native recording

[pikachu937]

Describe the feature

Hello. I think that I will express my general gratitude to you for implementing the stream recording function, but I would like to ask you to expand the functionality.

Add record options

• recordFormat - mpegts
• disable audio track recording
• balance recording between drives

    # record format: mpegts, fmp4
    recordFormat: mpegts
    # enable or disable audio track
    disableaudio: false
    # list mount point 
    drivers: /mnt/rec1, /mnt/rec2…/mnt/recN

[aler9]

Hello @pikachu937,

disableaudio and drivers won't be implemented for reasons already explained in #2255.

Regarding mpegts, the main issue is that the format currently doesn't support AV1, VP9 and VP8. In case of AV1, there's a specification available but most player don't implement it yet, in case of VP9 and VP9 there is not even a specification. Therefore, support may be added in the future but it can't be considered a general purpose solution.

[pikachu937]

about disableaudio in 2255 you suggested

"No. If you want to remove audio, you can create a reader that reads from a path, removes audio and republishes the stream to another path."

but that's not what I'm talking about! I suggest disabling the saving of the audio track, this will be much easier to accept by the server than re-opening a new stream without an audio track

about drivers in 2255 you suggested

No, load balancing can be achieved by using multiple instances of the server, while fauly tolerancy can be achieved by deploying the server inside Kubernetes with a storage plugin like Longhorn or GlusterFS or a hardware solution like a SAN. Nowadays load balancing and fault tolerancy are delegated to the infrastructure (K8s), there's no need to natively implement them.

I’m not talking about balancing the load on the server itself, but about balancing the storage of records between drives installed on the server, be it local drives or connected over the network. the server can monitor the fullness of a particular drive and save the recording to a less loaded one; this will be very important when using local drives.

about mpegts
your server already knows how to save fragments in mpegts in hls operating mode, besides, you write “mpegts - uses MPEG-TS segments, for maximum compatibility”, so I think not only I would find this format useful (even if it doesn’t support AV1 , VP9 and VP8) in the context of further compilation of hls playlists, this is at least an expansion of functionality and an increase in the user audience

[aler9]

Anyway, MPEG-TS support will come very soon.

[pikachu937]

This is great news, you are the best. What about disabling the saving of an audio track without restreaming and balancing the recording on the drive?
Thank you.