gundamguy

I mean the webgui itself, right now you can lock-down things like shares but anyone on the network can simply go to the webgui and change anything they want. It would be nice to have the account system tied into the webgui to prevent unauthorized access, and of course passwords would be pointless without SSL (HTTPS). Please keep in mind I am only asking if anyone is thinking about this/working on this. A version of this already exists if you change the root password under users it will prompt you to log in before it loads the webgui. (This will also effect your log in via Telnet and I think SSL.) I do not remember if this makes it HTTPS... but I don't think it does, because unRAID is designed to be used in trusted environments.

I was under the impression that the green bar in the upper right and corner of Version 6b12 was an status indicator... perhaps I am / was mistaken. Not to hijack this, but this did bring up a few minor visual thoughts about the current interface. Does it really make sense for the Disk Health Indicator to be the same as the Spun Up/Spun Down indicator? Would a disk with an error be red even if it's spun down? For example my parity is spun down right now along with everything else but the cache. So they are all gray... meaning I quickly see if my parity is good, and unless you know that a disk will only be Gray when it's spun down if it's also good... you'd have to spin it up to know. Also when you mouse over the disk status indicator it says two green arrows allow you to spin up the disk... but I don't see that as an option anymore in 6b12.

Based on what Weebo said, that seems likely. That's extremely intresting if that is what's going on... becasue it means that unRAID is using extra ram like a temp cache disk and might reduce the need of a cache disk for systems with infrequent large writes. Now are there any drawbacks to this, for example if your not using ECC ram does this increase the risk of bitrot?

EACH of these has an impact. The 4TB 7200 RPM HGST is capable of 160MB/s on the outer tracks. uRAID6 allows more buffering of writes in the buffer cache, especially with a large amount of ram. Perhaps housekeeping on XFS is managed better. There used to be a single kernel lock in reiserfs for years that hurt performance. The destination drive type plays a big role in this too. I'm on unRAID 5 and I burst at 110MB/s from windows with data on SSD to an HP micro server. There were kernel tunings I used plus I always make sure I purchase a high speed parity drive. Thx Weebo. The thing that confused me was that even with the new parity drive, my parity check speeds still clocked in around 75MB/s. Would the outer tracks not be a factor here also? John I am having a hard time believing this as well. First a speed of 113MB/s is basically saturating a gigabit LAN connection. Second that matches the speeds I get writing to a SSD cache drive. I thougth the parity calculations slowed writes to the array down somewhat... (This was my experience pre-cache) which is the basis of why people use cache drives and mover in the first place. If you can write to the array at the same speed (or better) which you can transfer data over your LAN to the array why bother with a cache and open yourself up to data loss.

This can be done, the better question is how do you do this safely since unRAID was not designed to be internet facing.

Thanks. It would be nice if they stopped making changes like this.

Getting a weird error where even though I just updated Plex to the latest version and it says that version is running that an update button is still there asking if I want to update again... "Installed Version: 0.9.11.6.800-831ffad? Update to Version: dist-ninja/0.9.11.6.800-831ffad" Also today this is the first time I've seen "dist-ninja" and the download went really slow. "Dist-ninja/" was there before the last update as well. Did they change something?

I can confirm that Plex does NOT [automatically] update files from the cache drive. I have Plex set to use /mnt/user and it does not add new shows and movies until after the mover runs at night. Edited your mostly correct post. Plex will find them if you manually scan for new media. Plex may not automatically add them (unknown to me, and not my use case). [Edited for word choice, manually is the better word.]

Probably to avoid running out of memory, or avoid killing the flash drive. But I have 16 gigs of ram and a whole array of drives for storage. If your looking for a better stats package, you could try the Dynamix System Stat's Plugin. It gives you a longer history if you select it from the drop down.

Moving this to the roadmap as it's not a defect, but a feature request. Thank you, still learning where things are supposed to go in the forms here. I think NAS put it way better then I did and addressed even more concerns then my basic one. I guess end of the day I'd like some sort of notice beyond having to dig into the overall system log if a notification attempt fails for some reason.

I'm not sure that's true in all cases. It is known that it's a reporting issue when you are booted in Xen mode, and the actual way to determine CPU frequency in Xen mode is xenpm get-cpufreq-states | grep "current" If you are not booted in Xen mode then it's possible that this is not just a reporting issue. At least I've not heard any authoritative source confirm this is just an reporting issue for non-xen users... yet. I also think the conversation is a little muddied because of different groups of people here (Xen, non-Xen) are giving conflicting reports to all cases "It's just a reporting issue", "It's not just a reporting issue" when it might be the case that it's different for both groups. For the record, I'm not in Xen mode (default boot options) and when I check the cpu frequency using the following code awk '/^cpu MHz/ {print $4*1" MHz"}' /proc/cpuinfo or cat /proc/cpuinfo |egrep -i mhz (Both from another thread about this issue) my cpu seems to be running at max frequency even when cpu usage is minimal and not stepping down. That said I'm not an expert on this or anything linux related, and those methods to extract the cpu frequency could be incorrect, or there could be something else going on. I'd like some feedback on this from an authoritative source for us Non-Xen users (which is why I brought this back up a few posts ago)... would be good to know one way or another if it's just an reporting issue in both Xen and Non-Xen modes, or if it's another issue which needs to be fixed. There is no correlation between percentage of cpu used and actual MHz. The two commands are not showing conflicting information from what I could see. Shouldn’t the processor be throttling? Maybe this is just confusion on my part about what the expected behavior here is and should be, but my understanding is that with modern processors should include some sort of dynamic throttling and not run at the maximum frequency all the time. We know that the webgui is reporting the wrong info and this throttling is occurring if you are booted in Xen mode. What I have not seen conclusively stated yet is if the webgui is reporting the wrong info and this throttling is actually occurring under default boot options. Resulting in the CPU running at max frequency at all times. If someone could provide the correct way to check the current cpu frequency under default boot options I’d be glad to check my system. All I know is that so far using the commands that I’ve seen posted here (not the Xen one since I’m in default boot mode) is that those results support the notion that my CPU is not throttling as I would expect. This post suggests that the files that govern cpu scaling might just be missing. I’ve not checked my system for this yet.

I'm not sure that's true in all cases. It is known that it's a reporting issue when you are booted in Xen mode, and the actual way to determine CPU frequency in Xen mode is xenpm get-cpufreq-states | grep "current" If you are not booted in Xen mode then it's possible that this is not just a reporting issue. At least I've not heard any authoritative source confirm this is just an reporting issue for non-xen users... yet. I also think the conversation is a little muddied because of different groups of people here (Xen, non-Xen) are giving conflicting reports to all cases "It's just a reporting issue", "It's not just a reporting issue" when it might be the case that it's different for both groups. For the record, I'm not in Xen mode (default boot options) and when I check the cpu frequency using the following code awk '/^cpu MHz/ {print $4*1" MHz"}' /proc/cpuinfo or cat /proc/cpuinfo |egrep -i mhz (Both from another thread about this issue) my cpu seems to be running at max frequency even when cpu usage is minimal and not stepping down. That said I'm not an expert on this or anything linux related, and those methods to extract the cpu frequency could be incorrect, or there could be something else going on. I'd like some feedback on this from an authoritative source for us Non-Xen users (which is why I brought this back up a few posts ago)... would be good to know one way or another if it's just an reporting issue in both Xen and Non-Xen modes, or if it's another issue which needs to be fixed.

Is that only in Xen mode that you're seeing this issue? Thanks for looking into it.. For now I manually lowered my multiplier to 12, so at least it's not just wasting energy for no good reason. No, I see this in KVM also. All 4 cores show about 3000 in my dashboard. Same, mine is 3800... OK, we are investigating. With xen, someone else already gave us the command we needed to get the proper CPU frequencies. Fix is being implemented. Still trying to figure out the right approach with KVM. There are pretty major differences between KVM and Xen in this area. I just manually lowered my CPU multiplier to 12 for now.. It'll do for now until a fix is figured out. I don't think there is a "fix" to figure out. The issue is reporting, not actual cpu freq. I used xenpm get-cpufreq-states | grep "current" before and after starting a plex transcode and I can see the freq go from 1400 to 3500. so cpu scaling is working, it just isn't being reported correctly. With xen, its just reporting. With KVM, I'm not sure yet... Any updates on this? Is this actually an issue or just the webgui reporting bad info? Using code provided in this thread it appears that it's an actual issue. I am in the default boot mode which I believe is KVM, so it's not Xen related.

Applies to unRAID b6v12 and up Story Long Version: Last night I was making some changes to my configuration of apcupsd by adding an ethernet connected slave (Windows PC) to my master (unRAID) which is both on the same UPS. After configuring the apcupsd client on the Windows PC I pulled the plug on the UPS (to simulate a power failure). Immediately the unRAID webgui notifies me, via the new in browser notification system, that I am on battery power (awesome). Eventually the Windows PC automatically powers down (awesome) and I plug the UPS back in and almost immediately the unRAID webgui notifies me that power has been restored (again awesome). Now here is where I get confused… during this time I got zero alert emails, so I checked my email configuration again and a test email was sent almost immediately with no problems. So the configuration was good. I asked dlandon if apcupsd is hooked into the new notification system (it is) and after a little bit of discussion I determined that human error strikes again… My router/switch is on battery backup, but my modem is not (might want to fix this.) No internet connection means no ability to contact mail servers which of course means I can’t expect an email from unRAID. Don’t I feel stupid for not thinking about this first… Story Short Version: I shouldn’t be surprised that I didn’t get an email notifying me about power outages / restoration from unRAID when unRAID has no internet connection and therefore is unable to send email notifications. Suggestion: What I am writing about is that I somewhat expected to get a notice in my browser telling me, hey we tried to email you because of this alert but we were unable to. Would it be possible to add a function to the browser notification system which triggers another alert when an alert event happens but the email notice fails to send? I know this seems kind of redundant, but I can think of a few scenarios where this could be useful information. Even more so if you are set to receive alerts via email only. Or maybe a system which queued the alerts and attempt to resend x number of times (2-3, maybe even user customizable) every x min (1-10?) would be awesome... and then after the final try it still fails it would give you an in browser alert, notifying you that something prevented your email alerts from happening and check that out again to make sure it's working or makes sense that it was unable to send. Sorry I get long winded at times.

Looks good. I am a fan of these Mini builds. I also built a Node 304 Unraid Box ealier this year. I used a slightly different MOBO but also with 6 SATA III ports. I also decided to go with the Intel I3-4130T 2.90 Dual Core CPU because of its energy efficiency, have had no problems yet with Plex transcoding of HD streams, but thankfully I have room to grow if desired. Your CPU should be no problem for your desired purpses. One thing that bugged me about my Node 304 build (and most Mini-itx builds) was due to limited HD bays at most you are going to have 1 Parity Disk and 5 Data Disks (with no Cache). Almost immidately after finishing my build I discovered just how useful and awesome Cache disks are. However I also wanted to have 5 possible data bays (2 are still empty) to maximize my storage space and didn't want to give one up for a Cache drive which needed to be at most 500 GB. My solution was to buy a Apricorn Velocity Solo x1 off of Amazon and an 250 GB SSD and have my Cache in my PCIe slot and not taking up a drive bay. I wasnn't 100% sure this would work but did a little research on the chipset included and went for it. Plugged it in and it Worked without any additional effort on my part. Note if you are thinking of doing the same the Velocity Solo x2 has faster read and write speeds but uses the Marvel 9182 chip which I'm not sure is supported by the Linux Kernel used in Unraid... yet. Overall I like how my Unraid box looks and functions. It had to be small, make little noise, and look good because I live in an apartment with my wife so it was going somewhere it would be seen but not heard.

I also just deleted my stale plugins via the webGUI. After deleting the last one, the tab showing stale plugins also disappeared. The folder called "plugins-stale" remains, but is now empty. Ah ha, that's the issue. I was expecting them to show up like they did before and now they are on a new tab that I must have over looked. I'll check later and verify that it was just an oversight on my part. Will / should stale O&S's show up in this location too?

You are not wrong. They don't do anything, they also don't take up a lot of space. It's more an expectation of mine that if I can click update / install and add files to my flashdrive via the webgui, I'd like to be able to remove / delete old files via the webgui. I wouldn't even have noticed this at all if the functionality didn't exist in beta610 (and prior to that I think). It isn't a big deal, but that's also why I am surprised this feature went missing, as it doesn't seem hard to implement either. A couple of additional thoughts on plugins. Does the status check run on first boot? Based on what I've seen it seems statuses for plugins are all "unknown" after a restart / power-up until I first go to the webgui plugins tab and hit check for updates. Would it be possible / make sense for this to run once when unRaid boots? That would seem just a little more user friendly. Additionally, plugins are moving to being installed via the plugin tab, which is great, but what was a little counter intuitive to me way back when I first started using unRaid was that I didn't go to the plugins tab to check on the status / settings of my plugins. Update status is contained on the plugins tab, but why couldn't the plugin name / image be the link to that plugin's specific page? Wouldn't that be more intuitive to new users?

Just noticed, in the Plugin Manger in 6b10 there as the option to delete stale plugins (old versions which were archived from updates) that doesn't seem to be the case in 6b12? Any thoughts on adding this functionality back in? Also similarly O&S updates done via the plugin manager tab are archived on the flash as well, anyway we could have an option to remove those under a stale plugin / O&S listing in the plugin manager? I don't think this ever existed even in 6b10. It's not difficult to delete them from the flash drive, but I've been doing that by first enabling SMB sharing of the Flash and then deleting the files, then disabling SMB sharing... and would prefer something more native to the webgui.

Good to know, didn't realize that was happening. Been a while since I saw the boot menu. So if I'm booting into Unraid with KVM support, how can I correctly check CPU speed?

I saw that as well but I suppose I'm confused here because I didn't think I was using xen. My desired configuration is to use vanilla Unraid without vitalization. Treat me like a total noob here, I'm really a later comer to the Linux bandwagon and learning the ropes as I go. So maybe I just have a total misunderstanding of how Unraid works, but I thought xen with unraid 6 in dom0 was a option not the default?

I want to say thanks for this. Following PhAzE's directions and using your code, I was able to get plex working again much easier then I had originally anticipated. For others looking to do this but unsure what to do the steps were were first navigate to Go file (under config on flash drive) then copy and past code (correcting plugin names) into Go file below existing code save then reboot.

Installed the 6.0-beta12 just now. CPU speed says 2900 (max) just about all the time. Not using XEN or KVM. Just idling. When I tried some code someone posted earlier in this thread to check the CPU watch -n .5 grep \"cpu MHz\" /proc/cpuinfo it appears that the CPU cores are actually running at that frequency. This seems to be a problem, but I am also not skilled in Linux so maybe I just don't know what I should be seeing.

I am super excited for this! I was going to ask if you had any plans to make your Plex Plugin compatible with the Plug-in manager in Unraid 6, and I am extremely happy to learn that you are already working this!

First I want to say thanks for the great plugin's. Your Plex Plugin is awesome, i just updated my plex install with such ease, and it's using the latest Debian release so we don't have to wait for an unraid specific one anymore! Double awesome. Now for my issue, perhaps my config is messed up or something but all of my TV metadata and only my tv metadata disappeared Or at least all of the Posters and Banners did. Is this something I did wrong or how can i check this? BTW your app did say my config would be persistent. Also I'm running it off of a cache only share, any ideas of what I did wrong or how to solve this issue for future updates?

Trying to install your 64bit plex plugin, but it gave me the following error. Warning: simplexml_load_file(): I/O warning : failed to load external entity "/b oot/config/plugin/plexmediaserver.64bit.plg" in /usr/local/emhttp/plugins/plgMan /plugin on line 125 plugin: warning: no name attribute, installing anyway Warning: simplexml_load_file(): I/O warning : failed to load external entity "/b oot/config/plugin/plexmediaserver.64bit.plg" in /usr/local/emhttp/plugins/plgMan /plugin on line 125 plugin: xml parse error Did I do something wrong? Wait, got it to work the second time. Not sure if this did it, but I decided to seed the packages folder with the latest deb release of plex before trying again, then it worked.