PrimeNetEncoder - HDHomeRun Prime Network Encoder

[El Jeffe']

@Waylo - I've done some additional troubleshooting of my "bad" music channels. I attempted to tune one of them through the SageTV interface and then I immediately logged into the HDHomerun PRIME being used to tune the channel. According to the PRIME's System log, it receives subscription requests for the bad channels from SageTV and it responds by streaming data to my SageTV IP address via the appropriate port number. The "Tuner Status" portion of the HDHomerun interface reflects 97% signal strength and 100% for both Signal Quality and Signal Strength. So, things seem to look okay from a HDHomeRun perspective.

Like you, I think that a few missing music channels is a pretty small inconvenience in the scheme of things. It would be nice to get these channels working at some point though.

[Sparhawk6]

Well, I gave yours a shot, hoping that it would fix some of my choppiness and pixelation I experience in SageDCT. However, I never could get it to work. I kept getting the "-10 no tuner" error. I swear I set the firewall open to all the ports you identified. I even tried disabling the firewall completely. No dice.

I had to scramble back to SageDCT before the wife found out I tried to tinker.

[nyplayer]

Originally Posted by Sparhawk6

Well, I gave yours a shot, hoping that it would fix some of my choppiness and pixelation I experience in SageDCT. However, I never could get it to work. I kept getting the "-10 no tuner" error. I swear I set the firewall open to all the ports you identified. I even tried disabling the firewall completely. No dice.

I had to scramble back to SageDCT before the wife found out I tried to tinker.

3 Things.

1. Did you disable SageDCT service?
2. Did you try it as a console?
3. If you ran it as a runnable somethings there is a delay finding the tuners.
http://forums.sagetv.com/forums/show...&postcount=126
http://forums.sagetv.com/forums/show...&postcount=129

[nyplayer]

For all you SageDCT users you really do not have much to do if you want to convert. You can use the same properties entries you use in SAGEDCT all you have to do is change the ports in the prime encoder.
Just change in prime.properties.
tuner0.port=7000 to tuner0.port=6969 or whatever it is do the same to the other tuners.

That way if you have to revert you can just shutdown Primeencoder and re-enable and restart SAGEDCT.

As a matter of fact your tuners will be already setup with correct epg.

[RangerZ]

My "Tuner Collection" grew to over 500 this morning. I have learned a few things however. If one closes SageTV and then restarts it all but 1 set of PrimeNetEncoder tuners will disappear.

As soon as the first channel change is performed on one of the PrimeNetEncoder tuners a new set of all (3) tuners will be added to the tuner list. For each additional channel change in the logs, another set of tuners is added. They will be added (inserted) directly below the existing tuner of the same number.

As yet I have not had an event where the recording continues on the same file, so I am not sure that the issue is file creation related or channel change related, and assume it to be channel change related.

Can someone please check their "Video Sources" for extra tuners, restart SageTV, do some channel changes, and report back on what happens with tuner count.

If you see tuners increasing, please include your sage.properties and PrimeNetEncoder.properties.

I believe that has something to do with the prime sage properties files ... I had made some changes in mine and I do do get duplicate tuners.

I am working through a sage.properties compare and will post on this later. Is the "do do" suppose to be a "do not"?

RangerZ

[nyplayer]

Originally Posted by RangerZ

My "Tuner Collection" grew to over 500 this morning. I have learned a few things however. If one closes SageTV and then restarts it all but 1 set of PrimeNetEncoder tuners will disappear.

As soon as the first channel change is performed on one of the PrimeNetEncoder tuners a new set of all (3) tuners will be added to the tuner list. For each additional channel change in the logs, another set of tuners is added. They will be added (inserted) directly below the existing tuner of the same number.

As yet I have not had an event where the recording continues on the same file, so I am not sure that the issue is file creation related or channel change related, and assume it to be channel change related.

Can someone please check their "Video Sources" for extra tuners, restart SageTV, do some channel changes, and report back on what happens with tuner count.

If you see tuners increasing, please include your sage.properties and PrimeNetEncoder.properties.


I am working through a sage.properties compare and will post on this later. Is the "do do" suppose to be a "do not"?

RangerZ

I already addressed the multiple tuners here.

http://forums.sagetv.com/forums/show...&postcount=176

[waylo]

Originally Posted by El Jeffe'

@Waylo - I've done some additional troubleshooting of my "bad" music channels. I attempted to tune one of them through the SageTV interface and then I immediately logged into the HDHomerun PRIME being used to tune the channel. According to the PRIME's System log, it receives subscription requests for the bad channels from SageTV and it responds by streaming data to my SageTV IP address via the appropriate port number. The "Tuner Status" portion of the HDHomerun interface reflects 97% signal strength and 100% for both Signal Quality and Signal Strength. So, things seem to look okay from a HDHomeRun perspective.

Like you, I think that a few missing music channels is a pretty small inconvenience in the scheme of things. It would be nice to get these channels working at some point though.

Just remembered that with SageDCT, I was using the DUMP setting rather than ffmpeg, so that may be a crucial difference. Could never get SAGEDCT's older ffmpeg to play nice with my clients (on the server I think it was passable), but I probably never looked specifically at the music stations.

[RangerZ]

Yes, you did address the issue, and I had read your post, but you also indicated you did not know what the specific fix was. As indicated above I am working through a comparison to see if I can specifically identify the parameter(s) that will solve the issue.

Please confirm that my question on the typo in your post 176.

The attached spreadsheet compares your file to mine and the default. Mine and the default have the same parameters with some minor value variations. Your file has some additional parameters (5) and there are 2 parameters removed.

I have color coded the entries as follows

• Light green is a direct match
• Dark green is a setting that I believe to only effect the video quality, and probably not relevant to the issue
• Yellow - Parameters that exist in only one or the other file
• White - Not smart enough to know the impact

Yellow nyplayer
mmc/encoders/1012343954/audio_capture_device_index=-1
mmc/encoders/1012343954/audio_capture_device_num=0
mmc/encoders/1012343954/audio_processor=
mmc/encoders/1012343954/dshow_tv_type=
mmc/encoders/1012343954/video_processor=

Yellow default
mmc/encoders/1012343954/1/0/broadcast_standard=
mmc/encoders/1012343954/fast_network_encoder_switch=false

The first dozen or so lines in each section of your file have part of the parameter string changed. A 1 is changed to 100.

I already have tested changing the IP address and the "delay_to_wait_after_tuning" with no change to the tuner behavior.

I am working through the balance of the parameters one at a time in hopes to be able to document the specific fix. I do not expect to complete this today.

I am not sure if this analysis will help you recall the specific fix.

My post was more a request to identify the magnitude of the issue. I do not think most people would check to see their tuner count unless asked.

[nyplayer]

Originally Posted by RangerZ

Yes, you did address the issue, and I had read your post, but you also indicated you did not know what the specific fix was. As indicated above I am working through a comparison to see if I can specifically identify the parameter(s) that will solve the issue.

Please confirm that my question on the typo in your post 176.

The attached spreadsheet compares your file to mine and the default. Mine and the default have the same parameters with some minor value variations. Your file has some additional parameters (5) and there are 2 parameters removed.

I have color coded the entries as follows

• Light green is a direct match
• Dark green is a setting that I believe to only effect the video quality, and probably not relevant to the issue
• Yellow - Parameters that exist in only one or the other file
• White - Not smart enough to know the impact

Yellow nyplayer
mmc/encoders/1012343954/audio_capture_device_index=-1
mmc/encoders/1012343954/audio_capture_device_num=0
mmc/encoders/1012343954/audio_processor=
mmc/encoders/1012343954/dshow_tv_type=
mmc/encoders/1012343954/video_processor=

Yellow default
mmc/encoders/1012343954/1/0/broadcast_standard=
mmc/encoders/1012343954/fast_network_encoder_switch=false

The first dozen or so lines in each section of your file have part of the parameter string changed. A 1 is changed to 100.

I already have tested changing the IP address and the "delay_to_wait_after_tuning" with no change to the tuner behavior.

I am working through the balance of the parameters one at a time in hopes to be able to document the specific fix. I do not expect to complete this today.

I am not sure if this analysis will help you recall the specific fix.

My post was more a request to identify the magnitude of the issue. I do not think most people would check to see their tuner count unless asked.

You can use the ones I posted ... I am officially on vacation see you all sometime next Wednesday.

[wayne5w]

Originally Posted by nyplayer

When you were getting no signal did you have the SAGEDCT service disable? I found that if you have sagedct enable and the service starts it ties up the tuners.

Ps I will almost all packed.... Chambers Not too far from where I am in Portland OR.

I had the service stopped but not disabled. Will try that too.

[El Jeffe']

Originally Posted by waylo

Just remembered that with SageDCT, I was using the DUMP setting rather than ffmpeg, so that may be a crucial difference. Could never get SAGEDCT's older ffmpeg to play nice with my clients (on the server I think it was passable), but I probably never looked specifically at the music stations.

Waylo - FYI, my previous SageDCT configuration used the ffmpeg option and all of my music channels were working.

[realcap]

Originally Posted by RangerZ

Yes, you did address the issue, and I had read your post, but you also indicated you did not know what the specific fix was. As indicated above I am working through a comparison to see if I can specifically identify the parameter(s) that will solve the issue.

Please confirm that my question on the typo in your post 176.

The attached spreadsheet compares your file to mine and the default. Mine and the default have the same parameters with some minor value variations. Your file has some additional parameters (5) and there are 2 parameters removed.

I have color coded the entries as follows

• Light green is a direct match
• Dark green is a setting that I believe to only effect the video quality, and probably not relevant to the issue
• Yellow - Parameters that exist in only one or the other file
• White - Not smart enough to know the impact

Yellow nyplayer
mmc/encoders/1012343954/audio_capture_device_index=-1
mmc/encoders/1012343954/audio_capture_device_num=0
mmc/encoders/1012343954/audio_processor=
mmc/encoders/1012343954/dshow_tv_type=
mmc/encoders/1012343954/video_processor=

Yellow default
mmc/encoders/1012343954/1/0/broadcast_standard=
mmc/encoders/1012343954/fast_network_encoder_switch=false

The first dozen or so lines in each section of your file have part of the parameter string changed. A 1 is changed to 100.

I already have tested changing the IP address and the "delay_to_wait_after_tuning" with no change to the tuner behavior.

I am working through the balance of the parameters one at a time in hopes to be able to document the specific fix. I do not expect to complete this today.

I am not sure if this analysis will help you recall the specific fix.

My post was more a request to identify the magnitude of the issue. I do not think most people would check to see their tuner count unless asked.

I'm having the same problem, multiple tuners are being added to video sources by the prime network encoder change. After rebooting, it goes back to 3 tuners but overnight many additional tuners are added. Thanks for working on finding the problem. Please let us know if you find a solution. I appreciate your efforts.

[RangerZ]

Working through the parameters today I made the change below and it seems to have cured the "Growing Phantom Tuners" issue.

For each of the three tuners there is a section that looks like this.

mmc/encoders/1012343954/1/0/available_channels=
mmc/encoders/1012343954/1/0/brightness=-1
mmc/encoders/1012343954/1/0/broadcast_standard=
mmc/encoders/1012343954/1/0/contrast=-1
mmc/encoders/1012343954/1/0/device_name=
mmc/encoders/1012343954/1/0/encode_digital_tv_as_program_stream=false
mmc/encoders/1012343954/1/0/hue=-1
mmc/encoders/1012343954/1/0/last_channel=
mmc/encoders/1012343954/1/0/provider_id=
mmc/encoders/1012343954/1/0/saturation=-1
mmc/encoders/1012343954/1/0/sharpness=-1
mmc/encoders/1012343954/1/0/tuning_mode=Cable
mmc/encoders/1012343954/1/0/tuning_plugin=
mmc/encoders/1012343954/1/0/tuning_plugin_port=0
mmc/encoders/1012343954/1/0/video_crossbar_index=0
mmc/encoders/1012343954/1/0/video_crossbar_type=100

The last parameter "video_crossbar_type" has a value of 100. I scrutinized my sage.properties this morning and noticed that for other tuners in my system (not Prime) that the value of this parameter is the same as the forth value in the string ( the "1" between slashes => "mmc/encoders/1012343954/1/0/video_crossbar_type=100"). I changed this to match the 100 for all of the strings above ("mmc/encoders/1012343954/100/0/....") and for all three tuners. This is also the value for the parameter in the sage.properties file suggested by nyplayer, however he has other changes in his properties file.

I do not know what the parameter is for, but trying to figure this out I found this post which seems to describe the same problem and corroborates the solution and that the index parameter value is for the next segment (which is "0" in both cases). http://forums.sagetv.com/forums/arch...p/t-51168.html

Someone earlier indicated this was a "theme" issue, however I am using the delivered theme.

I have been running for a few hours now and not created ANY new tuners. While it's still to soon to tell, I have only had 1 message indicating a tuner failure, which recovered almost immediately. I have been getting dozens of these tuner failures per day.

The attached text file is a modified version of what is included in the PrimeNetEncoder Install Guide.doc. I also am successfully using the 127.0.0.1 IP as opposed tot he PC's assigned IP.

Make sure to stop the Sage Service before editing the sage.properties file.

Can someone tell me what the 2 video_crossbar parameters do?

RangeZ

[Fuzzy]

Originally Posted by RangerZ

Working through the parameters today I made the change below and it seems to have cured the "Growing Phantom Tuners" issue.

For each of the three tuners there is a section that looks like this.

mmc/encoders/1012343954/1/0/available_channels=
mmc/encoders/1012343954/1/0/brightness=-1
mmc/encoders/1012343954/1/0/broadcast_standard=
mmc/encoders/1012343954/1/0/contrast=-1
mmc/encoders/1012343954/1/0/device_name=
mmc/encoders/1012343954/1/0/encode_digital_tv_as_program_stream=false
mmc/encoders/1012343954/1/0/hue=-1
mmc/encoders/1012343954/1/0/last_channel=
mmc/encoders/1012343954/1/0/provider_id=
mmc/encoders/1012343954/1/0/saturation=-1
mmc/encoders/1012343954/1/0/sharpness=-1
mmc/encoders/1012343954/1/0/tuning_mode=Cable
mmc/encoders/1012343954/1/0/tuning_plugin=
mmc/encoders/1012343954/1/0/tuning_plugin_port=0
mmc/encoders/1012343954/1/0/video_crossbar_index=0
mmc/encoders/1012343954/1/0/video_crossbar_type=100

The last parameter "video_crossbar_type" has a value of 100. I scrutinized my sage.properties this morning and noticed that for other tuners in my system (not Prime) that the value of this parameter is the same as the forth value in the string ( the "1" between slashes => "mmc/encoders/1012343954/1/0/video_crossbar_type=100"). I changed this to match the 100 for all of the strings above ("mmc/encoders/1012343954/100/0/....") and for all three tuners. This is also the value for the parameter in the sage.properties file suggested by nyplayer, however he has other changes in his properties file.

I do not know what the parameter is for, but trying to figure this out I found this post which seems to describe the same problem and corroborates the solution and that the index parameter value is for the next segment (which is "0" in both cases). http://forums.sagetv.com/forums/arch...p/t-51168.html

Someone earlier indicated this was a "theme" issue, however I am using the delivered theme.

I have been running for a few hours now and not created ANY new tuners. While it's still to soon to tell, I have only had 1 message indicating a tuner failure, which recovered almost immediately. I have been getting dozens of these tuner failures per day.

The attached text file is a modified version of what is included in the PrimeNetEncoder Install Guide.doc. I also am successfully using the 127.0.0.1 IP as opposed tot he PC's assigned IP.

Make sure to stop the Sage Service before editing the sage.properties file.

Can someone tell me what the 2 video_crossbar parameters do?

RangeZ

On a capture device, the 'crossbar' is basically the source selector part of the device. So on a typical analog capture device, that has a tuner, composite video, and s-video inputs, the 'crossbar' is what selects which input to capture from. The reason it is the same as that /?/ number in sage is so that sage can be configured to use multiple inputs on the same capture card, so there would be different configurations for each of the inputs on a given source. For devices like network encoders, the crossbar is not utilized, but there is an issue in sage that if it is not matching, it will cause the issues you describe (by creating new entries every time it is parsed out).

[RangerZ]

Thanks for the education.

This explains what I see in sage.properties for my Hauppauge WinTV 885 (multiple inputs). Do the values 1,2,3 have a specific meaning (ie input type) in Sage or just need to be unique?

RangerZ

[Fuzzy]

Originally Posted by RangerZ

Thanks for the education.

This explains what I see in sage.properties for my Hauppauge WinTV 885 (multiple inputs). Do the values 1,2,3 have a specific meaning (ie input type) in Sage or just need to be unique?

RangerZ

There is no specification for what crossbar value equates to what input. It varies by each capture device. As far as network encoders are concerned, the 'crossbar' doesn't matter, but it does need to match itself to avoid the problem described herein.

[RangerZ]

I have been experiencing tuner failures like some of you. Mine have been in the dozens per day. Not new, had them from the start with PrimeNetEncoder and also with SageDCT. I waited to report this until I ferreted out the noise.

I did have less failures with SageDCT, but also had numerous recordings with pixelation. I am of the opion that some of the "soft" failures are "hard" failures in PrimeNetEncoder, but not really sure.

I have reread the majority of the thread.

• I am running the 1.0.1 version
• I currently have the FFMPEG delay at 1300 and think this has reduced the rate, but I still had 9 failures in 18 hours.
• It does not sound like the Prime Firmware upgrades are helping. http://forums.sagetv.com/forums/show...1&postcount=49
• I am running in Service mode. It sounds like this is not an issue other than for the convenience of testing. http://forums.sagetv.com/forums/show...3&postcount=51
• I am running on a 64-bit Vista platform, but am not clear if the problem exists on 32-bit or not.
• Firewall off on the Server\Client, but not clear if there is anything in the router that can be set.
• I tried the 32-bit FFMPEG and it has less failures, but not sure that this is statistically valid.
• All the failures are during recording, as opposed to starting the recording. PrimeNetEncoder indicates a channel change to the same channel.
• I'm not following the logic around post 100 and green drives. Ok, if the drive is sleeping, that's one thing, but in virtually every case of Halt I have the drive is currently recording.
• I have a UPS on the Prime Tuner, but no power conditioning. (noted by KryptoNyte in http://forums.sagetv.com/forums/show...8&postcount=56)
• I have replaced the PC GB MB LAN card with a good Intel GB card, and upgraded to Netgear GS-10X V3 switches.

Failure Types/notes

• Most are Yellow flag events and show in the PrimeNetEncoder as a channel change
• A few are events show no Stop in the PrimeNetEncoder logs, just a channel change.
• One halt resumed with no video (black)
• About 20% of the time multiple tuners fail at about the same time, but then I am not recording on multiple tuners all the time
• Not all failures recover, or recover right away. Some recover 10-20 minutes later, though I think this is playback recovery, not recording.
• Some will recover by toggling the screen, some will require clicking the time line past the red section.

Another poster reported issues with music channels. I also have problems with these. When tuning to a music channel, it takes about 5-7 seconds for the display to read No Signal. About another 5 seconds later it goes to the Music channel static image and begins to play. After a few minutes playing fails and the image is left. I am unable to navigate at any point. The menus do not respond, though sometimes I can get to the main menu after a long delay, but unable to select anything. I have had to close SageTV to continue.

For trouble shooting, I have attached logs, properties, some screen prints of a failure that does not show in messages and a readme.txt file that summarizes the findings of the search string "encoder halt detector" in the included sagetv_X.txt files. It's a rar file with an added zip extension.

RangerZ

[CptKirk]

Just a side note:

After SiliconDust did the big software/firmware update on 6/5/15 they respun/updated just the firmware on 6/15/15. If you are not on the 6/15/15 firmware I would suggest doing that update.

It would be a good idea to add the Prime firmware version you are using to your signature.

Release 20150615:
- CC models: Improvements to channel scan with a tuning resolver present.
- CC models: Fix problem causing device to temporarily stop authorizing channels.
- CC/DC models: Fix problem causing device to temporarily stop responding.
- DC/HDHR4-2DT models: Fix problem causing some channels to be missed in the channel scan.
- HDHR4/TC/CC/DC models: Improvements to diagnostic logging.

[Tiki]

Originally Posted by RangerZ

I have been experiencing tuner failures like some of you. Mine have been in the dozens per day. Not new, had them from the start with PrimeNetEncoder and also with SageDCT. I waited to report this until I ferreted out the noise.

I did have less failures with SageDCT, but also had numerous recordings with pixelation. I am of the opion that some of the "soft" failures are "hard" failures in PrimeNetEncoder, but not really sure.

I have reread the majority of the thread.

• I am running the 1.0.1 version
• I currently have the FFMPEG delay at 1300 and think this has reduced the rate, but I still had 9 failures in 18 hours.
• It does not sound like the Prime Firmware upgrades are helping. http://forums.sagetv.com/forums/show...1&postcount=49
• I am running in Service mode. It sounds like this is not an issue other than for the convenience of testing. http://forums.sagetv.com/forums/show...3&postcount=51
• I am running on a 64-bit Vista platform, but am not clear if the problem exists on 32-bit or not.
• Firewall off on the Server\Client, but not clear if there is anything in the router that can be set.
• I tried the 32-bit FFMPEG and it has less failures, but not sure that this is statistically valid.
• All the failures are during recording, as opposed to starting the recording. PrimeNetEncoder indicates a channel change to the same channel.
• I'm not following the logic around post 100 and green drives. Ok, if the drive is sleeping, that's one thing, but in virtually every case of Halt I have the drive is currently recording.
• I have a UPS on the Prime Tuner, but no power conditioning. (noted by KryptoNyte in http://forums.sagetv.com/forums/show...8&postcount=56)
• I have replaced the PC GB MB LAN card with a good Intel GB card, and upgraded to Netgear GS-10X V3 switches.

Failure Types/notes

• Most are Yellow flag events and show in the PrimeNetEncoder as a channel change
• A few are events show no Stop in the PrimeNetEncoder logs, just a channel change.
• One halt resumed with no video (black)
• About 20% of the time multiple tuners fail at about the same time, but then I am not recording on multiple tuners all the time
• Not all failures recover, or recover right away. Some recover 10-20 minutes later, though I think this is playback recovery, not recording.
• Some will recover by toggling the screen, some will require clicking the time line past the red section.

Another poster reported issues with music channels. I also have problems with these. When tuning to a music channel, it takes about 5-7 seconds for the display to read No Signal. About another 5 seconds later it goes to the Music channel static image and begins to play. After a few minutes playing fails and the image is left. I am unable to navigate at any point. The menus do not respond, though sometimes I can get to the main menu after a long delay, but unable to select anything. I have had to close SageTV to continue.

For trouble shooting, I have attached logs, properties, some screen prints of a failure that does not show in messages and a readme.txt file that summarizes the findings of the search string "encoder halt detector" in the included sagetv_X.txt files. It's a rar file with an added zip extension.

RangerZ

Given the large number of failures and the fact that you are also seeing failures with other tuners, it might be good to check for hardware issues with your system.

I would start by checking your Windows logs for any errors. I didn't see what version of Windows you are using, but in Windows 7, you can go to control panel -> administrative tools -> event viewer. I would be on the lookout especially for anything related to your hard disks or disk controllers, networking, or memory.

You may also want to get a a memory test program to check your RAM.

Troubleshooting the network can be a little more challenging (unless you happen to have some spare gear around that you can easily swap out). But, a bad cable or a flaky Ethernet switch can cause all kinds of weird problems.

[RangerZ]

I upgraded the SD firmware to the latest version (20150615). I had some initial problems that seemed to disappear after a couple of reboots of the PC and the tuners. Unfortunately this has made no impact on failures. I averaged 1 per hour, and have two incidents overnight where 2 tuners failed at the same time.

At first when watching a station it would go back to the program guide where I saw a red circle with a white center on the active program. I have an old HDHR-US that I use for OTA. On those stations the when the time line was clicked the top and bottom overlay flashed white. The menus on these channels lost animation, they just were or were not.

Regarding hardware, it's worth checking, but I was down this road a while back and found nothing. Not sure I can do it this weekend. I replaced the video card with a newer ATI and upgraded the LAN card to an Intel (based on other posts for SageDCT) about a year back. I do have issues with Catalyst Control Center crashing. It's been an issue from the start and I have tried different versions and different cards. All of the current failures are with it still running.

I have looked at the logs and there are very few problems. The attached are examples of the type of failures I see in the logs. I can not link any error to a tuning failure based upon time.

• There is an e1qexpress issue every 3-4 hours. I lost the part where the connection comes back about 3-4 seconds later. There is always the iScsiPrt in the middle.
• The asstor64 seems to come in groups, but there have been none for a few days. Not using raid per say.
• Beep should be a non issue.
• I also have the WMI failure, that there does not seem to be a good fix for in general.

RangerZ