Hi, just wanted to add my case.
Same situation: Argentina - Cablevision/Fibertel.
Hi, just wanted to add my case.
Same situation: Argentina - Cablevision/Fibertel.
As a workaround, I have switched the order of the agents on the Plex Movie settings. I moved CineMaterial and Fanart.tv to the top, Iām getting the posters from them. Seems to be working fine by now. I know it is not ideal, but I prefer that to the screenshots or the blank posters 
My agents are as follows:
Hope this helps!
The problem is being investigated by tmdb / cloudflare - they have asked for anyone with the issue to provide the output from this request
https://image.tmdb.org/cdn-cgi/trace
There is no harm in also providing traceroute image.tmdb.org output
If anyone still affected in Argentina please post the results here
Iām in Argentina - Cablevision-Fibertel is my ISP
https://image.tmdb.org/cdn-cgi/trace
fl=44f40
h=image.tmdb.org
ip=190.194.xxx.xxx
ts=1525131268.382
visit_scheme=https
uag=Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
colo=EZE
spdy=h2
http=h2
loc=AR
tracert image.tmdb.org output:
Tracing route to image.tmdb.org [104.16.58.155]
over a maximum of 30 hops:
1 <1 ms 2 ms 1 ms testwifi.here [192.168.86.1]
2 * * * Request timed out.
3 * * * Request timed out.
4 * * * Request timed out.
5 21 ms 22 ms 23 ms 86-165-89-200.fibertel.com.ar [200.89.165.86]
6 17 ms 17 ms 18 ms 200.0.17.181
7 17 ms 17 ms 18 ms 104.16.58.155
Trace complete.
Thanks!
fl=44f21
h=image.tmdb.org
ip=186.136.xxx.xxx
ts=1525220382.499
visit_scheme=https
uag=Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.139 Safari/537.36
colo=EZE
spdy=h2
http=h2
loc=AR
Tracing route to image.tmdb.org [104.16.57.155]
over a maximum of 30 hops:
1 <1 ms <1 ms <1 ms READYSHARE [192.168.1.1]
2 * * * Request timed out.
3 * * * Request timed out.
4 * * * Request timed out.
5 11 ms 23 ms 15 ms 86-165-89-200.fibertel.com.ar [200.89.165.86]
6 19 ms 9 ms 9 ms 200.0.17.181
7 9 ms 11 ms 17 ms 104.16.57.155
Trace complete.
@luquiyahni said:
As a workaround, I have switched the order of the agents on the Plex Movie settings. I moved CineMaterial and Fanart.tv to the top, Iām getting the posters from them. Seems to be working fine by now. I know it is not ideal, but I prefer that to the screenshots or the blank postersMy agents are as follows:
- Cinematerial
- Fanart.tv
- Plex Movie
- Local Media Assets
- OpenSubtitles.org
Hope this helps!
Where did you get that cinematerial agent?
edit: my bad, itās already on plex.
Same here with Cablevisión/Fibertel, I wish I had found this post sooner, I tried so many different things and couldnāt find any issueā¦
fl=44f41
h=image.tmdb.org
ip=190.191.XXX.XXX
ts=1525581058.446
visit_scheme=https
uag=Mozilla/5.0 (Windows NT 6.3; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.139 Safari/537.36
colo=EZE
spdy=h2
http=h2
loc=AR
1 <1 ms <1 ms <1 ms wifi.fibertel [192.168.1.1]
2 2 ms 2 ms 1 ms 192.168.0.1
3 * * * Tiempo de espera agotado para esta solicitud.
4 * * * Tiempo de espera agotado para esta solicitud.
5 * * * Tiempo de espera agotado para esta solicitud.
6 * * * Tiempo de espera agotado para esta solicitud.
7 16 ms 14 ms 15 ms 149-161-89-200.fibertel.com.ar [200.89.161.149]
8 19 ms 15 ms 15 ms 150-165-89-200.fibertel.com.ar [200.89.165.150]
9 12 ms 12 ms 12 ms 200.0.17.181
10 13 ms 20 ms 13 ms 104.16.61.155
Thanks to everyone for the extra information requested by cloudflare tech support
What we need to do now is to try and replicate the issue through making direct requests to image.tmdb.org - when we tried making individual requests before through curl, they did not fail - the issue may be only arising when a batch of requests are made
I have been looking replicating what was seen in the wireshark capture provided by @gonse and to produce a windows batch file that makes similar requests as was in the wireshark capture and to capture the results
@gonse If you are available to help, I have a windows batch file that will make 84 image fetches batched in similar way to what I saw in the wireshark and then we can see if any fail. I am adding similar delays between the batches of requests as was observed
I will send you zip of the batch file in a private message. Please create a directory where you place that batch file in and also copy curl.exe to the same directory
Then in a command line window, run the batch file. It will spawn a number of command line sessions for the requests made in parallel and will then suspend for a number of seconds before firing off another batch and so on until all 84 requests are made. The image files fetched will be stored in that directory as well as the response headers received for reach request
At the end please zip the directory and send it to me
Thanks
Looking for a couple more users running Plex Media Server on Windows in Argentina with this issue to try out a couple of test batch files - retrieving a number of images and to save the results for each of the 2 tests and zip and send to me
@sa2000 said:
Looking for a couple more users running Plex Media Server on Windows in Argentina with this issue to try out a couple of test batch files - retrieving a number of images and to save the results for each of the 2 tests and zip and send to me
Iām the OP of this thread with the same issue, and willing to test the batch file.
@EmilianoTalamo said:
@sa2000 said:
Looking for a couple more users running Plex Media Server on Windows in Argentina with this issue to try out a couple of test batch files - retrieving a number of images and to save the results for each of the 2 tests and zip and send to meIām the OP of this thread with the same issue, and willing to test the batch file.
sent zip with two test batch files
@sa2000 said:
Looking for a couple more users running Plex Media Server on Windows in Argentina with this issue to try out a couple of test batch files - retrieving a number of images and to save the results for each of the 2 tests and zip and send to me
Fell free to send it to me too. Iām also located in Argentina.
@sa2000 said:
Looking for a couple more users running Plex Media Server on Windows in Argentina with this issue to try out a couple of test batch files - retrieving a number of images and to save the results for each of the 2 tests and zip and send to me
Feel free to send it my way as well.
Currently, when I looked at the Wireshark capture it shows a lot of TCP retransmission errorsā¦
Hi @sa2000, I also have this issue (Argentina + Fibertel) and Iād love to help testing. My logs are:
fl=44f39
h=image.tmdb.org
ip=190.192.xxx.xxx
ts=1526818740.713
visit_scheme=https
uag=Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.181 Safari/537.36
colo=EZE
spdy=off
http=http/1.1
loc=AR
traceroute to image.tmdb.org (104.16.59.155), 30 hops max, 60 byte packets
1 * * *
2 * * *
3 * * *
4 * * *
5 * * *
...
30 * * *
Yes, my traceroute is a blatant failure 
@AgustinCordes sent you test batch files
Tests so far for other users did not get failures using the batch with curl
@sa2000, sent! And please note that I was issuing the traceroute on the Ubuntu subsystem, which for some reason fails. My proper trace is this:
Tracing route to image.tmdb.org [104.16.59.155]
over a maximum of 30 hops:
1 1 ms 1 ms 1 ms 192.168.0.1
2 * * * Request timed out.
3 * * * Request timed out.
4 * * * Request timed out.
5 468 ms 453 ms 389 ms 133-161-89-200.fibertel.com.ar [200.89.161.133]
6 337 ms 343 ms 197 ms 150-165-89-200.fibertel.com.ar [200.89.165.150]
7 * * * Request timed out.
8 442 ms 439 ms 470 ms 104.16.59.155
Trace complete.
Itās seems the problem has been fixed.
Since a few hours ago, Iām able to pull TMDB posters and backgrounds from the Plex Media Server, without any issues.
Thanks @gonse for the feedback - could others on this forum thread please check
The test scripts I sent out to a few of you did not fail to fetch images - the scripts downloaded 84 images each and with 4 tests the maximum would have been 336 images and whilst there was evidence of some fetches taking a few seconds (up to 4 seconds noticed) there were no failures.
I am guessing that with a much higher volume of image requests that the times to complete some requests may have exceeded the 20 second timeout we have coded in within the agent in Plex Media Server.
Lets hope that the issue has been resolved - if not, I will increase the number of fetches in the scripts and see if we get failures
Back to square one.
Problem downloading poster, background and actor images is back.
I dontā know if this is the solution, but since Fibertel change my router Technicolor (Cisco) to Segemcom, all the problems solve.
No se si fue eso, pero desde que Fibertel me cambio mi router del Technicolor (Cisco) a un Segemcom no tuve mas problemas, ahora anda a los pedos, incluso el error de editar una pelicula dentro de la pelicula (lo que habiamos hablado @gonse)