And the memory leaks get closed one after another? Dont they? Just because there are still issues does not mean it gets improved upon.

Media matching is no issue if you follow the naming sheme.

I am not upset at all, not sure why you think that.

Jellyfin will and connot be the replacement you wish it to be. Exposing something to the Internet is not a solution for the normal person. Heartbeat, Log4shell etc. etc. all of those are the reason why, not necessarily the service you are hosting by itself.

Especially in an age where tailscale is available to install on every major smart TV or other devices i do not get why you even want to recommend ppl to expose it.

No, not really. But what should i expect from someone who states as an ‘objective opinion’ “I do not like the programming language so the project is bad”

If i had to guess, since you are jumping on the memory leaks, you got an issue, reported it and did not get treatet like they fix it with a priority.

You keep jumping on “They had an RCE so the security must be completely broken”

Once? No jellyfin has had about 4 major RCE issues since the fork. At least 4 that I’m aware of. Blaming it on the previous code only makes sense if the split is recent. They have had time to completely rewrite if they really want.

It absolutely makes sense, otherwise they would have had to throw everything away.

The EFcore refacotring was like 6 years in the making.

And all that from just a few single ppl. Look at the ckntributer list, and how many contribution. Not many active devs are working on jellyfin on their free time. The problems that jellyfin has, is not from a lack of trying but a from a lack of finger and arms.

And you need to take it like it is.

But for complete other reasons than RCEs or similar.

As an FOSS project that inherited lots of shitty code this is basically the best thing they could do.

Not sure why, but you get specific about once RCE but not about other problems and keep vague about them. Is it the lack of understanding or disingenuousness?

It has had a pretty high number of RCE exploits including one recently the architecture of the web service is just very poor and leads to a lot of basic problems.

So they had an RCE that got fixed therefore the software is bad and insecure. Therefore every OS and basically any enterprise software that was ever used is insecure.

Got it.

And Plex doesn’t require any. It’s okay to accept that one product can be more polished than the other, and Plex has a lot of stuff that “just works”

And it is ok to accept that Plex is getting worse and worse. Only reason why ppl use it these days is because they still have an old lifetime pass. As soon as they take it away or introduce a new tier of features or even removing features of it, they will swarming away from Plex.

And they will!

OC never said anything to do with your comment, you seem to be really offended by recommending an alternative to a tool that you use.

As I said, when you know the exact path of a media item on the server then you can check if the item exists.

If you choose a none standard filepath its not an issue.

Should that be fixed yes.

Whats the scenario? A law firm could brute force check all media items on open jellyfin servers? Highly illegal to exploit something like this in a lot of jurisdiction. And would also not proof the existence of the media on the server, just a file named like it.

Mitigation? Just add another random letter in the docker-compose mount path.

Have you even read the issues and understood them?

Yes, those should be fixed, but unless you are worried about someone hijacking a video stream when you use a generic media path, there is not that much to worry about.

7.0-rc7 is probably due to the 7.0 release early mid april. So the fix was in the mainline on 1st of April. The commit on 11th from GKH was probably due to the release.

I am not that familiar with the commit and release structure to get more into detail. But to me it clearly looks like the statement on copy.fail is correct, that the fix was in mainline on 1st of April.

From my point of view, I would suggest that maybe the communication downstream to the distros was not handled that well? But who would be to blaim? The researches that would need to communicate this issue to most existing distros? Linux maintainers? Distro maintainers?

Hard to say, without knowing the communication of the related mailinglists and disclousre etc.

Honestly, thats a really bad take. Yes obviously, you should not let attackers access the terminal, but there are linux servers that rely on multiuser operations, like Servers that are meant for terminal access, like HPC.

Then services get hosted via container these days, so even with rootless containers you get root access if you only get RCE on one service. And even if there are additional VMs for more isolation between host, you still get root on the whole VM.

Looking at the CVE on NIST,i found following commit which dates to 30.03

https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=a664bf3d603dc3bdcf9ae47cc21e0daec706d7a5

The patches where proposed over a month ago and the patch to the kernel was commited on 1th of April.

Either the Vulnerability was not proper communicated to the distro maintainers or they were the ones sleeping.

This was probably executed as a responsible discllsure where clear timelines and release dates get communicated from the beginning.

I find it hard to blame the security team here when there was 1 month of time between first commited patch and release of the PoC.

I have q xyz Domain as my main Domain. And there are basically no issues here. But in 1 or 2 places i noticed that the domain gets blocked, a couple of free/open wifis or a couple of DNS servers. But nothing major imho.

By default this applications allows when adding a server, that the communication is not encrypted between the app and the server. This should be configured by default to enforce TLS encryption. If someone would want to disable dis behavior and allow unencrypted communication, then this should take extra steps.

As i commented somewhere else, to say that since it is turned off it is secure by default, is like saying: “The SSH server is turned off by default so the configuration that comes with it does not need to be secure when shipped”

Thats like saying:

“The SSH Server configuration does not need to be secure because the SSH Server is turned off by default”

Yes, this is what we’re discussing… Are you a bot?

Obviously no. But you keep dodging the point here. And instead of comming up with an argument against my point, you seem to try to attack me personally.

In security and development there is a statement, called “secure by default”. That means the default settings are secure. This would encapsulate something like enforced Transport encryption.

Does this mean that the config can not be changed to fit the thread model? No.

Not sure why you’ve chosen to be indignant about this particular implementation.

We are talking about a tracking App. Most selfhosted projects do not store such private data. You may can mage the argument for immich but only for ppl who take a picture every 5 min.

If the target server is compromised or taken by LEA the data is gone.

Laying the responsibility into the hands of the user is not ok for such an data aggregating service. Such highly critical, private and intime data should be protected and secure by default.

Not even transport encryption is enforced in the project. At first glance, http is allowed on local connections?!? Generate a self signed SSL cert on start and pin it in the app. Easy.

It is no excuse that other services do not follow these state of the art protection measures.