ranasats
-
Posts
9 -
Joined
-
Last visited
Content Type
Profiles
Forums
Downloads
Store
Gallery
Bug Reports
Documentation
Landing
Posts posted by ranasats
-
-
6 hours ago, aptalca said:
Also keep in mind that plex only does hw encode on nvidia, not decode (silly omission of a flag in their ffmpeg build). I let them know about it a few times a long time ago but they dismissed it by saying they don't officially support nvidia.
Plus, most consumer nvidia cards are limited to two simultaneous transcodes.
Although I tested all sorts of hw transcode options, I'm still using sw transcode on my production server, which can handle at least 7 simultaneously.
I'll probably consider switching to hw transcode only when hevc content becomes ubiquitous, because sw transcode won't cut it for that.
Nice, I'll keep that in mind as well. I actually did a test, my Xeon 5690 was able to transcode 6-7 streams and CPU usage was at 100%, no one said they experienced any lag or loss of quality. I think 3 streams were 1080p and the other 4 were 720p. That's through docker. At most I've had 10ish connections but that rarely happens so I think I'd be okay with keeping it on docker.
Thanks for the development of this =) Everything works great.
-
On 1/5/2018 at 12:29 AM, aptalca said:
Plex only (officially) supports intel at the moment.
For nvidia one can install the nvidia drivers on both the host and inside the plex container to make it work. However, you probably won't be able to install them on unraid.
Dang, I must rethink my Unraid choice then. The main purpose of my server is to be a Plex/media server. I have a 970 which was working fantastic on windows 10 to do multiple transcodes in addition to my Xeon. Might have to either make a windows 10 VM for just plex and pass through the GPU or stick with just windows instead of unraid if the performance isn't as good...
Thanks so much for your input!
-
So, I had a few problems w/ my cache drive. I added a new parity drive, and rebuilt the parity the last day. While this was occuring I noticed the cache drive was completely full (250GB SSD). After parity completed I started the mover which wasn't starting. Read up and found out I had to disable/enable the cache drive from settings which I did. I also ran the permission fixer to see if that helped.. Started the mover and my log file is completely filled now with the following messages:
I've attached my diagnostic file if anyone can help me out.
-
15 hours ago, wgstarks said:
Awesome, thanks. I was wondering why it "automatically" corrected itself after a restart. Good to know!
-
I searched for GPU/Hardware acceleration, but was still slightly confused on whether or not GPU acceleration was supported in this container. I have a GTX970 connected to my unraid server as my primary video output (MB/CPU does not have an integrated GPU). I've enabled hardware acceleration in the container, but will this use the GPU without any additional configuration, or will I need to build my own container for plex?
-
Hey, Updated to the latest 6.4.0 rc19b build and I'm getting:
Jan 3 22:47:59 Mk4Alpha root: error: /plugins/unassigned.devices/UnassignedDevices.php: wrong csrf_token
Looked at the unassigned devices and everything seems to mount fine , no problems or anything. I can access the unassigned devices as well. Any reason this error is "bad" and I should uninstall the plugin until fixed?
-
Hi,
I just mistakenly added a 8TB disk as part of my array and booted it up. I wanted to add it as my parity drive. Currently the array has 3x 6TB drives and a 250GB SSD as cache with no parity (I just booted the server last night). I added the new 8TB and mistakenly used it as an array disk. How can I remove this disk and set it as parity without losing any data from my current array? It will need to build parity when I add the 8TB as parity. 18TB array is around 50% utilized without backup (the data was moved from the 8TB to the array!), so I don't want to create the new array from scratch.
Thanks!
NVM, read up on the "New Config" option and all is well =)
Unassigned drive - NTFS SSD - "Read only file system"
in General Support
Posted · Edited by ranasats
Hey,
I'm trying to delete files from my SSD which I mounted as an unassigned drive. It was originally my windows drive but now is mounted on my server. I was trying to remove files from the unraid shell but I keep getting the "rm: cannot remove '<file>': Read-only file system" error. However I can delete files from a networked computer that has the SSD mounted via smb...
Any idea how I can mount it as a "readable" disk?
Thanks!
NVM: Fixed.
unmounted drive > checked info > hit fix. Then it was mountable as read/write.
EDIT 2:
Half the files are now getting "Cannot access <file>L Input/output error"
Hmm