thebaldconvict

Stupid question but I didn't see anybody else ask. I use NVidia encoding and sometimes the resulting file is bigger than the original. Is there a way if that is the case to keep the original file? Maybe even fallback to CPU encoding for that file? Also, I have the audio set to convert to AAC and create a stereo track also. Why do I end up with two AAC stereo tracks if I feed it with a movie with a single 5.1 channel AC3 track?

Does it support nvidia now?

Hi there, I give in, I’m not usually to bad at this sort of thing but this not starting of my Plex docker has me stumped. I can see any error messages anywhere so I’m wondering if I am looking in the wrong place? If the docker is running fine and gets shutdown by the backup routine and the routine says it is starting the docker, where can I look to find out why it didn’t start? Is there any obvious reasons that have affected others for things not starting. If it gives any clues, I am on The NVidia build of Unraid and my Plex is using NVEnc transcoding. I don’t think the docker has started on its own since installing the NVidia card.... It is annoying because it used to work fine... Any assistance at all will be gratefully received ha ha

I get no reasons in the logs either that I can see. I'm using the standard Linuxserver Plex docker. It is strange as the problem only started a few weeks ago having worked fine for ages.

It isn't by any chance Plex that doesn't restart is it? Mine always used to but recently doesn't auto restart even though it does just fine manually. Only docker it doesn't restart.

Thanks both for your help and thank you for that explanation PWM and just thinking about that description, I might want to swap out my parity disks which from what I've been reading are shingled disks... Although I've also read that Unraid is fine with SMR disks so who knows. I was swapping my parity disks to 8TB disks but have only done one so far so I have a ST4000LM024 and a ST8000AS0002 as the parity disks, maybe wise to swap them with 2x PMR drives as I've read that even the 4TB is an SMR?

If I'm doing a reconstruct write I thought that removed the parity disks from the equation while doing that write? is 15-20MB/s really all I can get? What happens if I copy two files to the array and they go to different disks? Is that why sometimes I see it drop to 5/6MB/s?

Hi all, I am wondering would anybody with the Dell PERC H310 flashed to IT mode be able to comment re write speeds to their array? I currently have a cache drive, an unassigned disk and two parity disks on my motherboard with all the data disks of my array on the H310 and am having some speed issues. sorry for the long ramble that is coming!! So if I use the script I found here on these forums called write_speed_test.sh and run it on each disk individually all of the disks start at about 200MB/s and slowly drop to around 130MB/s, except the SSD cache drive which appears to start at around 500MB/s and drop to around 200MB/s, a little slower than expected but I can live with that. So, if for example I test the speed of disk6, I get 130MB/s, the unassigned disk I see about the same (maybe a little slower, it is an old 1TB WD Blue) If I test the SSD at the same time as testing the unassigned disk (both motherboard disks) I get pretty much the same results. If I test say array disk6 and a motherboard disk I get normal speeds from the motherboard connected drive but only around 60MB/s from the H310 disk. If I then test say disk5 while writing to disk6 (Both on the H310 and directly in reconstruct write mode) I get around 15-20MB/s on each and this is how it behaves most of the time. Is this a limitation of the H310 or my motherboard/cpu or something else? Its been this way for a while now but it only annoyed me yesterday when it took 4.5 hours to move a 256GB file from the unassigned disk to the array. Cheers all Tim

Hi all, back again just to give feedback on what I ended up doing. So I didn't want the array offline for ages (and it would have been), it is school holidays and the Mrs (and 5 kids) would have murdered me for Plex been offline and it has a Pi Hole VM also so internet would have been down too soooo I replaced the 4TB. So basically I bought a 4TB to go alongside the 8TB that I wanted to swap in rather than risk the array. I replaced the 4TB with the new drive and rebuilt using the standard process keeping dual parity throughout, then when that had finished I shutdown and pulled the 4TB parity 2 disk replacing it with the new 8TB and set it off doing a rebuild. Well now a parity rebuild takes 21hours so at a guess if I had done a parity swap my array would have been offline for 21 hours? Anyway, the array remained available throughout and at all times I had a protected array. Thank you all for your advice, it cost a touch more but now I am left with a tried and tested spare data disk (The original Parity 2 disk) and was protected all the way. AND I'm halfway to being able to use 8TB disks too Cheers Tim

Oh I get it, yes, perfect! Thanks again all Edit.. Any way to keep the array usable while it rebuilds the Parity, it says in the guide to'll be offline for the duration and thats going to be a long duration on an 8TB.

Forgive me if this is a stupid question but is that not what I was planning to do albeit manually and with parity 2? If there is a better way to do it though I can swap Parity 1 no worries, it is exactly the same type of disk at Parity 2.

I wont lose protection completely though will I? Actually would I? I would be down a parity disk AND I'd be down a data disk (potentially).... Hmmm, I was originally thinking I would be protected while parity 2 was being rebuilt but since one drive is having read errors I might not be... I haven't thought this through have I ha ha...... I could replace this data disk with another 4TB and then replace the parity with the 8TB and then move that parity and add it as a new disk to the array I guess.... Would keep it protected though out the process that way, I definitely want one of the parities to be an 8TB though for upgrading in the future (and I already have it) too LOL

And then after it has rebuilt the disk just add the new 8tb in as parity 2? Will it be ok with Parity 1 still being a 4tb, I assume it will be? Sounds easy guys, thank you for your help

As per the title really, I have a disk that is showing read errors, a long SMART test also shows them. Have swapped the cables over with a different drive in the array and the problem follows the disk. Anyway, all my disks are 4TB ones including my two parity drives, so what I'd rather do is pull Parity 2, put that in place of the failing drive and rebuild it. Then pop an 8TB in in place of Parity 2. Would I need to preclear the parity disk again before adding it to the array as a replacement data disk or am I ok to just stop the array, unplug the failing drive and start the array with parity 2 in its place?

Needs the tape on my Gigabyte h61 board too Sent from my SM-G925F using Tapatalk

Mine was as well, would stay up for months quite happily and it seemed to lock on the backup days mostly too. It's run through two CA backups since removing that stick and also I have a script that stops the VM's, backs them up to the array, starts them and then zips the disk images which must be pretty taxing since they are about 120gb each. It's done that twice also and people have been streaming off it via plex pretty much constantly this week with it being school holidays. One thing I never tried was downgrading unraid. Sent from my SM-G925F using Tapatalk

No sorry, I meant mine has been behaving for just over a week now (early days still I know) but was behaving just like yours before the ram died so I was wondering if maybe yours could be a similar fault? Hence maybe run memtest for a couple of passes. ☺ It isn't unheard of in the Windows world for later versions of Windows to show up ram issues so didn't know if maybe the Linux kernel could have the same effect.

Is it possible it could be a similar thing? I'd it possible that a newer kernel or kvm version is driving ram harder? Could be worth running the memtest86 on the startup menu for a few passes... Sent from my SM-G925F using Tapatalk

Ran my server with only 8gb for a couple of days while I found a matching stick of ram and all seemed well. I have since put a new 8gb on taking it back to 16gb and have got an up time of 8 days so far, not huge but much better and it's still climbing... Fingers crossed that was it! Better still the matching stick cost a grand total of £20, a new set would have been over £100.

Actually, I might have figured it out.... Flakey RAM. Ages ago (around 2 years ago) my server had gone into a boot loop situation where the fans would come on, then it would power off after about 2 seconds and cycle round in a loop, it would never show POST. I fixed that by doing a BIOS reset and all was well until this all started. Well I tried to reboot after the message above and I got the same boot loop thing as before but after stripping everything from the pc (all drives and cards) and a BIOS reset it was still doing it. If I take one of the RAM DIMMS out it posts fine, swap the other DIMM in and looped again, it follows the chip regardless of the socket, whenever that stick is in the board it wont post. It is now limping along with only 8GB RAM and showing 86% used on the dashboard, not sure it is enough but it is on and working for now... I'll monitor it like this and see how it goes, if it runs out of RAM or is stable I'll replace the stick, only problem is I can't get the same spec so will have to replace the pair.

Mine went again at 08:59 yesterday morning, spread a load of stuff across the screen that is plugged in and hard locked with only this is the syslog: kernel:page:ffffea000d092b00 count:0 mapcount:0 mapping: (null) index:0x1 Wierd huh?

OK so mine is still alive (just) It is sending loads of call traces to the syslog at the moment, have attached my latest stats. Is there a safe way to downgrade Unraid as I suspect it might be to do with the latest version(s)?... tower-diagnostics-20170326-1152.zip

Have any of you got a keyboard and screen plugged into yours? Mine when it did it was sat at the login screen with no new messages or anything but the keyboard was completely unresponsive also. But it runs the backups at 2am Tuesday morning so it isn't that on mine, in fact it completed them successfully this morning...

Well this is the current output, I will turn on troubleshooting mode also so I can get those when (if) it does it again.. tower-diagnostics-20170320-1933.zip

OK stupid question time, I see jamesj2 has managed it but how do I get the last diags file? Ah ok so I throw FCP into diagnostics mode and then when it does it again I just grab the latest diagnostics file....