as I can't delete my account here at my own (at least not found)
Foren admin do it for me and drop me a mail afterwards
bug reports I opened (and if they are still open) are now *your* bugs.
I'm done with it !
good night and good luck ![]()
as I can't delete my account here at my own (at least not found)
Foren admin do it for me and drop me a mail afterwards
bug reports I opened (and if they are still open) are now *your* bugs.
I'm done with it !
good night and good luck ![]()
too early
@reardonia reopen this bug
waiting for the fix hitting LE ...
P.S.
for audience: link to the bug is in comment #15
some more action ...
Question:
where is the stuff under /storage/.kodi/addons/service.tvheadend43/dvb-scan/dvb-s from
TvH or LE ?
===
if I remove Astra-1M I get all of my channels, but with hickups: it hangs for a long time at ~43 %. Never seen this last time (~1 or 2 year(s) ago).
why did I do that ?
to me it seems (debugged) that only Astra-1M is loaded, but Astra-1M is dead. even LyngSat doesn't list it anymore.
1M contains 16 entries, what matches the 15 in the above screenshot.
also in the scan IF before the scan has started: with 1M only 15 muxes, without a lot more and all services afterwards
for 1N and 1P it seems there are more or less (cosmetic ???) mistakes:
- Labels e.g. [Astra TV] and the like
- no Modulation e.g. [Deutschlandfunk, CHANNEL]
don't know if the last one is really needed, but I mean scan is speedier when they are there.
too make friends:
I offer to fixe 1N, 1P and attach both files when ready or do you own some tools to do it quicker then "visual grep" and hand editing ?
P.S.
scan issue can only be seens with a clean TvH, on an complete setup a re-scan doesn't show this issue !
for the record:
I had my guisettings.xml full of entries like this one
<resolution>
<description>3840x2160 @ 50.000000 Hz</description>
<subtitles>1080</subtitles>
<pixelratio>1.000000</pixelratio>
<overscan>
<left>0</left>
<top>0</top>
<right>1920</right>
<bottom>1080</bottom>
</overscan>
</resolution>
Display More
and that was it what caused the trouble. No idea how they came in ...
anyway, now it's resolved !
I've read that...
I survived by exchaning the TvH directory from my other setup and re-scanned the channels, cause the EPG stayed empty.
anyway, cause I couldn't reproduce my main issue (quartered display runnning 3840x2160 on 50 Hz, you remember ?) I plan to iron my main disk/setup too.
if the above Tvh issue occurs there too (probably, cause both are the same nightly) what do you suggest:
open a bug report at TvH ?
+++ Edit +++
no need to do a fresh install.
just exchanged a faulty guisettings.xml to fix my quartered display bug.
as the test disk still exists, I could help to nail the TvH scan bug.
I need to set up an clean install with nightly to reproduce an issue/bug.
I'm unable to find any muxes with last TVH and maybe with new scan tables.
My environment is an Digibit R1 with 2x Sat-cable.
Scanning on it's Web-IF => all is fine: 308 x TV, 132 x Radio (FTA only)
scanning with TVH immediately jumps to 15 found muxes but nothing to map.
scanned frequencies seems to be un-usual (see attached Muxes.png). Nothing where I usually find my TV channels.
tried a download/update of the scan tables within TVH settings, though.
TVH IF-setup is pretty default
EPG language: German
astra 19.2
I tried several times with with sequence:
1. systemctl stop service.tvheadend43.service
2. delete /storage/.kodi/userdata/addon_data/service.tvheadend43
3. systemctl start service.tvheadend43.service
as said it's an test environment with last nightly installed on an addional disk.
the environment on the other disk (also last nightly with the same(?) TVH) is okay.
so I guess something wrong with the scan tables ...
esp. see the frequencies in the attached Muxes.png and 13.tlist.txt (homebrewed-who-needs-bibleTV- FTA-scan list for my DigiBit from 2025.11.01)
- I got the same issues with the original Digibit-scan-list (before someone complains ...) -
I tried to find some of the frequencies (from muxes.png) at https://www.lyngsat.com/Astra-1N-1P.html
nothing !
and it seems TVH scans only for that non-existent 15 (see service.log)
kodi.log is clean
ls -lR /storage/.config/system.d/
/storage/.config/system.d/:
total 56
-rw-r--r-- 1 root root 28816 Jan 5 2025 README
-rw-r--r-- 1 root root 1781 Jun 22 11:58 cifs.mount.sample
drwxr-xr-x 2 root root 4096 Jun 22 23:02 kodi.target.wants
-rw-r--r-- 1 root root 1763 Jun 22 11:58 nfs.mount.sample
-rw-r--r-- 1 root root 1089 Jun 22 11:58 openvpn.service.sample
lrwxrwxrwx 1 root root 79 Jun 22 23:02 service.tvheadend43.service -> /storage/.kodi/addons/service.tvheadend43//system.d/service.tvheadend43.service
lrwxrwxrwx 1 root root 53 Jun 22 23:02 tvheadend.service -> /storage/.config/system.d/service.tvheadend43.service
-rw-r--r-- 1 root root 421 Jun 22 12:35 wireguard.service.sample
/storage/.config/system.d/kodi.target.wants:
total 4
lrwxrwxrwx 1 root root 79 Jun 22 23:02 service.tvheadend43.service -> /storage/.kodi/addons/service.tvheadend43//system.d/service.tvheadend43.service
tried to de-install/install TvH (without reboot), but got the same with last nightly
aha, I wasn't aware of the report.
well, it is like it is.
===
alas I can't mark this thread as resolved
I'm aware Xe is declared as experimental.
Thanks for support, but ...
my LE13 Nuc has a uptime of 100 h and most of the time it was running Xe.
Also my desptop box with Intel Corporation RocketLake-S GT1 [UHD Graphics 730] is running Xe since moons, through all kernels (I'm one of the Greg KH stable kernel tester). I'm on 7.1.0-rc7.
Both boxes only have Gen1 Xe GPU's und I've never seen bugs.
for the nuc I reported 1 or 2 issues regrading Xe, what -IIRC - have been config issues.
the 7.0.8 kernel for LE is out since some moons and was running flowlessly until yesterday.
===
we should wait until heitbaum will release kernel 7.1
===
heitbaum care to reproduce my finding with your similar NUC12 ?
since yesterday I see a bug (?) here.
just me ?
to produce:
1. run nightly-20260612-5c96e6
2. with xe-gpu-driver (i915 is fine)
3. start foot
4. to end foot type "exit"
since yesterday I see:
- foot exists
- Kodi desktop just pop up for a second
- screen becomes and stays black
via ssh I see (dmesg):
nuc kernel: xe 0000:00:02.0: [drm] *ERROR* [CRTC:149:pipe A][PLANE:33:plane 1A] fault (CTL=0x94001000, SURF=0x4000000, SURFLIVE=0x4000000)
log:
https://paste.libreelec.tv/parklike-brendan.log
addional info:
does not happen with i915-gpu-driver
no fix with previous nightly
box got a new bios yesterday, but the release notes do not tell something gpu related has changed.
otherwise (unsure) I would see this with i915-gpu-driver too.
I own an "AirMouse Orbsmart AM-1" which seems to be the same.
versus
Rii Mini i25 Wireless (QWERTZ) [RT-MWK25 DE]
===
when you're living in germany you might think about to get my one ?
specs:
german QWERTZ-Layout
bought July 2019, price: 39,9 €
Since 2023 unused, cause of "NO backlit keyboard" (=> moved to an Rii_Mini_i8+, Nov. 2023)
has only 24 load cycles (accu lasts ~10 weeks)
looks like new (could send a photo)
was used 1569 days
No defects
comes with original packaging and all what was in the box (dongle, charge cable), Invoice
let say: 20 € include delivery (DHL, in germany only )
1 month Return (but sending costs are then yours, usually 2.9 €)
I'm not aware of LE is writing to NVRAM.
What I could image:
when you unplug both other drive (recommendeted !) during install of LE on an 3rd drive your drive order in the bios could change.
Take a photo before unplugging for adjustments afterwards.
unfortunately you said nothing about how you boot debian/win
debian's grub chainloads win ?
- I'm long time out of the windows world, but what I remember: Win is picky mixing drive order, so I would (don't know if really necessary) also remember the sata/nvme (?) ports where the drives are now plugged -
after LE install you could plug in both drives again and (maybe) adjust the boot order in the bios (your photo !).
both other OS'ses should work as usual, but your LE will probably nowhere show up in boot menues (as least during first boot).
I guess debian should pick up your LE as an bootable OS.
if not: you ever could start LE (and usually the other OS'ses) via Bios boot menue key (see manufactur manual, something like F8, F11, ...)
===
if you're running with 3 drives:
and boot LE, LE will automount both (?) other drives to /var/media/ (unsure). No harm, LE operates in /storage. But don't blindly write data to /var/media/
if you boot win11 it will probably write a directory named "System Volume Information" to your LE drive in the directory /flash . No harm !
win11 will see from the LE drive /flash (fat filesystem, accessible, LE boot partition !) and /storage (ext filesystem, un-accessible and therefore offers to format it. Danger !!!)
...
Using D2P (as you were) is wrong for an Intel GPU device, it should use EGL. It's unclear if you were also trying to use DRMPRIME paths but that would also be wrong.
I've seen reardonia has tested it on Intel (N100 or so => somewhere under his patch sets/changes) so I tried it to.
You told this is wrong, so I changed it.
...
Running a 1080p desktop normally means the GUI looks sharper and the HTPC device uses less CPU/IO resources resulting in smoother navigation.
that's my main attempt too (lower resources/temp (!)) and fullfill what manufactur recommands (3840@2160@60)
...
You have confirmed that different and better configuration works. However you are unhappy at having a different configuration.
This *is* not (complete) the case (see comment #12)
again: just take a look at the photos (comment #10, #12 => "Video Resolution"
#12: manufactur recommendation hurt.
it was working before. I mean the 3840@2160@50, what -btw- the monitor offers by design...
(temp.) vanished ...
"You can please some of the people some of the time, not all the people all the time"
this works on both ends of us !
This sentence initial goes:
"You can fool all people some of the time, and some of the people all the time, but you cannot fool all the people all the time."
(Abraham Lincoln)
where I translate "fool" to german "belügen" oder härter "verarschen".
*NOT* my Intension !!!
P.S.
we should just wait what reardonia will tell ...
- yes I've read from him something like: GUI only 1080, but up to now I can't believe he is fine with "a monitor can't anymore run a resolution it is design for" -
well, I wasn't satisfied with what was said here.
I didn't know that you're on XBMC too.
I'm not after "winning new friends" and thought to post the link to the bug report here too, but was otherwise involved.
I'm after getting a fix for an - IMO - regression.
Clearly displayed in the photo in comment #10: in an quarter of an 4K monitor it says this quarter *is* 4K ...
well, I was slightly aware of the possibility "getting new friends" (but I referenced this discussion in the bug report, too, keeping it not under the radar somehow)
I hope to shift this, if you realize that I'm trying to get -IMO- bugs/regressions fixed [and as long until they are], not for me only then the community (before they getting aware of them, when nightly becomes release) .
Okay ?
P.S.
don't shift reports here to "erledigt" until the user is telling so or isn't convinced it's "erledigt".
P.P.S.
I'm following the git's (do not understand all, though) and I'm aware that something has changed since 18.05.2026 and the change might not be finished yet.
My obvious change was that I could do something before that date, but couldn't do afterwards => regression.
anyway, it's now there where the main developer is acting ...
well, first of all: I'm absolutly no expert and therfore I get confused with what is displayed pressing o-key => video resolution
there are always physically 4K pixels in the screen
how can I find that out ?
with your settings from comment #7 I get a full screen and the video resolution (o-key) says it is 1920x1080 => 2K (?)
edit: wiki tells 1920x1080 is full HD
see photo