[Weird behavior with hard drives, not sure if Unraid related.]

22 minutes ago, itimpi said:

How many drives do you have when they are all connected?

11 when it is still able to Launch (7 data, 1 parity, 1 unassigned, 2 cache pool) and 12 when I am no longer able to. I just checked my psu (Evga 450 BT Bronze), and its a single rail PSU, so that idea is out the window. My last thought is that maybe I didn't pay close enough attention, and either Cable A or Cable B was used twice, and one of the cables is the problem after all... I'm ordering a Molex-to-SATA power cable that a different community of server builders strongly recommends, hopefully this will fix the issue.. Though I don't know how it could! If it were simply a bad quality cable, I don't know why the server wouldn't start at all.. 


Edit: With a parity rebuild operation in progress, and two people transcoding Plex, my UPS shows me pulling 280 watts. So I really don't think its an overloading issue...

[Weird behavior with hard drives, not sure if Unraid related.]

Hello, I ordered 3 new hard drives, 1 to replace a failing one and two for upgrading storage.

Replacing the failing one provided no problems, I did it by itself before adding the two new drives. 

However, when I went to add the other two drives, the server would turn on / run, but unraid was not accessible. In addition, it seemed like whenever both drives were connected, the server didn't "properly start up" - for example, when I clicked the power button on front, it instantly powered off with no beeps. When the server is running properly, it has a few beeps and takes maybe 30-45 seconds to actually turn off when you click the power button.

 

So, I tested each Cable and Drive independently, and everything was fine. I plug in Drive A with Cable A, server starts up, WebUI is accessible, new drive exists. Same story when I plug in Drive B with Cable B. So both cables and drives are proven to be OK. However, when I plug them both in at the same time, the server does its weird "improper" startup, and WebUI is inaccessible...

Frankly, my case is an absolute mess of cables and stuff, so i'm not sure where exactly Cable A and Cable B source their power from. I *THINK* one is on the sata rail, and one is on the molex rail, but they both might be on the molex rail? (Using Molex->Sata converters) Is it possible i'm overloading the molex rail? (I know the PSU itself isn't overloaded, I run at ~250Watts peak and I have a 450W PSU)

I don't know what to check, at this point, lol. Any advice would be greatly appreciated..

[(SOLVED) Added some hard drives, no longer able to connect to Tower//]

Hey, I added two drives to my server today... However, when I booted it back up, I only saw one drive. So I powered off again, changed how those two new drives received power, and tried to boot back up to see if that fixed it. However, when I booted back up, I was no longer able to connect to Tower//, and after a few restarts i'm still not able to connect to Tower//, which means I can't bring the server online or do anything... Any suggestions?

Update:

I used a cheap, MOLEX - to - 2SATA splitter, and for some reason, even being plugged into the system prevented it from booting. Who knows why.

[[Support] Linuxserver.io - Qbittorrent]

Updated this docker and experiencing lots of problems now. WebUI shows speeds as 'udnefined', and lots of stuff (eg, torrent name, size, etc) are all blank. 

Removed the docker/reinstalled, Tried again with additional step of deleting all appdata for qbittorrent, problem persists... Not sure what else to do. 

[Everything's working great. Why is my log being spammed?]

3 minutes ago, johnnie.black said:

You can ignore those errors, it's because SAS drives don't spin down, it's a current known issue.

 

You can also disable spin down to get rid of them for now.

Well, such is life. They are spun up 24/7 for seeding anyways. Only the parity I really care about and it sounds like ya'll are doing what you can.

Thanks you and the other poster for time. Have a good night 

[Everything's working great. Why is my log being spammed?]

28 minutes ago, saarg said:

 

If you have it now, then run the diagnostics and post the zip here. 

Sorry folks, i'm realizing now just how bad my original wording was. I really do apologize. The diagnostics above is exactly what you've stated, I just personally felt it didn't contain much because the error is spammed every few seconds and has been going on for an unknown amount of time. I don't know how to get logs from when the error begins because I don't know exactly what is causing it, or when it starts. (thats what my terrible wording was trying to imply)
 

 

Edit:

I have confirmed its related to spinning the disks down. I set default spindown delay to 'Never' and my log is silent. So, i'm guessing the errors occur somewhere around 20mins after unRAID starts. Lol!

I use this RAID controller: SAS9201-8I LSI

 

Perhaps it is related to my issue. The only drive i've ever sucessfully seen spin down is one of my SATA drives, which obviously isn't on the controller. What a bummer if it is related...

[Everything's working great. Why is my log being spammed?]

Sorry if I was unclear johnnie.black, the errors are currently being spammed in my log. I just said they're useless because they've been spammed for an unknown amount of time, and so the only info in there I copy-pasted in my post.

 

I'll assume you already know know this and i'm being dumb, so after the next reboot i'll post back with a new log (before/after, just to be safe)

[Everything's working great. Why is my log being spammed?]

Sep 12 23:42:26 Tower kernel: mdcmd (183756): spindown 7
Sep 12 23:42:26 Tower kernel: md: do_drive_cmd: disk7: ATA_OP e0 ioctl error: -5
Sep 12 23:42:26 Tower emhttpd: error: mdcmd, 2639: Input/output error (5): write
Sep 12 23:42:28 Tower kernel: mdcmd (183757): spindown 0
Sep 12 23:42:28 Tower emhttpd: error: mdcmd, 2639: Input/output error (5): write
Sep 12 23:42:28 Tower kernel: md: do_drive_cmd: disk0: ATA_OP e0 ioctl error: -5
Sep 12 23:42:28 Tower kernel: mdcmd (183758): spindown 7
 

 

My log being spammed over and over with this junk.

Varies between Disk 0 and Disk 7.

 

Disk 0 is (I think) my parity drive - seems to be working great (no smart errors..)
Disk 7 is my 7th disk, (of course!), and it also.. *seems* to be working good? I have space settings set to fill-up, and disk 6 isn't full yet so I haven't really been able to check its performance.

Only similarity between these drives is their lack of use (I am writing to my cache pool so parity isn't necessary, and disk 7 isn't up to bat for storing data yet)

So it seems like unraid *wont* spin these disks down?

I don't really care, just curious why/what's going on. Most of my disks are spun 24/7 anyways.. but these two should be spun down I think. 

 

 

Edit:

I added diagnostics but they're probably useless. I restarted my unraid box yesterday and within first few mins of booting didn't get these issues. (So restarting and getting new logs probably won't help, either.)

Any advice?

tower-diagnostics-20180912-2346.zip

[[SOLVED] JBOD Cache pool.. and adding the 2nd drive?]

21 hours ago, johnnie.black said:

If current cache is btrfs you can add and change the profile later.

Thankyou. My single 128GB SSD is indeed BTRFS. I will add/format the 500gb and then add those commands you outlined. Thanks for being so knowledgeable about BTRFS! And double thanks for sharing that!! Marked as solved

[[SOLVED] JBOD Cache pool.. and adding the 2nd drive?]

So after some googling I found johnnie blacks post about various formats for cache pool devices, raid0,raid1,jbod, etc

I have a 128gb Ssd that I currently use for docker appdata only.. I transfer data too much/too fast to bother with enabling cache, and having it always be full, and whatever. So I bought a 500gb ssd to supplement and I want to add it to my array, and enable cache.

Do I have to lose the data thats currently on my 128gb? I obviously can back it up (it already is), but maybe was hoping I didnt have to

anyways, is add drive to cache pool, let it format/wipe the new drive (not the old 128gb!!) , and then set the options in both drives to the commands specified by johnnie black, the correct procedure? 

 

thanks folks

[[DEPRECATED] Linuxserver.io - Rutorrent]

18 hours ago, luutuananh said:

Hi, a quick noob question. I can't write to folder that rtorrent are using to seed. How can I change that so I can write to that folder?

chmod -R 0777 /mnt/user/Media

 

redirect folder name to w/e you need, it runs on ALL subdirectories, I just run it on my /Media root folder every couple days.

[[DEPRECATED] Linuxserver.io - Rutorrent]

Hello, adding downloads to ruTorrent seem to be causing tons of lag to the docker (refresh wheel frequently lighting up, normal usage never lights up)

And i'm also getting terrible download speeds. It seems to 'freeze' everytime it sucessfully connects to a seed to download.

 

Please advise? I have 550+ seeding with no issues. Just downloading. I'm currently using a different client to handle downloads and then this one for uploading.

 

Edit: not sure if this matters (it is downloading to corrrect directory), but the bottom-left "Disk Available" counter is showing xx.xxG/20.00G used, so I believe its registering its path as inside the Docker image (which is 20GB?) .. but maybe i'm offbase! Thanks.


Edit2:

I will leave this up , but I believe the issue was my settings. Google "optimizing rtorrent" for understanding the settings and not just making assumptions like me.

Wished peers/minimum peers should never be set to 0, I believe this caused my issue. (I still am using a seperate client as I like the other functionality but the issue is likely resolved)

[[Support] Linuxserver.io - Qbittorrent]

On 3/30/2018 at 6:37 AM, sparklyballs said:

 

don't use windows lol

 

but seriously though read the readme https://github.com/linuxserver/docker-qbittorrent#usage linked to in the OP

 

and set umask accordingly http://man7.org/linux/man-pages/man2/umask.2.html

 

 

Not a windows problem.. On my mac, the files inside these "bad permissions" directories don't even show up ?

 

I tried using the above command, set -e UMASK_SET=000,  in my qbittorrent advanced options under 'parameters' and it ruined the docker, requiring reinstall. Anybody have any suggestions there?