Roku Direct Publisher

[jtalbott]

Does anyone know what IP address ROKU accesses DirectPublisher feeds from? We need to open our CloudFlare firewall to ROKU but have no idea where to find the IP info.

The issue we are trying to resolve is that the feed is bombing on ingestion. The error says "Thumbnail could not be downloaded." If I upload the image to a different domain (without CloudFlare) the ingestion is good.

What is ironic is that the srt file we upload for closed captions has no problems and is hosted on the same domain that ROKU can't download the images from. I don't know why image files should work any differently than closed caption files as far as access is concerned.

[nabeards]

When I refresh a feed, these are the two IPs that immediately hit:
54.208.55.67
52.200.168.210

Both are AWS IPs, coincidentally, so could easily change in the near future. However, the second one is one I see often on feed updates so it's likely pretty stable.

HTH,
nabeards

[heyjones]

I'm having this same issue with WP Engine. They see a 403 error on their end when Roku attempts to get the thumbnail. That's interesting about the srt file, you'd think it would bomb out on both. I might try contacting Roku support and see if it's an issue on their end, as I'm not sure what else to do at this point.

[heyjones]

This is what I'm seeing in my nginx logs... same IP address. It looks like it has no problem getting the feed and the video file, but bombs out on the thumbnails:

52.200.168.210 - heyjones.com - [10/Jun/2018:14:55:44 +0000] "GET /feed/roku-direct-publisher/ HTTP/1.0" 200 948 "-" "Mozilla/5.0"
52.200.168.210 - heyjones.com - [10/Jun/2018:14:55:44 +0000] "GET /wp-content/uploads/2018/06/Screen-Shot-2018-06-09-at-9.01.52-AM-1920x1080.png HTTP/1.0" 403 162 "-" "Java/1.8.0_131"
52.200.168.210 - heyjones.com - [10/Jun/2018:14:55:48 +0000] "GET /wp-content/uploads/2018/06/Screen-Shot-2018-06-09-at-9.01.52-AM-1920x1080.png HTTP/1.0" 403 162 "-" "Java/1.8.0_131"
52.200.168.210 - heyjones.com - [10/Jun/2018:14:55:51 +0000] "GET /wp-content/uploads/2018/06/Screen-Shot-2018-06-09-at-9.01.52-AM-1920x1080.png HTTP/1.0" 403 162 "-" "Java/1.8.0_131"
52.200.168.210 - heyjones.com - [10/Jun/2018:14:55:51 +0000] "GET /wp-content/uploads/2018/06/TextInMotion-Sample-1080p.mp4 HTTP/1.0" 200 23725150 "-" "Mozilla/5.0"

[heyjones]

I tried disabling CDN and renaming the file to remove non-alphanumeric characters but still get the same error.

[nabeards]

You are blocking requests to the thumbnail URL because of the User-Agent. Here's `curl` commands you can run to see:

>curl -I -H "User-Agent:Java/1.8.0_131" "http://heyjones.com/wp-content/uploads/2018/06/Screen-Shot-2018-06-09-at-9.01.52-AM-1920x1080.png"
HTTP/1.1 403 Forbidden
Server: nginx
Date: Mon, 11 Jun 2018 21:19:20 GMT
Content-Type: text/html
Content-Length: 162
Connection: keep-alive
Keep-Alive: timeout=20

>curl -I "http://heyjones.com/wp-content/uploads/2018/06/Screen-Shot-2018-06-09-at-9.01.52-AM-1920x1080.png"
HTTP/1.1 200 OK
Server: nginx
Date: Mon, 11 Jun 2018 21:20:31 GMT
Content-Type: image/png
Content-Length: 1162332
Last-Modified: Sat, 09 Jun 2018 16:02:28 GMT
Connection: keep-alive
Keep-Alive: timeout=20
ETag: "5b1bfa14-11bc5c"
X-Type: static/known
Cache-Control: public, max-age=2592000
Vary: Accept-Encoding
Access-Control-Allow-Origin: *
Accept-Ranges: bytes


So you'll have to look at your config for why you are doing that...
nabeards