-06-05 07:02:02,289-ERROR:authenticate locast Unable to login due to invalid logins. Clear config entry login_invalid to try again

Do I keep the '< >' around the username and password in the config.ini file?

When I enter URL http://[IP]:6077/ I get to the cabernet page and can get to the XML/JSON links.

Should I try to integrate into Tvheadend or do I have some problems I need to resolve first?

Thanks

I'm just a user like yourself but the issue with the cryptography module is discussed earlier in this thread, but basically you don't really need it, it just adds extra security if you have it. It is completely optional. See https://tvheadend.org/boards/5/topics/43052?r=44150#message-44150

Don't know about the pip module thing in CoreELEC but did you try opkg install python3-pip ? May not be anything that obvious but worth a shot, although I have no idea how much space it takes up. But you don't need it if you have no interest in the cryptography. If you do install pip, then you should be able to do pip3 install cryptography to get the cryptography module.

You do not keep the angle brackets. For the username you use the email address that you log into the Locast site with, and for the password you use the password. But no angle brackets. A previous message in this thread mentions how to clear login_invalid (see next to last paragraph at https://tvheadend.org/boards/5/topics/43052?r=44145#message-44145), but it's just an entry in your config.ini file that you have to delete AFTER fixing your username and password. Once you get all that fixed you should be good to go.

Hope this helps. I suggest scrolling through the previous posts, right now this thread is only five pages long so it shouldn't take that long and it may help answer your current and even possibly some future questions.

I saw the cryptography comment but I didn't realize that was associated with pip. I don't need cryptography so I won't worry about not being able to install pip module.

I got my login fixed, thanks.

I guess I'm on to integrating into Tvheadend.

I have some questions about setting up the grabber.
How do I use the tv_grab_url file? Will it work on my CoreELEC box? Do I just copy and paste the text into a file? Where do I put the file and does it matter what I name it?

I already use Zap2EPG for my OTA EPG data. Should I just continue to use that for Locast also since the data is the same?

Thanks

G Kazaroth wrote:

Next the tv_grab_url is located where all the tv_grab* files are located. If you go to the tvheadend > Configuration > Channel/EPG > EPG Grabber Modules, The PATH variable normally shows where the files are located. This is normally found in the /usr/bin directory. After you place the file in the folder, restart tvheadend to pick up the file.

To be a bit more clear about that, if you go to tvheadend > Configuration > Channel/EPG > EPG Grabber Modules, in the left hand pane you will see a list of EPG Grabbers, all of which should be disabled (have an orange dot next to them) except the ones that are enabled (those will have a green dot with a check mark ✔ inside next to them). On my system, the one I have enabled is called "Internal: XMLTV: tv_grab_file is a simple grabber that just read the ~/.xmltv/tv_grab_file.xmltv file" and the path is /usr/bin. That just means that there is a script called tv_grab_file in the /usr/bin directory, but that script is looking for a file at ~/.xmltv/tv_grab_file.xmltv. In this case the ~ refers to tvheadend's user, not your user, which is normally the hts user unless you have changed it. So in reality it's looking for a file called /home/hts/.xmltv/tv_grab_file.xmltv. So what you have to do is copy or move the xmltv.xml file to that location and then run the grabber. For example, one way to do that would be this (the first line assumes that Tvheadend runs as the hts user):

sudo su hts
cd /home/hts/.xmltv
wget -q --retry-connrefused --tries=5 --timeout=180 http://server_ip_address:6077/xmltv.xml
mv -f xmltv.xml tv_grab_file.xmltv

Replace server_ip_address so the URL is correct. After doing all that then you have to run the internal grabber in Tvheadend - you can do that by going to tvheadend > Configuration > Channel/EPG > EPG Grabber and click the button that says "Re-run Internal EPG Grabbers".

If that works then you can automate the process by putting the above commands in a shell script (you may need to specify full paths to the commands and files in that case) but instead of putting sudo su hts in the script itself, you'd instead run the script as the hts user. What I'd do is become the hts user (using sudo su hts) and then use crontab -e to create a cron job that would run as the hts user, and have that run the script at whatever time you like. Then I'd set up Tvheadend to grab the file, by going to tvheadend > Configuration > Channel/EPG > EPG Grabber and in the "Internal Grabber" section, in the box that says "Cron multi-line" set the time for it to run the grabber, which would be maybe 15 minutes or a half hour after you run wget (to give it time for the download to complete), using the same time format you would use when creating a Cron job.

But that is just one way to do it, and maybe not the best way. There is another method that uses sockets. Quite some time back I found an article that discusses both methods at https://freetoairamerica.wordpress.com/2014/12/03/some-hints-for-getting-free-to-air-satellite-channels-into-the-electronic-program-guide-in-kodi-or-xbmc-or-another-frontend/ (that article primarily discusses the use of zap2xml) but since I already had it set up to use the grabber file I never bothered with the sockets method.

Tvheadend has a lot of moving parts, so to speak, and this is one area where it can get really confusing if you don't understand how all the parts work together. There may be an easier way to do this but if so I'm not aware of it.

EDIT: You may wonder why I'm not using the grabber mentioned in item 2 of the first post in this thread. It's partly because I'm already familiar with using tv_grab_file, but also because the instructions for using that one just aren't complete enough. For one thing the screenshot is so small as to be unreadable, but for another that grabber doesn't just automatically show up in Tvheadend's list of grabbers. I am guessing that at a minimum you'd need to copy that grabber file to the /usr/bin directory (and make sure it is executable and owned by the root user) and then restart Tvheadend before Tvheadend would even find it. But I am not certain of that. If that's all that is required then it could simplify things, I just wish that screenshot were about two or three times larger, so you could actually read it. But I tend to stick to what I am familiar with.

Sean Micklem wrote:

G Kazaroth wrote:

Next the tv_grab_url is located where all the tv_grab* files are located. If you go to the tvheadend > Configuration > Channel/EPG > EPG Grabber Modules, The PATH variable normally shows where the files are located. This is normally found in the /usr/bin directory. After you place the file in the folder, restart tvheadend to pick up the file.

To be a bit more clear about that, if you go to tvheadend > Configuration > Channel/EPG > EPG Grabber Modules, in the left hand pane you will see a list of EPG Grabbers, all of which should be disabled (have an orange dot next to them) except the ones that are enabled (those will have a green dot with a check mark ✔ inside next to them). On my system, the one I have enabled is called "Internal: XMLTV: tv_grab_file is a simple grabber that just read the ~/.xmltv/tv_grab_file.xmltv file" and the path is /usr/bin. That just means that there is a script called tv_grab_file in the /usr/bin directory, but that script is looking for a file at ~/.xmltv/tv_grab_file.xmltv. In this case the ~ refers to tvheadend's user, not your user, which is normally the hts user unless you have changed it. So in reality it's looking for a file called /home/hts/.xmltv/tv_grab_file.xmltv. So what you have to do is copy or move the xmltv.xml file to that location and then run the grabber. For example, one way to do that would be this (the first line assumes that Tvheadend runs as the hts user):

sudo su hts
cd /home/hts/.xmltv
wget -q --retry-connrefused --tries=5 --timeout=180 http://server_ip_address:6077/xmltv.xml
mv -f xmltv.xml tv_grab_file.xmltv

Replace server_ip_address so the URL is correct. After doing all that then you have to run the internal grabber in Tvheadend - you can do that by going to tvheadend > Configuration > Channel/EPG > EPG Grabber and click the button that says "Re-run Internal EPG Grabbers".

If that works then you can automate the process by putting the above commands in a shell script (you may need to specify full paths to the commands and files in that case) but instead of putting sudo su hts in the script itself, you'd instead run the script as the hts user. What I'd do is become the hts user (using sudo su hts) and then use crontab -e to create a cron job that would run as the hts user, and have that run the script at whatever time you like. Then I'd set up Tvheadend to grab the file, by going to tvheadend > Configuration > Channel/EPG > EPG Grabber and in the "Internal Grabber" section, in the box that says "Cron multi-line" set the time for it to run the grabber, which would be maybe 15 minutes or a half hour after you run wget (to give it time for the download to complete), using the same time format you would use when creating a Cron job.

But that is just one way to do it, and maybe not the best way. There is another method that uses sockets. Quite some time back I found an article that discusses both methods at https://freetoairamerica.wordpress.com/2014/12/03/some-hints-for-getting-free-to-air-satellite-channels-into-the-electronic-program-guide-in-kodi-or-xbmc-or-another-frontend/ (that article primarily discusses the use of zap2xml) but since I already had it set up to use the grabber file I never bothered with the sockets method.

Tvheadend has a lot of moving parts, so to speak, and this is one area where it can get really confusing if you don't understand how all the parts work together. There may be an easier way to do this but if so I'm not aware of it.

EDIT: You may wonder why I'm not using the grabber mentioned in item 2 of the first post in this thread. It's partly because I'm already familiar with using tv_grab_file, but also because the instructions for using that one just aren't complete enough. For one thing the screenshot is so small as to be unreadable, but for another that grabber doesn't just automatically show up in Tvheadend's list of grabbers. I am guessing that at a minimum you'd need to copy that grabber file to the /usr/bin directory (and make sure it is executable and owned by the root user) and then restart Tvheadend before Tvheadend would even find it. But I am not certain of that. If that's all that is required then it could simplify things, I just wish that screenshot were about two or three times larger, so you could actually read it. But I tend to stick to what I am familiar with.

Thanks, but this sounds a little beyond my skillset at this point. But, I agree, it would be better if you could see the example image.

I've made some progress. The Locast channels show up in my Kodi Tvheadend client and they will play. However, the channel numbers start at 10000 and increase sequentially from there but not in actual channel number order. Are they supposed to be that way? The correct channel number does show up in the description.

But the EPG is not populating. I copied the text of the tv_grab_url file into a text file called tv_grab_url and saved it in the same location as the other tv_grab* files. I've enabled it in the EPG Grabber Modules tab, left the Priority as 3 (Zap2EPG has a Priority of 5), put http://[HOST]/xmltv.xml in the Extra arguments field, and re-ran the internal grabbers. It looks like there may be some extra characters in the example screenshot but I can't read them. And no Locast channels are showing up in the EPG Grabber Channels tab as far as I can tell. Locast channels do show up in the main channels tab.

Also, what would be the best way to automatically start Tvheadend-locast when I reboot CoreELEC? I'm assuming I can put the command in an autostart.sh file (any hints on the syntax?) but don't know if that is the proper way. I saw the locast.service file but I don't know if I'm knowledgeable enough to use it or if it will work on CoreELEC.

Thanks