Edrikk

Upgrade from 6.8 to 6.8.1 went without issue. If you don't mind, 1 suggestion, and one behavior comment: Suggestion: Can the login page have the cursor start in the username field upon load? right now it takes a click to get there, which can be saved given username entry is the first action on the page. Comment: In the versions prior to 6.8, one could jump directly to a URL (e.g. http://tower/Docker). i.e. entering username/pass would take you to that page. With the new login form, if the user is *already* authenticated, the browser will take you directly to that URL (given that you're already signed in). However, if you are not logged in, going to (again e.g. above url) will pop the login form as expected, but upon successful authentication you are always taken to "Main".

Fully realize that this is more of a usability/user experience request more than anything else: @bonienl is it possible to have the new login screen have focus on the username field when it loads? Currently one has to click/tab into the username field to start the process... It's an extra click that doesn't seem necessary. PS. I'm on Windows 10, and have sanity checked behavior on Firefox and Chrome.

Hi, The second report I received from running OpenVAS against my unraid box (6.5.0) follows:

This as I mentioned wont work. Users might skip a (few) versions then upgrade. In this most recent example, if the user jumps from 6.1 to 6.3, the 6.3 upgrade wouldn't have a message not point to the thread for 6.2, which is where Users are now told to delete docker.img as an example. Release notes are good, but the upgrade mechanism needs to make the call and do the work. Maybe at the end, tell the User what it did as it upgraded across versions. A good (smaller) example of this is how Kodi does DB scheme upgrades. It basically sees what version you are running, what you are upgrading to, and runs each intermediary update sequentially to get it to the version and state that it has to get to.

I think that either this should be automated, or made super clear as a message or something upon upgrading. I actually read the thread, and didn't notice that the User has to delete the file. Most people (rightly) assume that if an update is available, they can just click the button. I think LT would be wise to improve the update process to handle such exceptions; and the ones that can't be should be popped to the user in the GUI. This becomes more and more important as updates happen on updates, and potentially each needing independent "action". You can't expect the Users going from (for example) 6.1 to 6.8 to go through every update thread to make sure they see every upgrade minutiae.

Fair enough! It seems both are needed then to make this feature useful (with ability to truly disable one or both).

Actually I don't see the need for this at all -- filling up a disk is hardly an "error" of any kind. A warning when the TOTAL storage gets low would be useful ... but I could care less if an individual disk hits 99 or even 100%. The utilization on my disks on the Dashboard shows 100%, 100%, 100%, 100%, ... until the last 4 disks, which are then at 27%, 47%, 50%, and 45%. While it's VERY unlikely I'd get low enough to actually trigger a warning without realizing I was getting close anyway, it would be nice if there there was a warning when the TOTAL space available (preferably on a by-share basis, since not all shares are allowed to use all disks) was below a specific threshold [which, candidly, I'd prefer was set in GB instead of a percentage of the overall space]. Couldn't agree more. I don't think 99% of the people could care less if a specific single disk is full... That's actually pretty much going to happen by design. Now if we're talking about the total space, that's actually useful information for an alert. The current implementation not useful. As seen by the number of people inquiring, and setting to to 100% to shut if off....

Thanks again folks! A hard reboot indeed got things back. I then ran the "New Permissions" task to bring everything up to a known state. Thanks again!

Thanks folks. It's good to see at least that my recollection was correct (searching wasn't bringing up what I remembered reading). As long as I can ssh into the box, the rest will be alright, so it's great to see that a reboot should bring it up. Will provide an update later tonight (12 hours probably). Thanks so much!

Hi, This morning I was in a rush, and manually ran a shell script which changes permissions, based on input parameters which make up the path and file name. Accidentally, I didn't give input parameters (early morning rush, etc). Since there was no parameters, the script basically started recursively changing ownership and permissions from /. As soon as I noticed this I killed the script, but it was too late. It appears that the permissions on OS file such as bash (sh) were changed, and I can't execute anything as root. In fact, I had to go to work, and I fact SSHing into the box fails (post authentication) with an "/bin/sh permission denied" error. I copied the contents of the "flash" directory from the USB (as SMB still works), in case recovery will be "involved". The Web interface also doesn't load any longer. My questions are: 1). Since I'm at work, I can't try this till later tonight, but would a hard reboot of the box reset the permissions? (I don't believe the actual OS files are static and the ram disk is flushed, resetting the permissions if I'm not mistaking). 2). If I'm wrong on "1" any other suggestions? Running 6.1.4 Thanks so much. Moral of story: don't do things in a rush. Felt so dumb driving to work.

regarding lightbox issue on iOS, No I had only reported on the beta and RC threads. I have just created this ticket: https://github.com/bergware/dynamix/issues/3

Hi guys, Can anything be done in dynamix to make sure that the lightbox pop ups work on iOS? Right now when you finger to scroll through (for example) the update progress lightbox, the page behind the lightbox is what scrolls.

Makes absolute sense! Thank you!

Please excuse the very dumb question: If the VMs are set to use br0 and get an IP address from the router, why is it that I can't VNC to that IP address / appropriate port? I have to VNC to the UNRAID IP to have VNC connect which is weird....

Your share should be listed as a sub folder under "user". Your other issue sounds like a bug. Thanks Jon, and agree. For some reason the user folder wasn't expanding before, but when I just went back in to try again, it did. Probably a browser cache issue... Thanks again!

When trying to create a new VM, I cannot see the shares (above). I only see Disk1 through Disk5, as well as "user". Meaning I can't select my "VM_Images" share as where the VM should be installed. I see the below is a known issue, please ignore this part: Does anyone else's shares show up with a yellow triangle ("Some or all files are on unprotected storage.") I don't have a cache drive, and all my shares are on the "normal" (for a lack of a better term) disks, XFS formatted.

This is a bug that will be fixed in an upcoming release that only affects users creating vdisks on non-btrfs formatted devices and when traversing through user shares. See this to fix: http://lime-technology.com/forum/index.php?topic=40251.0 Perfect thank you! Indeed that fixed the issue... Even my old VM ran!

But that would mean that I can't access the VM from the LAN... I'll try it, but that's no good.. I asked this question before and you can still access via lan. I'm doing it right now in fact. http://lime-technology.com/forum/index.php?topic=39343.msg366969#msg366969 Interesting! Unfortunately I just tried (both the ISOs, as well as a vdi file converted qcow2) and they all give the same error... I am guess the qcow2 file won't be smooth since it was created on a different machine all-together, but the fresh ISOs..

But that would mean that I can't access the VM from the LAN... I'll try it, but that's no good..

Trying to setup a KVM for the first time... I downloaded Linux Mint (both 64 and 32 bits just to see if that was the issue), but regardless of what I try, I get the following error: Error creating VM: internal error: process exited while connecting to monitor: 2015-05-28T19:18:12.791976Z qemu-system-x86_64: -drive file=/mnt/user/VMs/LinuxMint/vdisk1.img,if=none,id=drive-virtio-disk1,format=raw,cache=none,aio=native: Could not find working O_DIRECT alignment. Try cache.direct=off. Nothing Fancy in the setup: - I've defined br0 under network settings - I've created a share for the ISOs and a Share for the VMs - I've assigned br0 in the VM Manager page - Click "Add VM" and: --- OS = Linux --- Passthrough CPU --- Using Core 0 and Core 1 (out of 4 cores) --- Init Ram 512MB (out of 4GB) --- Graphics Card = VNC (it's a headless computer) --- Soundcard None --- USB devices None checked (option being UPS) --- BIOS = SeaBIOS --- Machine --> Tried both i440 and q35 Any tips?

As I reported during the betas (I believe b9), I just wanted to share that the "update Plugins" pop-up has two issues: 1) On iOS devices, the contents of this screen (i.e. during updates) roll past the bottom of the box, but the user cannot scroll down to see the results of the update. 2) Minor: Capitalization of each word in "update Plugin" is weird (i.e. first word uncapitalized, second word capitalized).

When it's ready.... (I wanted to post a question, but then I remembered the answer... )

Ah oh course... Will be doing that when I get a chance... Let's call it a sanity test performed over vpn from work...

Not quite sure what that means... The test case that was tested and has failed is "initial setup with sane out of the box defaults are/may be not actually being respected/followed."

Try changing the values in display settings, apply them, and then change them back. The temperature alerts work 100% perfect for me regardless of whether I use Celsius or Fahrenheit. As an aside, with every one that has stated that it doesn't work, the one common denominator is that when its not there is no F or C in the email, which again mine displays perfectly Tried setting the display setting to "F" (apply) and then back to "C" (apply). It appears that it has worked, as I haven't received any emails as yet.. So there's definitely a bug somewhere here... PS. My first ever install of unRaid was 6b9, and am now on 6b14.