Leaderboard

It's not only about remote access or not. The whole api key thing needs a granular control similar to user accounts - server management, item deletion, library access, remote access. That it needs work was even acknowledged by the devs multiple times over several years. But as is usual in the Emby world, that's all we get - at most an acknowledgment of the problem, but not a solution.

Hello all, Here of late, I have found myself with an abundance of free time on my hands, and up to now I've been selfishly spending that time almost exclusively playing video games, but even that has grown tiresome so I started thinking of other ways I could occupy myself and the idea hit me one day while digging through old files on my computer, when I found a small collection of Icons I started developing years ago, but I never released publicly because of reasons not worth getting into, that I might like to revive the old project. Then I thought about all the people in this community that have managed to keep my old projects alive, and wondered if after all this time, would there be any interest in a new style being developed? I haven't used any kind of media manager (including emby) for a very long time so I don't really know what sort of styling is being used these days and if my old design would even fit in to the modern UI. So that's where you come in, because I'd like to know if there would be any desire for a new set such as these, and if I were to pick up where I had left off building up this collection, will you guys even use them? Here is a link to download the set so far, for you to play around with to determine if this something you'd enjoy and like to see more of. If you do load these up into emby, please share a screen shot so I too can see how they look in the modern UI! My vision for this new collection is to use the highest quality images possible when creating the new disc's, and to try to not use true disc labels that are covered in badges and unwanted clutter. So these won't be an accurate representation of the disc's in your IRL collections but instead a more aesthetic representation of what I think they should look like. That's about all I have to say on the matter, so go try them out, see how they feel, and then come back and tell me all about it!

Yep - I allude to that above Agree 100% - and tbh why I copied in softworkz as no doubt you are aware that he was leading the previous security incident investigation. If api keys are being scraped from the forums - then I expect to see many more posts on this issue. Not allowing any manually created API keys remote access seems to be a sensible first step .. If system api keys are out there, then emby have a problem ...

Yep, there's some nice workarounds and some nice lateral thinking, it would just be really nice if it wasn't necessary...

Here's something I found out recently, and others might find it helpful: I'm using an old home theater system, a Samsung HT-Z220 https://www.samsung.com/ph/support/model/HT-Z220T/XTC/ which came very cheap with the 32" TV (also a Samsung) from around 2008. I haven't used the DVD player for some time, but it works as a sound system. I connected a Xiaomi TV Box S: https://www.mi.com/global/product/xiaomi-tv-box-s-2nd-gen/ to it and the TV. In the TV box, I set the audio to S/PDIF, with an optical cable from it to the theater system. For the software, I used Kodi in an Android OS and the Emby add-on, and they access videos in a local Windows 10 i5 PC in the same network. The Kodi audio setting was set to 5.1-channel, with upmixing, and passthrough for everything that the player could handle. Videos with 5.1-channel aac audio played fine but the audio was never that strong. When I tested the system using a 5.1-channel test video, sound from the, say, left sat speaker was loud when the sound test was played on it, but I could still hear the test sound slightly when the left inner speaker was tested. I discovered that the Emby Android app could be set to transcode all audio to Dolby, so when I went to the Emby add-on in Kodi and configured it transcode all audio to ac3. the audio suddenly became louder and in the test video separation of sound per speaker much clearer. You might want to try this if the audio isn't that clear and the server can handle transcoding.

Having movies shown both in any collection AND individually was never supported, when "Group items into collections" was selected, regardless is it Decades Collection or any other franchise collection - it was always or. Edit: FWIW I never keep that grouping, as Collections tab is just a click away and Decades collections are listed first in the list, so: individual movies in main movies library, Decades collections a tab/single click away.

I just went to check this myself, and have found the one and only API key I've created, in plaintext within my anonymized log file. The key is used for ombi, and is seen in a line beginning with: Info Server: http/1.1 POST http://[local ip manually redacted]:8096/emby/users/authenticatebyname (I logged into ombi using an emby user, then triggered a sync between ombi+emby to ensure some activity before grabbing this log) The file is definitely supposed to be anonymized as most lines are something like: Info Server: http/1.1 Response 401 to host3. Time: 0ms. POST http://emby_remote_ip/Sync/data So potentially every log file on this forum, anonymized or not, contains keys into people's servers... (at least those that have api keys in use) Those should probably be removed as well.

What about the other items - should user created api remote access be allowed by default ? @softworkz- Any thoughts here ?

Just put the see all option back! @luke How many of these post do we need to see in the forum before you see removing it was a bad idea? Not everyone likes to scroll horizontally for everything. When you clicked on the see all option button it displayed everything in a very useful vertical grid, which made things easy to see.

Just need to add list view links back as episode and Trailers have them but not Movie & Shows since you combine them or individual links per Movies and Shows. http://localhost:8095/web/index.html#!/list/list.html?type=Movie,Series&personId={id}&serverId={serverid}}

I'm glad I saw this before allowing the update, else I'd be just as annoyed as rexerm. The changelog note "group collection items by media type" really doesn't do justice to the carnage that's been done here. It's really disappointing that breaking changes to how we view our collections have been forced on us seemingly without consideration to how people might use the software in ways that aren't specifically the same as the developers expect. 1) It's incredibly useful to be able to mix content types. 2) Forced horizontal scrolling is a total blight! You've mentioned that "an option to control this is certainly possible", but you don't say that you're actually going to restore the original functionality, or when. Could you please give us some certainty here?

Far from it. I agree a database of intros would not be of great value for the reason you stated. But I use the intro skip all the time as it is now and it works great. And it is one single button click that gets me to the end of the intro, not "somewhere near the end of the intro" which is what that 10 sec advance would get you. And some series would require more than one 10 second advance to even get there.

I'm really sorry but offering camera upload and having them exposed by this vulnerability (for years now!) is simply ridiculous. People should be informed when activating the camera upload functionality or even better, get this fixed eventually!

Create a new topic under respective forum section and describe your issue in detail, as it appears unrelated to broken playback/files not being probed, this ain't 4.8.4.0 general issues thread but very specific issue with certain platforms (Linux-based).

@Lukewhy not use the schedule to pre call the upcoming and save it into the Cache/DB. it is not that the upcoming changes every second to call the TVDB on the fly. for someone who has a very large media, this could take forever.

Welcome back Demon!

Working for movies ..

I may do a little experimentation today and add this into the MediaInfo plugin to simply add the date as a decade tag. At least you can then filter by decade if you wish (or combine with other tags/genres.) But agree 100% with the sentiment above - how hard could this be to to add into the date filter - that is afterall where you would naturally goto to use such a filter ..

Well, 6 months on from me implementing "Decade Smart Playlists" for music, which are working very well for listening to Songs by Decade, I've still been trying to figure out a nice way for easily browsing Albums by Decade. Sure, using a competitor system is an option, but waiting for another half a decade isn't...! My current Emby method is to use Genres. Yes, I know that a Decade is not a 'technically correct' Genre, but it is an available field within Emby that can be relatively easily imported, from embedded music metadata, at a Song/Album/Artist level. (Tags cannot yet achieve this, as far as I know?). These are my current "Decade Genres": To create these manually for a large library would be extremely painful, but it can be done semi-automatically by using MP3tag scripting functions: E.g.: Format value “GENRE”: $if($and($geql(%year%,1950),$leql(%year%,1959)),%genre%; #1950s,%genre%) Format value “GENRE”: $if($and($geql(%year%,1960),$leql(%year%,1969)),%genre%; #1960s,%genre%) Format value “GENRE”: $if($and($geql(%year%,1970),$leql(%year%,1979)),%genre%; #1970s,%genre%) Format value “GENRE”: $if($and($geql(%year%,1980),$leql(%year%,1989)),%genre%; #1980s,%genre%) Format value “GENRE”: $if($and($geql(%year%,1990),$leql(%year%,1999)),%genre%; #1990s,%genre%) Format value “GENRE”: $if($and($geql(%year%,2000),$leql(%year%,2009)),%genre%; #2000s,%genre%) Format value “GENRE”: $if($and($geql(%year%,2010),$leql(%year%,2019)),%genre%; #2010s,%genre%) Format value “GENRE”: $if($and($geql(%year%,2020),$leql(%year%,2029)),%genre%; #2020s,%genre%) For the non-coders (like me), these Genres were created by combining the following 4 boolean functions $if(x,y,z) if x is true, y is returned, otherwise z. $and(x,y) returns true if x and y are true. $geql(x,y) returns true if x is greater than or equal to y. $leql(x,y) returns true if x is less than or equal to y. I expect that any coders reading this could do this more elegantly? The Genre field(s) also appears in convenient locations near the top of the music pages, without needing any CSS manipulation: Lastly, by using color-coded emoji icons, it's possible to group the Genres so they appear in the grouping and order that you want (I'm still experimenting with this). If (when?) Emby implement RoonAlbumTag importing, I'll probably try out Decades as Tags, although they'll then be near the bottom of the pages. It would be great if the talented Emby developers could nicely integrate some useful built-in Decade options for Music/Movies/etc. - perhaps one day?!?!

Hi, yes we do plan to address this. Thanks for reporting.

I know we already have it as a feature when you resume a show after stopping it but waaaay back in the day when I had a Tivo if you paused a program to say go flip the laundry or answer a phone call when you unpaused it would jump back 5 seconds ( could have been more it was a long time ago) which was nice because it was like a bit of a refresher. That is a feature I miss and would be cool if it emby could adopt something similar.

I updated my emby server this morning to the most recent version and now when I click to go into the page for a specific person where their movies and shows section is no longer has a "view all" button as before and now lists everything in a horizontally scrolling list and I want to know if there is anywhere in the settings to revert this change back to before so you could bring up their complete list of entries in a page that looked similar to the collections and could be sorted and filtered like any other. The current one is really unwieldy and difficult to use for people who have a large filmography or if you want to get certain information at a glance (like their oldest film or most recent one). I took a cursory glance in the settings but I can't see anything that would affect this so I'm hoping I just missed it and not that its been removed entirely in favor or a less customizable system.

Hence (I thought) effectively having movies showing up in two places, in a collection and individually. Though you're right, maybe I was somehow expecting different treatment for movies depending on whether they end up in normal Indiana Jones type collections vs Decade collections. So yes applying a filter as you described does sound much better.

Actually, I got the NAS recovered... I'll start up Emby shortly and report back if I find anything further. Thank you.

Yea I realised that we had never actually catered for Itemtype=audio in the MediaInfo plugin - only video (shows,movies). So I'ne now added it as a library type that you can choose - but of course all the codec types are different vs video - so it's a fair bit of work. I'll add the tagging bit first (tagging 'FLAC at both an Album and Track level - so you can sort all your FLAC Albums for example. Currently Music/Audio seems VERY light on filterable metadata ..

Hey @rbjtechgreat work with what you have achieved for movies in a very short time. I get a similar filter pop-up window with what I implemented via Genres so far: You beat me to my next comment... This is still the limitation with using either Genres or Tags. I can select one OR more decades to filter by, but I cannot select (say?) #1970s AND Classic Rock. Ideally Decades should be their own filter or within the existing year filter because you could still OR decades and years, plus it would be much less likely that you would ever want to AND them? My next tentative idea is to repurpose the Music Studio filter because I already use another (workaround) method for viewing selected Record Labels... If you do get around with doing anything with music, I'll be interested to see what you come up with, if not, no worries at all!

ah, good to know. u.u I don't frequent the forums much only when something goes wrong or need tech support. Glad i'm not the only one complaining tho. Hopefully it can be updated, it incredibly unuserfriendly.

Introskip/ chromaprint as a technology works very well. Yes it's not 100% accurate, but both the Plugin and to some extent the core implementation are certainly good enough. I have mine set to auto-skip - it's that good, and it's usually within 0.5 secs of the content start before playing - that's pretty darn good in my book. If you do have a rogue entry, then I use the Plugin tools to edit it, or use ChapterAPI for the core. btw - an online database of intro's works very well, but not in the way you are thinking. You don't actually hold the 'timings' - as that would obviously not work - what you hold is the 'chromaprint fingerprint' of the intro - that way, you do not need to create it each time - you just need to 'detect' the intro on your version of the file, using the online fingerprint. It works perfectly - but it needed support from the community to upload the fingerprints - which it never got. It's actually incorporated into the Chaper API plugin - or was.

Only time we have really seen this is library path overlap. That is why assigned folder paths per library is requested.

I was going to start a new thread, but wanted to let people be aware of a change to the fail2ban config required based on my setup. The logs now don't look identical to the 2019 view that was originally at the start of this thread, which means that the regex created does not work any longer. The second bit, and this could be 1 of 2 reasons: 1) I am running emby, via the linuxserver docker and 2) running in docker, either way on my setup the logs have some strange characters (half width chracters). After some playing around and trying to get myself banned (mobile into my own network) I have found the following reports the correct host in my case: [INCLUDES] before = common.conf [Definition] failregex = ^.* Info Server\: http\/1\.1 Response 403 to \u200c\u200d\u200d<HOST>\u200c\..*$ Note the \u200c\u200d \u200c means : ZERO WIDTH NON-JOINER \u200d means : ZERO WIDTH JOINER These characters appear to break <HOST> in fail2ban, but luckily are set as visable (\u200c is not in the log file, but is parsed in fail2ban) Hope this helps someone else. Cheers

You guys are doing great. For me, Emby has been working great and with very very very few issues for several years. From that perspective, the current release cadence with a sufficiently long QA period seems to work well. Be wary of growing the team; more coders require more revenue and that may require bigger adjustments to the business model than discussed elsewhere in the forum.

It's all done via the mediainfo plugin - so emby c#/api libraries - so you can't really 'break it' with tags - they are just useful to fill in the gaping holes with the current emby filter system. If they every get around to adding the 'AND' ability (pretty standard since the introduction of binary logic in 1847 (!), then tags will be even more useful...

So API keys obviously don't need passwords - that's the entire point - but of course allowing remote access via an API is a big no-no in the first place. @Lukeimo - API keys need a bit of work - a) are api-keys not obfuscated in the logs ? b) we should have the ability to disable an individual key - currently you can only delete it. c) remote access via manually created keys (usually created manually for local api access to 3rd party apps only) should not have remote access granted. You should not have to rely on nginx to block the request. Maybe you need to do the same as you did for users/passwords and have a 'remote access' column to grant permissions ? So were the previous API keys published in logs ? 2024-04-20 02:57:52.663 Info Server: http/1.1 POST https://‌SERVER_URL.com‌/Users/New?api_key=‌18ff6e1756264241a73b659936a04f91‌. UserAgent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36 X-Middleton/1 The log above is showing the api_key in plain text ... If the server URL was known - then a remote request to setup a new user is a simple thing to do ..

Working for TV Shows as well - it tags the decade in the episodes (of limited use), but more importantly, tags the show as well (using season 1 episode 1 air date as the decade). This allows you to simply filter all the 80's TV shows - for example, using the standard filters. I'll sort the music/audio out another day - I'll tag the tracks and the albums, again so you can easily list all your 80's albums for example.

Hi, yes the functions need to be added to TV mode. Thanks.

Yes, it's a premiere feature.

Hi, we are looking into this. Thanks for reporting.

Thanks for the ideas all. I've been trying all sorts of things, but keep coming back to Kodi because it just works (especially now that v21 supports Dolby Vision in MKV files). I'm moving house soon, so my networking set up will be entirely different, so I might give Emby another try then. I'll report back if I'm able to figure out what my issue is.

Version 1.7.8 came out on 4/9. SamES said it would be fixed in the next version on 4/11. My guess the fix will be in 1.7.9.

guys i have installed 1.7.8 and it is til the same as before with the flac audio transcoding when it does not need to.

Which server version? Which platform? If 4.8.4.0, there have been some issues with Linux-based systems with files not being probed, you can download and install replacement 4.8.4.0 build and see how does that compare.

I have reinstalled the update, now able to play live tv and recorded files

I have just manually loaded the beta: "emby-server-synology72_4.9.0.14_x86_64.spk" onto my Synology NAS (DSM 7.2.1-69057 Update 4). I tested with an ANSI playlist generated in Windows, and can confirm that all accented characters in a test .m3u file now work fine! Great job - thanks!

Got it (most of it) had to change parts of the plaqn but i got it Showed it didnt work out as long as the server was installed on my NAS (4K streaming keeped stopping) so I installed on own server on my HTPC Fun Fact for power dvd to work i had to check the internet stream box in external player - all other choices i ended up with the internal player. Still on to do list: Emby theater needs to be started as admin so every time i geht the Windows dialog to allow .. have to find out how to stop this just for emby theater Sometimes Power dvd start replay on the small screen and then emby theater screen ist spilt between the two screens. Will report back after i got those 2 done Greetings Omni

Don't worry about the next/previous bull, I'm not sure why that's there.. maybe to preview what your about to change? Just scroll down to the tags section and add your tag and select (add to) or what ever you want to do. Edit: Don't forget to add your tag to the textbox.

For context today is the 20th and my example is yesterday the 19th only showing via Upcoming as shown above. Same example different test server. Where is episode 9 (from the 19 as it should be Missing) as it only appears if I enable Upcoming? Another example of yesterday not showing as missing.

Oh no not at all. You aren't going to get high capacity with 2.5" drives. Believe me when I say parents can make some absurd show requests which will take up a tonne of space, all because they "might" want to watch it. Then when they do want to watch it, they will thank you and start requesting more oddball stuff, and you will quickly run out of space. I have family members that love British cop dramas, and they have not only a lot of those, but they have a lot of seasons!

And most people use pirated media of different rippers so intro detection becomes pointless for most. Far easier to just press fwd for the 10s skip than dealing with skip intro and the multi clicks. Chapters would work since you modify the file but you would need a plugin to read the chapters look for intro or what not and then replace the timestamps of the media item.

Ich denke dass du mit 32GB RAM gut hinkommst, auch wenn man noch z.B. eine Datenbank o.ä. laufen lässt. Das Transcoden sollte auch ausreichen, solange es kein 4k ist (keine Erfahrung, zu große Bandbreite, ich habe die Sachen auch in HD). Emby transcoded den Film bei Standardeinstellung mit voller fps in den Cache und streamt davon, d.h. er transcoded nicht die ganze Zeit. Wahrscheinlich geht dir der Upstream aus, bevor du zu viele Nutzer hast Ich würde wahrscheinlich auf 1 TB mit der SSD gehen, da die Mediacenter im Laufe der Zeit viele Metadaten wie Bilder ansammeln und die Config immer größer wird.

Many people appear to be missing the entire point of 2FA for emby, believing it's magically going to stop the bad guys - the simple answer is it's not, not even close. If you are naive enough to use the same password for multiple systems, then 2FA is going to get you some more protection when (not if) your password is available from a data breech. But the real risk to emby is not it's data or media, it's simply an 'entry point' into your home network that is now available and for the bad guys to wait for an 'emby toolset' vulnerability. It could be the emby web server, ffmpeg, it's database etc - if one of those gets compromised and remote code is executable from the internet - then bang, the attacker has access to whatever network your emby server is on. For those with experience on cyber security/networks - yes there are all sorts of extra protections available - Reverse Proxies, VPN's, Isolated DMZ networks, IPS etc - these would all sit 'in front' of the emby web service - which should sit 'inside' a reasonable protected perimeter 'defence'. So going back to the question - is 2FA going to add protection - my personal belief is it's not a priority - especially when HTTPS is not enforced, password strength is not enforced, brute force lockouts are not enforced etc. These are things which could be done with relative ease - and provide better basic security out the box. Once the basics are done - THEN maybe think about 2FA for the Admin account .. Just my 2p.