docker-plex-remote-transcoder
docker-plex-remote-transcoder copied to clipboard
deevus
Slave shows ups on check config and get cluster load, but no transcode
Any Ideas? For all a can tell the slave is there. I had to fiddle with the ssh keys to get them usable, but the master logs in no password with the plex user via ssh.
But any time I run a transcode, it’s handled by the master.
Could it be that the containers are not running as plex user?
Is it perhaps related to https://github.com/deevus/docker-plex-remote-transcoder/issues/13?
If you could provide some logging that may be helpful
Here's a "normal" transcode.
2019-09-16 21:15:49,888 - prt - INFO - Checking for orphaned PRT processes
2019-09-16 21:15:49,892 - prt - INFO - Found 0 orphaned PRT processes
2019-09-16 21:15:49,911 - prt - INFO - No hosts found...using local
2019-09-16 21:15:49,912 - prt - INFO - Debug mode - enabling verbose ffmpeg output
2019-09-16 21:15:49,913 - prt - INFO - Launching transcode_local: ['/usr/lib/plexmediaserver/plex_transcoder', '-codec:0', 'h264', '-codec:1', 'aac', '-analyzeduration', '20000000', '-probesize', '20000000', '-i', '/data/Videos_To_check/F1/Formula1.2019.Italian.Grand.Prix.Qualifying.1080p.WEB.h264-VERUM-Obfuscated/c2fb185bada244d8a71044919b0c1e38.mkv', '-filter_complex', '[0:0]scale=w=480:h=270[0];[0]format=pix_fmts=yuv420p|nv12[1]', '-filter_complex', "[0:1] aresample=async=1:ocl='stereo':osr=44100[2]", '-map', '[1]', '-metadata:s:0', 'language=eng', '-codec:0', 'libx264', '-crf:0', '23', '-maxrate:0', '521k', '-bufsize:0', '1042k', '-r:0', '25', '-preset:0', 'veryfast', '-x264opts:0', 'subme=2:me_range=4:rc_lookahead=10:me=dia:no_chroma_me:8x8dct=0:partitions=none', '-force_key_frames:0', 'expr:gte(t,0+n_forced*8)', '-map', '[2]', '-metadata:s:1', 'language=eng', '-codec:1', 'aac', '-b:1', '125k', '-f', 'dash', '-min_seg_duration', '8000000', '-skip_to_segment', '1', '-time_delta', '0.0625', '-manifest_name', 'http://127.0.0.1:32400/video/:/transcode/session/acx31spzt2yvg0h6mzau984u/69400979-8916-4e0f-9d83-5693b833052c/manifest', '-avoid_negative_ts', 'disabled', '-map_metadata', '-1', '-map_chapters', '-1', 'dash', '-start_at_zero', '-copyts', '-vsync', 'cfr', '-y', '-nostats', '-loglevel', 'verbose', '-loglevel_plex', 'verbose', '-progressurl', 'http://127.0.0.1:32400/video/:/transcode/session/acx31spzt2yvg0h6mzau984u/69400979-8916-4e0f-9d83-5693b833052c/progress']
Where it reports no slaves.
But when I exec into the container (master) and run /usr/lib/plexmediaserver/Plex\ Transcode I get a connection:
2019-09-16 21:18:35,707 - prt - INFO - Checking for orphaned PRT processes
2019-09-16 21:18:35,709 - prt - INFO - Found 0 orphaned PRT processes
2019-09-16 21:18:36,157 - prt - INFO - Host with minimum load is 'folco.local'
2019-09-16 21:18:36,157 - prt - INFO - Using transcode host 'folco.local'
2019-09-16 21:18:36,157 - prt - INFO - Launching transcode_remote with args ['ssh', '-tt', '-R', '32400:127.0.0.1:32400', u'[email protected]', '-p', u'8022', "export LANG=C.UTF-8;export PLEX_UID=1000;export TERM=xterm;export TZ=Pacific/Auckland;export SLAVE_USER=plex;export SHLVL=1;export CHANGE_CONFIG_DIR_OWNERSHIP=true;export HOSTNAME=d32d06b6cfe9;export ADVERTISE_IP=http://drogo.local:32400;export SLAVE_PORT=8022;export PWD=/;export SLAVE_IP=folco.local;export PLEX_GID=1000;export PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin;export;export LC_ALL=C.UTF-8;export MASTER_IP=drogo.local;export HOME=/config;export _='/usr/lib/plexmediaserver/Plex Transcoder';export PRT_ID=f5fffdfed86211e989220242ac1f0002;cd /;prt_local "]
So I think all the bits are there, and the ssh connection is being established, but for some reason the master doesn't seem to think so. I did see a note about the easyaudiotranscoder being used (being a reason for not using a slave)....but while the transcode was running I looked at the process list and the easyaudio transcoder wasn't running (so assume it's not that)
Hmm I'm not sure. I'm not currently using this myself, but if you can work it out I will accept a PR.
Yeah not knowing why it’s not using the slave I’ve asked on pro as well. I wonder if something with the plex user
So I fiddled with the prt python script and added some debugging to the ssh command
And saw 2019-09-17 08:35:42,283 - prt - DEBUG - stderr: ssh: /usr/lib/plexmediaserver/lib/libcrypto.so.1.0.0: no version information available (required by ssh) ssh: /usr/lib/plexmediaserver/lib/libcrypto.so.1.0.0: no version information available (required by ssh) ssh: relocation error: ssh: symbol ENGINE_load_builtin_engines, version OPENSSL_1.0.0 not defined in file libcrypto.so.1.0.0 with link time reference
Any ideas why (or how to stop it) from trying that libcrypto file and not another one?
Though I note the permissions on that folder as root:root where a lot of other files in the container are the Plex user.
The docker images had stopped building. Now that I've rebuilt them can you try again and see if the issue persists?
I'll give it a go now...actually a hack I'd made to the master /root/etc/cont-init.d/62-ssh-keygen
is after the:
printf "StrictHostKeyChecking no\n" >> /etc/ssh/ssh_config
I've added:
#update the ssh config and set the pub/private keys
sed -i 's+# IdentityFile ~+IdentityFile /config+' /etc/ssh/ssh_config
Which sets the pub/private key etc to the copy in /config/.ssh folder. Otherwise i found the keys didn't work (do yours?)
So no difference for me.
2019-09-17 19:05:51,465 - prt - INFO - Checking for orphaned PRT processes
2019-09-17 19:05:51,478 - prt - INFO - Found 0 orphaned PRT processes
2019-09-17 19:05:51,493 - prt - INFO - No hosts found...using local
2019-09-17 19:05:51,494 - prt - INFO - Debug mode - enabling verbose ffmpeg output
I assume it's that ibcrypto.so.1.0.0 issue (I have to hack around in the prt source to get that to show up).
Like I said I'm not using this currently. If you can work it out I'll happily merge it in
Cheers anyway. The only thing I can think of is to run as the plex user...maybe USER plex inth dockerfile. Will give that a go.