[Beta] Asuswrt-Merlin 380.65 Beta is now available

[RMerlin]

Is it normal that with Asuswrt-Merlin firmwares Minidlna gets a new uuid on every restart?

The UUID is generated by minidlna itself, it's not specified by the firmware.

 

[Diamond67]

The UUID is generated by minidlna itself, it's not specified by the firmware.

Thanks for the reply.

Is there any way to fix my problem? With current Merlin firmwares my Sony TV shows "phantom" Media Servers after reboots of the router. That never happened before. I haven't updated the firmware of the TV lately. Only the fw of my AC68U.

I even reformatted my attached usb thumb drives (EXT2 and 4) and added paths to the Media Server and renamed the Server.

BTW: I didn't use my asus router for about a year, so the latest fw before testing current Merlins and Merlin betas was "RT-AC68U_380.57.6_HGG-FINAL" last year, and some non-Merlin fw's before that. With those fw:s my Sony TV worked all right.

 

[john9527]

The UUID is generated by minidlna itself, it's not specified by the firmware.

Something broke getting the serial number along the line.....ran into this on my fork. Check out
https://github.com/john9527/asuswrt-merlin/commit/ce9a22edf637e1d1362379ce9a417a736f676b94

 

[RMerlin]

Something broke getting the serial number along the line.....ran into this on my fork. Check out
https://github.com/john9527/asuswrt-merlin/commit/ce9a22edf637e1d1362379ce9a417a736f676b94

Changing UUID seem to be by design. Minidlna's uuid.c hasn't changed in years, and it specifically mention that it's generating a UUID type that's time-based.

        /* Determine 60-bit timestamp value. For UUID version 1, this is
         * represented by Coordinated Universal Time (UTC) as a count of 100-
         * nanosecond intervals since 00:00:00.00, 15 October 1582 (the date of
         * Gregorian reform to the Christian calendar).
         */

        /* Set UUID version to 1 --- time-based generation */

So that part is normal. You'd have to ask Justin or some UUID expert as to the reason why the UUID is dynamic and not statically based on the specific device.


I'll have to investigate the matter of the serial number. Most glaring issue I see there is indeed the absence of a fallback block in case the LAN mac address would be blank. But then, I don't see why it would be either, as other things would probably be more seriously broken if that were to happen, so that'd me more of a precaution than a likely potential situation.

The rest of that block of code hasn't been changed since the initial 3.0.0.3.144 merge according to git blame, so it wouldn't be a new issue, unless minidlna itself used to process the provided serial, and that processing had changed at some point.

 

[joegreat]

For some reason I cannot upgrade from beta 3 to 4. It says it's upgraded. But still says on beta 3. I'll do sometime testing. If I have to default first i will. I had no problem updating from 123

Typically memory shortage: Remove USB drives and reboot - then try again to update firmware!

 

[Diamond67]

The rest of that block of code hasn't been changed since the initial 3.0.0.3.144 merge according to git blame, so it wouldn't be a new issue, unless minidlna itself used to process the provided serial, and that processing had changed at some point.

I checked my router config backup archives, and I have used at least these Asus-Official (Factory) firmware versions with my Sony Television:

Year 2016
RT-AC68U_3.0.0.4_380_1842-g7414eb9

Year 2015
RT-AC68U_3.0.0.4_378_7369-g8f33a45

And haven't ever got this problem with uuid/serial number (or whatever is the reason).

 

[john9527]

I'll have to investigate the matter of the serial number. Most glaring issue I see there is indeed the absence of a fallback block in case the LAN mac address would be blank. But then, I don't see why it would be either, as other things would probably be more seriously broken if that were to happen, so that'd me more of a precaution than a likely potential situation.

If I remember correctly, it was returning a random value for serial. And not all clients are sensitive to it. I found it on my Sony Blu-ray player. Same symptom as the OP. Making the change I did made it consistent (again if I remember correctly the first part of the UUID was hardcoded on my fork).

 

[RMerlin]

If I remember correctly, it was returning a random value for serial. And not all clients are sensitive to it. I found it on my Sony Blu-ray player. Same symptom as the OP. Making the change I did made it consistent (again if I remember correctly the first part of the UUID was hardcoded on my fork).

Thanks, I'll take a closer look later.

 

[RMerlin]

380.65 has now been officially released. Thank you for the feedback provided during that long test cycle.

Please use the release thread for further discussion regarding this release.