mirror of
https://github.com/community-scripts/ProxmoxVE.git
synced 2026-03-31 06:34:17 -04:00
ErsatzTV 25.2.0 Update Issues #1343
Closed
opened 2025-11-20 05:08:29 -05:00 by saavagebueno
·
33 comments
No Branch/Tag Specified
main
github-action-update-changelog
pr-update-app-files
add-script-bambuddy-1774853250
MickLesk-patch-1
fix/immich-maintenance-mode-redis-error
fix/npm-unmask-openresty-on-migration
fix/ollama-intel-gpg-error-handling
github-action-archive-changelog
add-script-yourls-1774732133
add-script-matter-server-1774638379
fix/dispatcharr-pg-port
cron_update_lxc
chore/immich-v2.6.3
add-script-geopulse-1774548387
cdn_improvements
add-script-birdnet-1774535320
fix/tools-func-exit-codes
fix/immich-update-db-hostname
update/frigate-0.17.1
fix/use-absolute-path-for-install
fix/pin-npm-version
shell_safe_fixes
remove_booklore
chore/update-url-community-scripts
komodov2
refactor/turnkey-modernize
add-script-nextexplorer-1774344421
add-script-homebrew-1774342032
fix/shell-security-hardening-v2
improve/build-func-performance-cleanup
fix/build-dns-prefix
fix/anytype-mongodb-wait
fix/frigate-cpu-model-path
copilot/fix-installation-failure-isponsorblocktv
fix/reactive-resume-add-git
copilot/scanopy-fix-apt-configuration-error
add-script-isponsorblocktv-1774009652
add-script-alpine-wakapi-1774008954
fix/coder-code-server-existing-config-and-reachability
add-script-teleport-1773928044
CrazyWolf13-patch-wealthfolio-1
refactor/tools-func-qol
fix/stirling-pdf-jdk-reinstall
fix/pinned-version-wording
MickLesk-patch-10
fix/reactive-resume-v5013
fix/tracearr-update-version-oom
copilot/fix-hdd-space-for-owncast
tremor021-patch-6
pocketbase_bot
disp_fix
fix/tdarr-binary-check-curl-retry
MickLesk-patch-9
refactor/podman-quadlets
alpine-ntfy
refactor/jellyfin
CrazyWolf13-patch-11
feature/autousermod_hwaccell
add-script-split-pro-1773677692
fix/frigate-openvino-fallback
fix/paperless-ngx-default-ram
fix/plex-restart-after-update
fix/gluetun-openvpn-env
MickLesk-patch-8
termix_add_guacd
MickLesk-patch-7
fix/tududi-nodejs-in-update
fix/sparkyfitness-npm-peer-deps
docs/website-metadata-workflow
fix-pbs_microcode
remove_jsons
michelroegl-brunner-patch-4
add-script-test-1773325265
cleanup_workflows
feat/remove-frontend
automated/update-github-versions
feat/mode-generated
fix/n8n-build-essential
fix/sparkyfitness-shared-deps
fix/rocm-path-escaping
fix/storage-validation-cross-node
fix/frigate-nvidia-version-regex
arm64-build-support
readme
michelroegl-brunner-patch-3
fix/coder-code-server-backup
copilot/fix-immich-update-dependency-issue
rust
fix/linkwarden-update-playwright
fix/powerdns-sqlite-permissions
fix/duplicate-nameserver-searchdomain
CrazyWolf13-patch-7
feat/ollama-rocm-support
fix/seerr-migration-update-script
preflight_tests
adgu_fix
2026-03-30
2026-03-29
2026-03-28
2026-03-27
2026-03-26
2026-03-25
2026-03-24
2026-03-23
2026-03-22
2026-03-21
2026-03-20
2026-03-19
2026-03-18
2026-03-17
2026-03-16
2026-03-15
2026-03-14
2026-03-13
2026-03-12
2026-03-11
2026-03-10
2026-03-09
2026-03-08
2026-03-07
2026-03-06
2026-03-05
2026-03-04
2026-03-03
2026-03-02
2026-03-01
2026-02-28
2026-02-27
2026-02-26
2026-02-25
2026-02-24
2026-02-23
2026-02-22
2026-02-21
2026-02-20
2026-02-19
2026-02-18
2026-02-17
2026-02-16
2026-02-15
2026-02-14
2026-02-13
2026-02-12
2026-02-11
2026-02-10
2026-02-09
2026-02-08
2026-02-07
2026-02-06
2026-02-05
2026-02-04
2026-02-03
2026-02-02
2026-02-01
2026-01-31
2026-01-30
2026-01-29
2026-01-28
2026-01-27
2026-01-26
2026-01-25
2026-01-24
2026-01-23
2026-01-22
2026-01-21
2026-01-20
2026-01-19
2026-01-18
2026-01-17
2026-01-16
2026-01-15
2026-01-14
2026-01-13
2026-01-12
2026-01-11
2026-01-10
2026-01-09
2026-01-08
2026-01-07
2026-01-06
2026-01-05
2026-01-04
2026-01-03
2026-01-02
2026-01-01
2025-12-31
2025-12-30
2025-12-29
2025-12-28
2025-12-27
2025-12-26
2025-12-25
2025-12-24
2025-12-23
2025-12-22
2025-12-21
2025-12-20
2025-12-19
2025-12-18
2025-12-17
2025-12-16
2025-12-15
2025-12-14
2025-12-13
2025-12-12
2025-12-11
2025-12-10
2025-12-09
2025-12-08
2025-12-07
2025-12-06
2025-12-05
2025-12-04
2025-12-03
2025-12-02
2025-12-01
2025-11-30
2025-11-29
2025-11-28
2025-11-27
2025-11-26
2025-11-25
2025-11-24
2025-11-23
2025-11-22
2025-11-21
2025-11-20
2025-11-19
2025-11-18
2025-11-17
2025-11-16
2025-11-15
2025-11-14
2025-11-13
2025-11-12
2025-11-11
2025-11-10
2025-11-09
2025-11-08
2025-11-07
2025-11-06
2025-11-05
2025-11-04
2025-11-03
2025-11-02
2025-11-01
2025-10-31
2025-10-30
2025-10-29
2025-10-28
2025-10-27
2025-10-26
2025-10-25
2025-10-24
2025-10-23
2025-10-22
2025-10-21
2025-10-20
2025-10-19
2025-10-18
2025-10-17
2025-10-16
2025-10-15
2025-10-14
2025-10-13
2025-10-12
2025-10-11
2025-10-10
2025-10-09
2025-10-08
2025-10-07
2025-10-06
2025-10-05
2025-10-04
2025-10-03
2025-10-02
2025-10-01
2025-09-30
2025-09-29
2025-09-28
2025-09-27
2025-09-26
2025-09-25
2025-09-24
2025-09-23
2025-09-22
2025-09-21
2025-09-20
2025-09-19
2025-09-18
2025-09-17
2025-09-16
2025-09-15
2025-09-14
2025-09-13
2025-09-12
2025-09-11
2025-09-10
2025-09-09
2025-09-08
2025-09-07
2025-09-06
2025-09-05
2025-09-04
2025-09-03
2025-09-02
2025-09-01
2025-08-31
2025-08-30
2025-08-29
2025-08-28
2025-08-27
2025-08-26
2025-08-25
2025-08-24
2025-08-23
2025-08-22
2025-08-21
2025-08-20
2025-08-19
2025-08-18
2025-08-17
2025-08-16
2025-08-15
2025-08-14
2025-08-13
2025-08-12
2025-08-11
2025-08-10
2025-08-09
2025-08-08
2025-08-07
2025-08-06
2025-08-05
2025-08-04
2025-08-03
2025-08-02
2025-08-01
2025-07-31
2025-07-30
2025-07-29
2025-07-28
2025-07-27
2025-07-26
2025-07-25
2025-07-24
2025-07-23
2025-07-22
2025-07-21
2025-07-20
2025-07-19
2025-07-18
2025-07-17
2025-07-16
2025-07-15
2025-07-14
2025-07-11
2025-07-10
2025-07-09
2025-07-08
2025-07-07
2025-07-06
2025-07-05
2025-07-04
2025-07-03
2025-07-02
2025-07-01
2025-06-30
2025-06-29
2025-06-28
2025-06-27
2025-06-26
2025-06-25
2025-06-24
2025-06-23
2025-06-22
2025-06-21
2025-06-20
2025-06-19
2025-06-18
2025-06-17
2025-06-16
2025-06-15
2025-06-14
2025-06-13
2025-06-12
2025-06-11
2025-06-10
2025-06-09
2025-06-08
2025-06-07
2025-06-06
2025-06-05
2025-06-04
2025-06-03
2025-06-02
2025-06-01
2025-05-31
2025-05-30
2025-05-29
2025-05-28
2025-05-27
2025-05-26
2025-05-25
2025-05-24
2025-05-23
2025-05-22
2025-05-21
2025-05-20
2025-05-19
2025-05-18
2025-05-17
2025-05-16
2025-05-15
2025-05-14
2025-05-13
2025-05-12
2025-05-11
2025-05-10
2025-05-09
2025-05-08
2025-05-07
2025-05-06
2025-05-05
2025-05-04
2025-05-03
2025-05-02
2025-05-01
2025-04-30
2025-04-29
2025-04-28
2025-04-27
2025-04-26
2025-04-25
2025-04-24
2025-04-23
2025-04-22
2025-04-20
2025-04-21
2025-04-19
2025-04-18
2025-04-17
2025-04-15
2025-04-16
2025-04-14
2025-04-13
2025-04-12
2025-04-11
2025-04-10
2025-04-09
2025-04-08
2025-04-07
2025-04-06
2025-04-05
2025-04-04
2025-04-03
2025-04-02
2025-04-01
2025-03-31
2025-03-30
2025-03-29
2025-03-28
2025-03-27
2025-03-26
2025-03-25
2025-03-24
2025-03-23
2025-03-22
2025-03-21
2025-03-20
2025-03-19
2025-03-18
2025-03-17
2025-03-16
2025-03-15
2025-03-14
2025-03-13
2025-03-12
2025-03-11
2025-03-10
2025-03-09
2025-03-08
2025-03-07
2025-03-06
2025-03-05
2025-03-04
2025-03-03
2025-03-02
2025-03-01
2025-02-28
2025-02-27
2025-02-26
2025-02-25
2025-02-24
2025-02-23
2025-02-21
2025-02-20
2025-02-19
2025-02-18
2025-02-17
2025-02-16
2025-02-15
2025-02-14
2025-02-13
2025-02-12
2025-02-11
2025-02-10
2025-02-09
2025-02-08
2025-02-07
2025-02-06
2025-02-05
2025-02-04
2025-02-03
2025-02-02
2025-02-01
2025-01-31
2025-01-30
2025-01-29
2025-01-28
2025-01-27
2025-01-26
2025-01-24
2025-01-23
2025-01-22
2025-01-21
2025-01-20
2025-01-19
2025-01-18
2025-01-17
2025-01-16
2025-01-15
2025-01-14
2025-01-13
2025-01-11
2025-01-10
2025-01-09
2025-01-08
2025-01-07
2025-01-06
2025-01-05
2025-01-04
2025-01-03
2025-01-02
2025-01-01
2024-12-31
2024-12-30
2024-12-29
2024-12-28
2024-12-27
2024-12-26
2024-12-25
2024-12-23
2024-12-21
2024-12-20
2024-12-19
2024-12-18
2024-12-17
2024-12-16
2024-12-13
2024-12-12
2024-12-09
2024-12-08
2024-12-07
2024-12-06
2024-12-05
2024-12-04
2024-12-03
2024-12-02
2024-11-30
2024-11-29
2024-11-28
2024-11-27
2024-11-26
2024-11-25
2024-11-24
2024-11-23
Labels
Clear labels
🛑 Failure to comply with the guidelines
breaking change
bug
bug
bugfix
deferred
delete script
dependencies
enhancement
external
feature
github
help wanted
Implemented in VED waiting push to Main
in project pipeline
invalid
investigation
json
maintenance
new script
new script
nice to have
not a script issue
not planned
organization
pull-request
question
refactor
rename script
security
update script
website
wontdo
Mirrored from GitHub Pull Request
No Label
bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
saavagebueno
Clear assignees
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: SVI/ProxmoxVE#1343
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @clipz98 on GitHub (Jun 24, 2025).
✅ Have you read and understood the above guidelines?
yes
📜 What is the name of the script you are using?
ErsatzTV
📂 What was the exact command used to execute the script?
update
⚙️ What settings are you using?
🖥️ Which Linux distribution are you using?
Debian 12
📝 Provide a clear and concise description of the issue.
The
updatecommand is unable to safely update ErsatzTV (ETV) the new 25.2.0 release that came our earlier today without breaking my existing instance. The update includes an updated ffmpeg bundle from 6.1 to 7.1.1 (in docker only I'm assuming) and is required for this version to properly work too.🔄 Steps to reproduce the issue.
After creating a snapshot prior to updating with WinSCP open and sshed into the LXC to monitor file changes:
updatecommand in the lxc console verbosely; update was a successupdateagain, but it confirmed that it was on the latest versionupdatecommand; it was rinse and repeat with it not working againopt/ErsatzTVfolder to compare and contrast file modification dates and other possible discrepancies/root/.local/share/ersatztvfolder for discrepancies; I noticed that ersatztv.sqlite3-shm & ersatztv.sqlite3-wal get deleted when theupdatecommand is ran (after my third rollback/test I did later on), but that may be inconsequential since those files are temp files generated while the db is actually properly running?/root/.net/ErsatzTVgets a new associated folder (with naming akin to a hash, e.g.nwoq+SbySr5SzMDQg_C0smTNkAarR34=for this update) each for each new update since there are also older folders corresponding to the last two times I updated (with matching modification dates for those eras), so I'm assuming there's a component of .NET that needs to be updated or referenced with each update?❌ Paste the full error output (if available).
Starting from top to bottom (from when the update script shut down the server to update it), this is part of the log in
/root/.local/share/ersatztv/logsthat's available from the broken updated instance. Everything prior to this is personalized to my instance (showing channels, libraries, etc.), so I've redacted that. These errors do not appear in any of my previous logs from a working prod environment:It mostly repeats, so I'm cutting it off earlier to not break the 65536 character limit
🖼️ Additional context (optional).
I'm by no means a Linux expert still, so I wouldn't be surprised if a fix was staring me in the face. The last time I updated with the
updatecommand was from .88 beta (or whatever pre-release version it was) to 25.1.0 in February. It broke my instance back then too. I was able to fix that by manually copying the ETV executable into the opt/ErsatzTV folder manually (at least I think that's what I did, it was back in Feb and I never documented it unfortunately since I fixed it and moved on). One thing I couldn't verify since the instance was not working was if FFmpeg was updated too. I'm the script accounts for also updating ffmpeg to the latest version since there's no direct upgrade path from 6.1 to 7.1.1 too apparently, so sharing that info JIC.Since the script is saying it's updateable, I'm its safe to assume others will face similar issues too (assuming they didn't change around directories and whatnot) whether they used Advanced or Default settings.
@tremor021 commented on GitHub (Jun 29, 2025):
Failed to bind to address http://[::]:8409: address already in use.indicates that something is keeping port 8409 opened. Its probably malfunctioning instance of ErsatzTV.Kill the process taking up the port. You can check which one it is with:
ss -tulpnYou can also do
systemctl stop ersatzTVand then use above command to check if instance is stopped. This would most likely prevent any issue with the update. If the instance is bugged after stop command, you must kill it or you will have same problem when you run update.Also, continuing on your additional comment about ffmpeg. Script grabs the latest available version at the install time with this command:
curl -fsSL "https://johnvansickle.com/ffmpeg/releases/ffmpeg-release-amd64-static.tar.xz" -o "ffmpeg-release-amd64-static.tar.xz"I don't see that we run a ffmpeg update alongside ErsatzTV. We just update ErsatzTV to the latest version. ffmpeg isn't really one of those softwares that needs constant updating, so i guess it shouldn't make any problems ever.
@clipz98 commented on GitHub (Jun 29, 2025):
Hey thanks for the response. I mentioned previously that I've been able to get it manually working by swapping out the old and new binary folders, but failed to mention that yes, I always stopped the
ersatztvservice and stuff before doing any of that. Doing this by hand allowed me to get it mostly working. For whatever reason, accessing via IP acts normally with all the menus clickable, but via my reverse-proxied url, the menu items aren't clickable (even after clearing cookies). That's probably out-of-scope, so I won't ask that to be troubleshot, but mentioning it for the sake of extra info in case it helps.The update does bring a bit of a change to how the ports are used in that you can now assign the web server admin UI and the streaming server separate ports (it keeps the default 8409 for both though per the changelog). At first I thought it was messing that up when used that
updatescript, but that's not an issue when I manually change binaries.I also brought FFmpeg because this specific version apparently requires 7.1.1 and will complain about earlier versions being out of date. I already analyzed the main ETV script, found the johnvansickle.com site, and learned that his version only goes up to 7.0.2 currently. I only found out about the error message when I manually updated ETV since that's presented on the main page immediately. It's worth noting the ETV author has a repo dedicated to docker ffmpeg builds (that are based on this project), so while I agree ffmpeg shouldn't normally need constant updates in normal circumstances, following the author's recommended version build would probably make sense for compatibility purposes and be considered for the updates like it already is for the first install. I made a test instance with the script from scratch just to share my the error too. It's a bit nicer than the one I got for having 6.x still installed:

For those that may not be as savvy enough to manually update and rely on the mc update feature, that's why I made this post in case it could aid the script devs. If the update script is meant to just focus on the app only without its dependencies, that's my bad since I guess that's a different story.
@tremor021 commented on GitHub (Jun 29, 2025):
If application itself complains about mismatching versions of ffmpeg, then i guess we should update it alongside ETV.
Gonna see if i can do this tomorrow or something
@MickLesk commented on GitHub (Jun 30, 2025):
hm, not so easy to switch to this ffmpeg, because its an completly different file structure then the "original" ffmpeg.
Edit: these files in there repo are windows only...
Maybe this Script get an refactor
@MickLesk commented on GitHub (Jul 8, 2025):
Full refactored
@clipz98 commented on GitHub (Jul 8, 2025):
Thanks for the the work! I snapshotted then ran the
updatecommand and unfortunately ran into this error now.This was originally installed via the script, so I made no major modifications to the OS itself (other than cron-automated updates through the LXC updater script). Going to attempt to see if I can fix the so-called broken packages when I have time, but sharing this for awareness in case anyone else ran into this issue.
@MickLesk commented on GitHub (Jul 8, 2025):
Better to recreate the LXC, because its an old Special ffmpeg Install, this is now the official
@clipz98 commented on GitHub (Jul 8, 2025):
This was a great hint, thanks Mick! Instead of starting over from scratch, what I tried instead was doing an
apt purge ffmpegTHENapt autoremovesince purging alone was not fixing the issue. It cleared off the main libraries with the first purge then cleaned everything else with the autoremove! My update was finally successful after that! I replicated these steps after rolling back snapshot again just to make sure as well.TL;DR for Everyone: Use
apt purge ffmpeg -y && apt autoremove -yis the way to go for anyone attempting to update BEFORE using the built-in mcupdatefunction!@clipz98 commented on GitHub (Jul 8, 2025):
Back for some more reporting now that I've had time to examine the update more:
I seem to be having an issue with the previous vulkan driver that was being used just fine before since it's complaining about "No compatible hardware acceleration kinds are supported by ffmpeg" now. The picture below is my updated instance that was previously using hw acceleration just fine. The FFmpeg Reports warning can be ignored:
For the sake of thorough testing, since @MickLesk suggested just rebuilding the LXC, I created a brand new test instance with the new refactored script. I used advanced settings for networking and stuff and nothing out of scope from a default, unprivileged LXC setup. It asked if I wanted to install the VAAPI driver (which I said "yes" to) and successfully installed everything. I opened the instance and was presented with the same issue again (minus the complaint about the VAAPI driver since I haven't set up any ffmpeg profiles yet):

I noticed in the new lxc that it didn't create any PCI passthrough devices unlike my production ETV instance. Left image is my current one (originally created by the script), and the right pic is the new test instance.
Here are the success reports from both installs (first one is the prod instance I updated directly, second one is the fresh install test instance):
My device is using Intel Xe graphics (it's an MS-01 with the i9-13900H) meaning it supports QuickSync. Has this been ran into by anyone else, specifically with a QSV device?
@MickLesk commented on GitHub (Jul 9, 2025):
whats the output of:
ls -l /dev/dri/
@clipz98 commented on GitHub (Jul 9, 2025):
Here it is:

@MickLesk commented on GitHub (Jul 9, 2025):
you are right. The VAAPI work fine so, but the entry is missing, will be added after: https://github.com/community-scripts/ProxmoxVE/pull/5875 merge
@clipz98 commented on GitHub (Jul 9, 2025):
Oops, I forgot to specify that the first picture was production Ersatz that's been used fine for months (I downgraded back to my snapshot). Here is the output of the test instance I made:
@clipz98 commented on GitHub (Jul 11, 2025):
Hello. I'm going to continue to be a pain with the feedback, so bear with me. I saw the new merge was completed and recreated my test LXC before attempting to update prod again.
Here are my results for the test LXC:
The the HW acceleration problem still exists. The new difference at least is that the PCI devices seem to be passed through correctly on new installs again at least after saying "yes" to VAAPI now that they have the proper UID/GID. Were you not seeing this issue on your end during your testing too @MickLesk? Do you have any QSV devices? Curious in case this is affecting Intel-based devices like mine alone.
This is my prod instance on the previous version:
Here's some info from the test instance in the Troubleshooting tab all the way in the bottom of the Support menu that might be useful:
General:
{ "Version": "v25.2.0-linux-x64", "Environment": {}, "Health": [], "FFmpegSettings": { "FFmpegPath": "/usr/local/bin/ffmpeg", "FFprobePath": "/usr/local/bin/ffprobe", "DefaultFFmpegProfileId": 1, "PreferredAudioLanguageCode": "eng", "UseEmbeddedSubtitles": true, "ExtractEmbeddedSubtitles": false, "SaveReports": false, "HlsSegmenterIdleTimeout": 60, "WorkAheadSegmenterLimit": 1, "InitialSegmentCount": 1, "HlsDirectOutputFormat": "MpegTs" }, "Channels": [ { "Id": 1, "UniqueId": "0788beee-29e4-411d-93a0-f49b0f4d36c0", "Number": "1", "Name": "ErsatzTV", "Group": "ErsatzTV", "FFmpegProfileId": 1, "StreamingMode": "TransportStreamHybrid", "SubtitleMode": "None", "MusicVideoCreditsMode": "None", "SongVideoMode": "Default", "ProgressMode": "Always", "WebEncodedName": "ErsatzTV" } ], "FFmpegProfiles": [ { "Id": 1, "Name": "1920x1080 x264 ac3", "ThreadCount": 0, "HardwareAcceleration": "None", "VaapiDriver": "Default", "QsvExtraHardwareFrames": 64, "ResolutionId": 3, "Resolution": { "Id": 3, "Name": "1920x1080", "IsCustom": false, "Height": 1080, "Width": 1920 }, "ScalingBehavior": "ScaleAndPad", "VideoFormat": "H264", "VideoProfile": "high", "VideoPreset": "", "AllowBFrames": false, "BitDepth": "EightBit", "VideoBitrate": 2000, "VideoBufferSize": 4000, "TonemapAlgorithm": "Linear", "AudioFormat": "Aac", "AudioBitrate": 192, "AudioBufferSize": 384, "NormalizeLoudnessMode": "Off", "AudioChannels": 2, "AudioSampleRate": 48, "NormalizeFramerate": false, "DeinterlaceVideo": true } ] }(Continued Next Post)
@clipz98 commented on GitHub (Jul 11, 2025):
@MickLesk commented on GitHub (Jul 11, 2025):
I think its an Display Error. Try an HW acc Test with ffmpeg. My Work fine, get the Same Message. But HWA Work
@clipz98 commented on GitHub (Jul 11, 2025):
Yeah I think my production instance has a mostly similar message for the VAAPI Capabilities troubleshooting menu if that's what you mean when you say "Same Message". Or did you mean you're getting the same message about HW acceleration from the first page with no compatible device?
How different can HW acceleration pieces change from how it was working just fine pre-refactor? Do you see HW accell options below like this (My working version):
Or this (The test lxc):

I've never done an ffmpeg HW acc test before, but this is what I tried. If there's a better way, let me know. I also did
ffmpeg -codecsandvainfobefore this, but I won't output those here since they looked fine like my prod LXC.Test (on 7.1.1)

Prod (on old 6.x Version & previous ETV)

@clipz98 commented on GitHub (Jul 15, 2025):
Nothing? Alright, perhaps I'll just migrate this to a docker setup instead if there's not a reliable way to upgrade future releases of the lxc I guess. HW acceleration is more or less the core of this, but not it's functional (meaning ffmpeg profiles for channels cannot be created at all) even though it previously was working fine.
@gb330033 commented on GitHub (Jul 15, 2025):
I'll go ahead and chime in here and say that I was banging my head against the wall trying to get hardware acceleration on my 13th gen Intel cpu (QSV and VAAPI) to work with ErsatzTV and it ended up being an issue with the ffmpeg build used by this refactored script.
I had never used ErsatzTV before, but apparently this refactor has made the installation use a version of ffmpeg that does not contain the necessary bits for hardware acceleration to work. I manually replaced this version of ffmpeg with the one built and maintained by the ErsatzTV dev (Referenced earlier in this thread - https://github.com/ErsatzTV/ErsatzTV-ffmpeg/releases) and hardware acceleration worked flawlessly.
In short, this refactored script is now broken in terms of using hardware acceleration with ErsatzTV. The build of ffmpeg being installed needs to be changed/updated to something that actually supports hardware acceleration.
@MickLesk commented on GitHub (Jul 15, 2025):
This is the newest patch of ffmpeg 7.1.1:
root@ersatztv:~# ffmpeg -hwaccels
ffmpeg version 7.1.1 Copyright (c) 2000-2025 the FFmpeg developers
built with gcc 12 (Debian 12.2.0-14+deb12u1)
configuration: --enable-gpl --enable-shared --enable-nonfree --disable-static --enable-libx264 --enable-libvpx --enable-libmp3lame --enable-libfreetype --enable-libass --enable-libopus --enable-libvorbis --enable-libx265 --enable-libdav1d --enable-zlib --enable-vaapi --enable-libdrm
libavutil 59. 39.100 / 59. 39.100
libavcodec 61. 19.101 / 61. 19.101
libavformat 61. 7.100 / 61. 7.100
libavdevice 61. 3.100 / 61. 3.100
libavfilter 10. 4.100 / 10. 4.100
libswscale 8. 3.100 / 8. 3.100
libswresample 5. 3.100 / 5. 3.100
libpostproc 58. 3.100 / 58. 3.100
Hardware acceleration methods:
vaapi
drm
Maybe i remove the whole script. The ffmpeg work fine, only ErsatzTV bullshiting this.
@jasongdove commented on GitHub (Jul 15, 2025):
Historically, I have only bundled ffmpeg with the Windows release of ETV - not the Linux release. With that said, it's much easier when everyone is on a consistent version, so it's my goal to host a compatible Linux build of ffmpeg (with all required options) at https://github.com/ErsatzTV/ErsatzTV-ffmpeg/releases going forward.
Right now there is an issue where the BtbN FFmpeg-Builds that I normally recommend require
libva2.21+ when some major distros only provide 2.20, so this specific build is compiled against the olderlibvahttps://github.com/ErsatzTV/ErsatzTV-ffmpeg/releases/download/7.1.1/ffmpeg-n7.1.1-55-gca5c0a959d-linux64-nonfree-7.1.tar.xz but I am waiting for confirmation from the user who surfaced the issue.@MickLesk commented on GitHub (Jul 15, 2025):
Thats new? 2 weeks ago only a Windows File exist?
@gb330033 commented on GitHub (Jul 15, 2025):
I can confirm that this update to the script (a "full" install of ffmpeg, instead of "medium") resulted in VAAPI working without any additional configuration needed in a brand new unprivileged LXC
@jasongdove commented on GitHub (Jul 15, 2025):
Yes, this is new - this is the first release where I am also hosting a Linux build of ffmpeg. As I said, I plan on hosting a compatible version for all releases going forward.
@MickLesk commented on GitHub (Jul 15, 2025):
Then i can build in this and the users can test, maybe in 2h or tomorrow
@clipz98 commented on GitHub (Jul 15, 2025):
Thank you for chiming in @jasongdove. I was very tempted to engage you earlier in the thread honestly, but figured your time was better spent not troubleshooting unofficial installs that were not directly referenced in the ETV's docs. Definitely didn't see the Linux-ready ffmpeg files in that repo last week either, so hosting the compatible Linux builds along with the Windows one will definitely benefit projects like this in the long run.
Thanks @gb330033 too for making me feel less crazy about things not working--I did a shit ton of tests before replying my findings each time.
Going to see if I'll have time later today to manually replace the ffmpeg build with that linked one like @gb330033 did. Otherwise, If @MickLesk finishes their revisions before I can get back to it, I'll test theirs and report back.
@MickLesk commented on GitHub (Jul 15, 2025):
ive build an script, but i get following issue:
The following channels use ffmpeg profiles that are not configured for hardware acceleration (Nvenc, Vaapi, Qsv): 1 - ErsatzTV
i deferr it until tomorrow
@jasongdove commented on GitHub (Jul 15, 2025):
This means ffmpeg reports hardware acceleration support (nvidia, vaapi, qsv) - but the FFmpeg Profile in ETV has not been configured to use one of the available accels. This is something users of ETV have to configure - it doesn't automatically enable accel.
@MickLesk commented on GitHub (Jul 16, 2025):
"Each user should configure this", is this documented somewhere so that I can add it to our Documentation Button or "note" on website?
I'm not a hardcore user for something like this, I'd just click it together and hope it works 😄
@clipz98 commented on GitHub (Jul 16, 2025):
@jasongdove was referring to something similar to my screenshot from 5 days ago where I shared my prod vs. the refactored LXC screenshots of the FFmpeg profile selections like this:
People that use ETV should be familiar with the setting, but here's a very brief nod to the setting in the "Create Channels" section. I guess it's technically more of an advanced thing to create profiles outside of the default profile that's preinstalled (if you know what you're doing).
@jasongdove commented on GitHub (Jul 16, 2025):
Please use https://github.com/ErsatzTV/ErsatzTV-ffmpeg/releases/download/7.1.1/ffmpeg-n7.1.1-56-gc2184b65d2-linux64-gpl-7.1.tar.xz - I accidentally built a non-free version yesterday. Assuming it tests correctly, I will leave this version forever and make a new release when a new ffmpeg is required.
@clipz98 commented on GitHub (Jul 16, 2025):
ffmpeg -hwaccelsand it came up empty (expected since I don't think Mick pushed anything out new yet)@jasongdove commented on GitHub (Jul 16, 2025):
Thanks for the confirmation - I will leave those artifacts alone now, and probably update ETV docs to link to them.