sub0ptimal

@trurl - I want to use a variety of backup drives, some of which were pulled from older systems and aren't large enough to hold the contents of my main array drives. I also do not necessarily want to back up everything in my array. Ripped DVDs can always be re-ripped. @pwm - After considering my options, I agree with you. I repurposed some old hardware and built a second server. I chose to use bare Linux as I did not necessarily need another Unraid system just for backup. I plan to add scheduled tasks to rsync Unraid shares to the backup server.

I want to create a second array in my Unraid server. I want to create offsite backups, and the simplest/cheapest way for me to do this to copy my array to backup drives which are later stored in my safe deposit box at the bank. I know that I can use Unassigned Devices, and this works well for targeted backups, but I am wanting to make a backup of my whole array. I will need an array of drives to store the data. I did a forum search on this topic, and it seems that most talk is about creating a second cache pool. My existing cache is fine. I just want to create a temporary array for backups.

I have a 24 bay Norco case, and I've labeled each physical bay 1 to 24. After troubleshooting a problem with one of my HBAs, I needed to move the drive that is in bay 18 to bay 7. I would like the device to be labelled "Drive 7" in the Array Devices list, to let me know where to find it if I need to physically move it again. Unfortunately, I don't see a way to move around drives in the Array Devices list. After stopping the array, If I remove the Disk 18 device and try to put it in the DIsk 7 slot through the array setup page, then it is warning me that my drive will be formatted. I could live with this (allowing the array to rebuild from the parity drive) but I think it will still try to emulate the DIsk 18 device, while also reformatting and adding what it thinks is a new Disk 7 device to the array. Is there a way to tell unRAID that the physical drive that it is calling "Drive X" is now "Drive Y" ? This brings up a bigger request that it would be helpful to add notes to each device that are visible in the Array Devices list, so I could, for example, record the physical location of the drive within the case, the date I installed it, etc. I am currently using a spreadsheet for this housekeeping. But that is more of an enhancement request.

Quite possibly this is not related to Docker, but both times this happened, I was starting a docker app. Maybe a coincidence. I am not the user who posted diagnostics earlier. I will try to troubleshoot some more, there is too much personal info in the diagnostics dump to upload. There are no errors in the log that I can see, other than a lot of complaints like these, which I assume are related to the web GUI: root: error: plugins/unassigned.devices/UnassignedDevices.php: wrong csrf_token root: error: webGui/include/DeviceList.php: wrong csrf_token root: error: plugins/preclear.disk/Preclear.php: wrong csrf_token I still have 10 hours on the parity check/rebuild, so I can't do anything until that completes.

Happened again today. Definitely seems to be related to docker. Restarted a docker app through the context menu on the Docker tab. Received popup "Execution error". All shares gone in the UI, /mnt/user is gone, shares not visible on the network. /mnt/user: Transport endpoint is not connected

Nzbget from linuxserver. I decided to stop the data rebuild and reboot, which solved the problem. Stopping and starting the array was not enough, I had to do a cold reboot.

I was working with a new docker today, and the same thing happened. All shares disappeared. I'm doing a parity sync/rebuild, so I can't reboot right now. Server will be kaput for another 19 hours.

Ah well, now I feel dumb for not seeing the previous question. I wonder if the trial version of Unraid allows dockers? I want this bad enough that I am willing to run a 0-disk array just to have the dockers. I guess I can run dockers without Unraid, but my experience with dockers is nil outside of Unraid.

I have a motherboard with two ethernet ports. One is currently unused. I'd like to use the second ethernet port (eth1) to allow my dockers to reach the internet (WAN) without impacting traffic on the primary port (eth0) which serves the LAN. eth0 and eth1 will be on different subnets. I suppose there are two steps here, one is getting Unraid to use eth1, and the other is to configure the dockers to use eth1 instead of eth0 for their attempts to reach hosts outside the LAN subnet. So for example, in the Deluge docker, I would want the host web UI to be served on eth0, so clients can access the UI on port 8112, but I want the docker's bittorrent activity to occur only on eth1. Also, I do not want Unraid to serve SMB clients on eth1, only on eth0. Has anyone tried this?

I have 2 8-port RAID cards (IBM M1015) and several SATA ports on my motherboard. Does anyone know if array writing throughput is faster, or slower, with the parity drive on the SATA port on the motherboard instead of the RAID card? I believe that both the RAID card and the MB port support 6 Gb/s.

Ok. Thanks, I will start moving stuff off disk13 and disk16. I have room in the array. I will check but I don't think anything on those drives is irreplacable. Once I empty out the disks, then I should replace each, one at a time, right? Rebuilding parity each time. Would I do a pre-clear on the new drives before the swap?

Ok. Looks like the SMART test ran ok earlier today, but now is not completing. Report attached. Should I try to move as much as I can off the failing drive before rebuilding parity? WDC_WD30EZRS-00J99B0_WD-WCAWZ0176723-20160316-1741.txt

My array is 16x3TB plus 3TB parity. Two of my 3 TB drives are giving SMART errors. Drive 16 is unable to complete a SMART self-test. Drive 13 197 Current pending sector 0x0032 200 200 000 Old age Always Never 3 Drive 16 197 Current pending sector 0x0032 200 200 000 Old age Always Never 4 198 Offline uncorrectable 0x0030 200 200 000 Old age Offline Never 1 These drives are 4-5 years old so I thought it best to replace them with new 6 TB drives. I've read the documentation on replacing drives in the array. Since I am replacing a 3TB drive with a 6TB drive, I need a 6 TB parity drive. The documentation describes a "swap-disable" procedure where you replace the old parity with the new larger drive, and use the old parity drive in place of the failed drive. Ok, no problem. My question is, when do I pre-clear the disks? I'd like to stress the new drives during the pre-clear so I can RMA them if there are any problems, before I put them in the array. Do I run pre-clear on the new parity drive? How about the old parity drive that I am moving? Are there any other tricks to swapping the parity drive? Thanks!

Well, at least it's not just me going crazy. And my symptoms were the same, it worked at one point and then stopped. Poking around in the HTML, the table is starts with <table class="share_status share "> and the body of the table (where the shares are listed) is <tbody id="share_list"> I changed the id of the tbody tag to gibberish, and the list was visible again. So something in the CSS is causing ABP to think it is an ad. Checking google, it looks like other people have had problems with the dynamix webgui and adblocking. Or not, this may be a wild goose chase.

How strange. I inspected the HTML in the browser, and saw that the rows for my shares are present in the page, but were not being displayed. I started disabling Chrome extensions and refreshing, and eventually the shares came back. It looks like, at least on this particular browser & PC, having Adblock Plus running messes up the web GUI. That seems totally bizarre to me, because there is nothing in the HTML that would trigger ABP. It's just a HTML table. ABP even shows 0 ads blocks on the page. I turned off ABP for the unraid host, which fixes the problem. This is a fine workaround, but I'd sure be interested to know more about the problem.