EPG fails to update

[shermanh4]

This is happening now for me too. Has been this way through several beta channel releases. Finally dug into this morning.

Here’s where things went south in the PMS log:

Sep	22,	2018	13:36.0	[0x700002133000]	DEBUG - EPG: Found a dupe show ‘House Flipping 101’ with a different GUID.
Sep	22,	2018	13:36.2	[0x700002133000]	DEBUG - EPG[onconnect]: Step 20/56 (2018-09-27T12:00Z) (network: 20.1 sec, database: 166.7 sec total: 185.7 sec)
Sep	22,	2018	13:36.2	[0x700002133000]	DEBUG - Activity: updated activity 255beb6c-a1da-483c-bc58-81ad3a6f2fe9 - completed 35% - Refreshing guide data
Sep	22,	2018	13:36.2	[0x700002133000]	DEBUG - Activity: updated activity 255beb6c-a1da-483c-bc58-81ad3a6f2fe9 - completed 35% - Refreshing guide data
Sep	22,	2018	13:36.3	[0x700002757000]	DEBUG - NetworkServiceBrowser: Parsing SSDP schema for http://172.16.***.***:8080/upnp
Sep	22,	2018	13:36.3	[0x700002757000]	DEBUG - HTTP requesting GET http://172.16.***.***:8080/upnp
Sep	22,	2018	13:36.3	[0x700002757000]	DEBUG - HTTP 404 response from GET http://172.16.***.***:8080/upnp
Sep	22,	2018	13:36.3	[0x700002757000]	DEBUG - NetworkServiceBrowser: found 0 SSDP devices via http://172.16.***.***:8080/upnp
Sep	22,	2018	13:36.3	[0x700002133000]	ERROR - EPG: JSON parse error. Wrote JSON data to “/Users/***/Library/Application Support/Plex Media Server/Diagnostics/EPG-onconnect-7f7a884a-a64a-467e-b2a7-63eea6182db8.json”
Sep	22,	2018	13:36.3	[0x700002133000]	WARN - EPG[onconnect]: Failed to load EPG.
Sep	22,	2018	13:36.3	[0x700002133000]	DEBUG - Database: Shutting down.
Sep	22,	2018	13:36.3	[0x700002133000]	DEBUG - Captured session 0.
Sep	22,	2018	13:36.3	[0x700002133000]	DEBUG - Captured session 1.
Sep	22,	2018	13:36.3	[0x700002133000]	DEBUG - Captured session 2.
Sep	22,	2018	13:36.3	[0x700002133000]	DEBUG - Captured session 3.
Sep	22,	2018	13:36.3	[0x700002133000]	DEBUG - Captured session 4.
Sep	22,	2018	13:36.4	[0x700002133000]	DEBUG - EPG[onconnect]: Total time to load EPG was 185.8 (HTTP details cached 95.5%, CloudFlare grid cached: 65.0%, 1 HTTP errors)

The JSON file says:
{
“message”: “Request forbidden by administrative rules”,
“__type”: “CloudSearchException”
}

[sipuncher]

Hi,

I’m having guide refresh issues. I’ve only been using DVR for a month now that community tuners are supported. Two issues:

• the guide takes hours to refresh with only 124 channels enabled
• I often find that the guide isn’t available for the next day and so have to manually refresh. This fixes it.

Any known issues here?

Latest beta server build and DVB-T2 tuner in the UK.

Thanks in advance.

[sa2000]

@sipuncher

’m having guide refresh issues. I’ve only been using DVR for a month now that community tuners are supported. Two issues:

• the guide takes hours to refresh with only 124 channels enabled
• I often find that the guide isn’t available for the next day and so have to manually refresh. This fixes it.

Any known issues here?

Latest beta server build and DVB-T2 tuner in the UK.

Thanks in advance.

I am looking into an issue of getting timeouts when fetching the guide data - but that would lead to failure and not the symptoms you describe

If you find the guide data is not available, please get logs. You may need to collect logs every few hours as the log files are recycled - you could increase the number of log files via editing advanced settings

To get logs, ensure you run with debug logging enabled
see https://support.plex.tv/articles/201643703-reporting-issues-with-plex-media-server/

Restart the server to get fresh logs created.

Collect the logs through the web interface - see https://support.plex.tv/articles/200250417-plex-media-server-log-files/

To increase the number of log files - see https://support.plex.tv/articles/201105343-advanced-hidden-server-settings/
The setting is named LogNumFiles

[Coxylox]

I’m having similar guide issues. Logs attached Plex Media Server Logs_2018-10-03_21-26-07.zip (2.1 MB)

Everything has been working fine (to my knowledge) until today, when I needed to delete and re-add my DVR in order to take advantage of the new HDHomeRun Premium TV services i.e. I needed to remap the channels to “Cable” vs. “Antenna”.

In any case, I can’t seem to get any consistent results. At one point I was able to get a few days worth of guide data but now it seems to just be silently failing and pulling 0% results. Any help is appreciated.

PMS: 1.13.8.5395
Windows 10
HDHomeRun EXTEND
zip: 92117
DVR—Cox Comm South County - Digital

[Coxylox]

Would be great to get any kind of update on this. Without the guide, I can’t record, which is my main source of content.

[sa2000]

@Coxylox

I’m having similar guide issues. Logs attached Plex Media Server Logs_2018-10-03_21-26-07.zip (2.1 MB)

Everything has been working fine (to my knowledge) until today, when I needed to delete and re-add my DVR in order to take advantage of the new HDHomeRun Premium TV services i.e. I needed to remap the channels to “Cable” vs. “Antenna”.

In any case, I can’t seem to get any consistent results. At one point I was able to get a few days worth of guide data but now it seems to just be silently failing and pulling 0% results. Any help is appreciated.

There is some unexpected data in the EPG - please zip the following file and attach here
C:\Users\coxyl_000\AppData\Local\Plex Media Server\Diagnostics\EPG-onconnect-654f8ed7-9fde-445d-824c-f92c871e17dd.json

There may be many files in this directory
C:\Users\coxyl_000\AppData\Local\Plex Media Server\Diagnostics
You can zip ones created in the last week or so and attach the zip as well as the above specific file

[Coxylox]

Here you go, let me know if you need anything else.

EPG-onconnect-specific-file.zip (628.8 KB)

The diagnostic files were too large to upload, so here’s a link to download:
https://drive.google.com/file/d/1-yHRi3wrh9wMiHRfwmSjCjNlNDmUu8BL/view?usp=sharing

[jester_tenet]

Just mentioning this forum post, in case it’s related;

UK EPG, any scripts that I can run to force a full EPG download

My Plex server currently only downloads a couple of days of EPG in the UK, used to be two weeks, but if I do a full download, then I get a lot more than the next two days. Does anyone know any way to have a script that every-night, would force a full download? Ideally, I would like Plex to fix this, as well as the current issue that Plex thinks channels are HD channels when they are not in the UK, but as no response to this from Plex I’m looking for a workaround

[sa2000]

@jester_tenet

Just mentioning this forum post, in case it’s related;

Not in this case.

[sa2000]

@Coxylox

Here you go, let me know if you need anything else.

The diagnostic files were too large to upload, so here’s a link to download:

Thank you for the diagnostics and logs

Whilst the JSON files do have errors, in each case it follows a curl error 56 which is a Curl Receive Error

Oct 03, 2018 21:25:16.037 [6540] ERROR - Error issuing curl_easy_perform(handle): 56
Oct 03, 2018 21:25:16.065 [6540] DEBUG - JSON parse error: Missing a closing quotation mark in string. (4511131)
Oct 03, 2018 21:25:16.096 [6540] ERROR - EPG: JSON parse error. Wrote JSON data to "C:\Users\coxyl_000\AppData\Local\Plex Media Server\Diagnostics\EPG-onconnect-654f8ed7-9fde-445d-824c-f92c871e17dd.json"
Oct 03, 2018 21:25:16.097 [6540] WARN - EPG[onconnect]: Failed to load EPG.

I suspect that the Error 56 CURLE_RECV_ERROR (56) is resulting in trnucated / incomplete JSON files.

In case it is a factor, could you tell me what the MTU Size is set to in the router

Also would like to see if we can get any clues for why the downloads of the EPG are repeatedly failing with CURLE_RECEIVE_ERROR

As this appears to be a repeatable issue - could you try to refresh with wireshark capturing the network traffic. and save into pcap file when you get the Refresh Error. Attach logs zip and any new JSON files created within the C:\Users\coxyl_000\AppData\Local\Plex Media Server\Diagnostics\ directory

For the wireshak capture, please zip the pcap file and send it to me by Private Message

Wireshark · Go Deep

[Coxylox]

I’ve sent you the files, let me know if you need anything else there. With respects to the MTU, I’m using Eero and I think the MTU is 1500. There’s no setting to change that unfortunately and that number I got from browsing the web, so I’m not sure if it’s accurate.

[Coxylox]

@sa2000

In case it is a factor, could you tell me what the MTU Size is set to in the router

using the below method I got a MTU of 1321

https://www.linksys.com/us/support-article?articleNum=134914#f

[sa2000]

@Coxylox

using the below method I got a MTU of 1321

This is the repetitive ping to see at what point it would fragment. This would tell you what you should set it to by seeing when it would fragment

MTU 1321 sounds a bit low

Could you go through this sequence

Start an elevated Command Prompt window (right click on Command Prompt and select Run As Administrator and type the following starting with 1452:

ping -f -l 1452 www.plex.tv

(That is a dash lower case “L,” not a dash “1.” Also note the spaces)

f you get an error that the packet needs to be fragmented , it means the message size is too high and you should try reducing the number in the ping until you do not get this error.

If there is no error, increase the message size by 10 and repeat the ping test, specifying 1462 as the message size. Repeat this until you get the packet needs to be fragmented error - when you do, then you need to establish the right value to use below this one. So test again with lower value - reducing this time a byte at a time

Add 28 more to this (since you specified ping packet size, not including IP/ICMP header of 28 bytes), and this should be your MaxMTU.

Note:If you can ping through with the number at 1472, you are done! Stop right there. Add 28 and your MaxMTU is 1500.

For PPPoE, your MaxMTU should be no more than 1492 to allow space for the 8 byte PPPoE “wrapper,” but again, experiment to find the optimal value. For PPPoE, the stakes are high: if you get your MTU wrong, you may not just be sub-optimal, things like UPLOADING or web pages may stall or not work at all

[Coxylox]

It came out to 1300 MTU (1272 + 28)

[sa2000]

I do not know if it is relevant to the issue

The wireshark capture shows RST received from cloudflare system that is responding to the EPG data requests - the RST is being received just after an EPG segment is received - I suspect the RST then gives rise to the CURLE_RECEIVE_ERROR and the partial EPG JSON file.

May be the MTU size is working out to be 1300 because of the existing setting in the router

Do you know why it is so low ?
Any way to set it to 1500 and then restart the router and then do the repetitive ping test

[Coxylox]

Why it is 1300, I do not know. That being said, since I’ve read conflicting information on what the MTU size if for the Eero router, I decided to see if it might be on the client side (Windows 10). Using this method, I increased my ethernet adapter to 1500 and retested. I was able to get an MTU of 1492 (1464 + 28) before getting the packet needs to be fragmented error.

I also pinged Eero support to get more clarity on what the default MTU is and if there is a way to change it.

Unfortunately Plex is still failing to refresh the EPG, even with these settings. I will PM you with the latest Wireshark, logs and diagnostic files.

[sa2000]

Sorry - that was just a thought - but I have confirmed myself that it is not a factor. I set my MTU size to 1300 and i got similar packet sizes to yours from Cloudflare seen in wireshark and i did not get the reset.

What is happening in your case is that connection that is downloading the Guide data is being reset by cloudflare. I do not know why that is happening

Each of the failures in the server logs was preceded by a CURLE_RECV_ERROR which appears to map to the RST TCP packet received

Is there a possibility of eliminating the router and ISP for a short test - eg connect the server to a cellular hotspot for a short while and trying a Refresh through http://127.0.0.1:32400/web on the server? (That would depend on whether you have ample data allowance for cellular)

[Coxylox]

@sa2000

Is there a possibility of eliminating the router and ISP for a short test - eg connect the server to a cellular hotspot for a short while and trying a Refresh through http://127.0.0.1:32400/web on the server? (That would depend on whether you have ample data allowance for cellular)

That was an interesting test. Took a while but still failed unfortunately. It failed somewhere around 55% refresh, which was much higher than I have seen in quite a while. Not sure if that’s relevant but thought I would mention it. I’ll PM you with the latest files to see if maybe something different happened.

[Coxylox]

@Coxylox

With respects to the MTU, I’m using Eero and I think the MTU is 1500.

For anyone following along or interested, here’s the response I got back from Eero support.

Hi there,

Thanks for getting in touch with us at eero. Good question, surprisingly no one has asked that before to my knowledge. The default MTU size on eero is 1500, this can not be changed however. If you have any other questions about this just let me know, I will be glad to help out. Thanks again.

Jeremy | eero Support