Squid

I decided to clear house because after rebooting all my dockers were stuck in "needs to update" but they couldn't update. EDIT: and I am an idiot, not sure why it was set to itself instead of my router 10.10.1.1.... but it has been like that for a couple years without issue.... I am 6.2b21 Sounds more like you need to set static DNS addresses in network settings sorry if this sounds dumb but my 5.x/6.x install has been set to automatic for a few years (my router is setup with a static IP), why would this change? last weekend I did a motherboard/cpu/ram swap, but even after that its been fine until today? Also, My DNS server settings have never changed, which i should clarify is set to my static ip (10.10.1.3) I would still set it up (8.8.8.8 and 8.8.4.4) - can't hurt, and I can say that the feed is definitely up. Using automatic settings for DNS within unRaid is usually OK, as then the router handles it. But, in my experience, most of the time the DNS settings that your ISP chooses suck (which is probably how your router is set up), and google's DNS addresses are always better. Also, since you got the message you should upgrade CA to the latest version, as that message is different (and more detailed) on the later versions. The version which you are running, if (for whatever reason) the appfeed fails (usually a connectivity issue) and CA reverts to legacy mode, if *any* of the repositories fail to download, then CA is unusable. On the later versions, CA will still function and just ignore the repositories which failed to download. As a secondary note, if your download bandwidth is completely maxed out, CA does NOT take as much time as needed to download its required files (either the appfeed or legacy mode individual repositories), and will abort the downloads if it takes too long. This is primarily because of how unRaid and emhttp work. Let's assume that for whatever reason the appFeed is going to wind up taking 20minutes to download. Without the automatic timeout (30 seconds), the entire unRaid GUI will wind up being unavailable for the 20 minutes.

I decided to clear house because after rebooting all my dockers were stuck in "needs to update" but they couldn't update. I am 6.2b21 Sounds more like you need to set static DNS addresses in network settings

Plugins - Check For Updates - Install the dynamix webUI update - Problem Solved

Download and destination folders are set by the download client - NZBGet / SabNZBD

That's an issue with your docker.img file (loop0)

My understanding is that since unRaid does not leverage hardware raid, the HBA will never signal that a drive is offline. (could be mistaken and the HBA *may* be smart enough to know that a drive has dropped offline and then signal). Easy way to test would be to pull the sata / sas cable from the back of the cage and see what happens with the error lights

within your go file (/boot/config/go) you would add something like cp /boot/myScript.sh /etc/cron.hourly/myScript.sh Using the .cron method is probably preferable, but tossing scripts via go into cron.hourly / daily/ weekly / monthly does allow you to easily adjust the time of execution fast (albeit for all of them at a time) And once you have the schedules plugin installed, under settings / schedules there will be a new tab called Fixed Schedules

You can either use the .cron method as described in wgstark's link, or alternatively if you want it to run hourly and be able to use the scheduler to manage it, within your config/go file, copy the script to /etc/cron.hourly When you go to fixed schedules, you will see your script listed there.

Did you set a location for the image file

Decent idea actually. I click it a lot during testing (sometimes not on purpose), and I usually use it as an excuse to go have a smoke. So I can see how other people would do the same thing. I've quit the fags, stop putting temptation in my way.... So you're back to straight now That was you 1000th post! Such technical insight, what a milestone... lol Personally, I think spending the milestone making fun of you is a perfect use of it. I can't wait til I hit 5000. Face it, next to sparklyballs you are a good whipping post

Deleted my docker.img file, and created a new one and it is working now Perfect. Strange why you had to do that but at least it worked and LTs implementation of docker allows for very quick restoring without recreating appdata

Decent idea actually. I click it a lot during testing (sometimes not on purpose), and I usually use it as an excuse to go have a smoke. So I can see how other people would do the same thing. I've quit the fags, stop putting temptation in my way.... lmao. ^^^ My boss keeps telling me to quit. I keep saying that if I didn't have to deal with him it be easy

This is the actual support thread: http://lime-technology.com/forum/index.php?topic=34009.0, but smdion isn't around that often, and all he did was create a template for a google application https://hub.docker.com/r/google/cadvisor/

On 6.1.9 if you search for cadvisor and install it, there's nothing you can ever do to get the webUI to ever show up in the drop down (known bug - fixed in 6.2beta 20) Not really a clue about what's going on with the container. I have never had any trouble with it, the defaults work 100% (since what they are doing is referencing unRaid system paths) While this may seem a bit weird, the internal workings of docker containers (the dockerfiles, how to make them, etc) I actually have zero clue about Quick search of these forums showed zero problems getting cadvisor to run using smdion's template (which is what I am), and I've installed, deinstalled, reinstalled, etc this on both my servers, so I'm not quite sure what to say exactly While my initial response would be to reset your server and/or restart your docker.img file, the net result is that you're going to get roughly the same information as what the resource monitor displays (cadvisor breaks it down further however) in a far far far prettier way.

Everything matches mine perfectly. And the dashboard shows that its running. I know that in my testing with cAdvisor it would occasionally take a minute to boot up the GUI. I would try again it again. Failing that (and I doubt that this will make a difference), remove the container and image through dockerMan, then search CA for cadvisor and install that (its identical except for a different host port) No idea what's up with the log not displaying, but it seems to be a recurring theme with 6.1.9 every once in a while

Can I ask for a bit of help with cAdvisor in this thread? When I click on a docker, the webpage will never load, and I end up getting a "ERR_CONNECTION_TIMED_OUT" message. If I click on "log" in the docker page for cAdvisor, it will not load. Do you need any more information, some logs or some screnshoot to be able to help me? I guess that here is as best a place as any, since it's me who's pushing it (but it is out of smdion's repository) I guess start with another screen shot of the volume mappings and the port. Are you able to hit it from the dashboard and selecting the WebUI for cadvisor?

Decent idea actually. I click it a lot during testing (sometimes not on purpose), and I usually use it as an excuse to go have a smoke. So I can see how other people would do the same thing. I click on it every time I make a change to one of my repo xmls to see if it looks fine. It only takes about 30 seconds for me Your mileage varies. Usually, my bandwidth is consumed by other containers running so it takes it a bit BTW, you realize that you're able to set up a "web-hook" into the appfeed so that whenever you make a change or addition into your repository the feed will automatically run a special scan just for you and then be available to the community at large within a minute or two http://lime-technology.com/forum/index.php?topic=41413.msg392752#msg392752 (But if you do this, I would chat up KodeStar on lsio's IRC channel for assistance as you do it - not sure if the instructions are 100% up to date) EDIT: Confirmation button on forced updates done. Will be released on the next release (sometime next week unless bugs show up elsewhere)

Decent idea actually. I click it a lot during testing (sometimes not on purpose), and I usually use it as an excuse to go have a smoke. So I can see how other people would do the same thing.

All paths are already correct as is (no modification needed, and you'll probably break things if you touch them) There is no appdata for cAdvisor. Only thing is modify the host port (9243) if it happens to conflict with something else (pretty much I picked it out of a hat so you should be good). Guess I should add a note that says no modifications at all needed. I don't want to bypass dockerMan completely and install it manually, because I do want it to be able to handle cAdvisors GUI, etc through the dashboard and what-not. EDIT: You actually gave me an idea though. On 6.2, I can set those mappings to be hidden so that nobody will even know that they are there under normal circumstances Thanks! Yeah, that is a good idea. What more information on the resource monitor do it get by installing cAdvisor? I did install it, but forgot to check before. I tried to delete it, but nothing got removed. Found a "bug", If you click the "update applications" button, there is no way to cancel it. So if I press it, then I x out the webpage because I want to undo it, the unraid gui will not load before 2-3 minuttes. It should still be possible to access the unraid gui, when the plugin is doing that update. If it is possible to fix, it would be perfect After cAdvisor is installed, within the resource monitor click on one of the icons for the apps and see for yourself. cAdvisor itself uses no resources (or next to nothing), and is something like a whopping 40meg in the docker.img, so its well worthwhile to have it installed, (and even if you never look at it, you'd never even notice the fact that its installed) As far as the Update Applications is concerned: #1 - The only time you ever need to hit that button is if you KNOW that there's a new application that hasn't yet been added to the appfeed. The appfeed itself is updated every two hours with all the latest and greatest, and CA checks everytime it loads to see if there's an update to the feed. If there is, then it automatically downloads the feed. This whole process takes around 5 seconds tops. #2 - Update applications does take a bit to run. There is no way to actually stop it short of outright killing the process. There are fail-safes however built into it. It will not try and take forever to download the particular repository, but will abort the download if it takes too long. In this case, a note pops up stating which downloads it aborted. #3 - Ultimately, the base problem here (and this issue is not limited to CA, but is across the board with unRaid) is that emhttp (the webUI engine) is completely a "single threaded" program. It does one thing at a time, and any other requests are processed in turn. This means that once it starts a process, that process runs to completion before the UI will respond to another request. For another example, max out your internet download speed with something like NZBGet. Now download a new docker app. While its downloading (and taking its sweet time because of the limited bandwidth available), try and open a new tab and go back to CA or anything else with the webUI. You won't be able to until the docker download is completed. Closing the tab won't abort the installation. All this being said, I have been planning some upgrades to the Update Applications button (more exactly to how the system downgrades itself to doing the same routing as what Update Apps button does). But in the meantime, just don't hit the button. Its there for a reason, but 99.9% of the time when people hit it, they are hitting it for the wrong reason. (Why the pop up more or less says Don't Hit This). To be quite honest, if I was in control of the application feed (and I'm not), the button wouldn't be there at all.

All paths are already correct as is (no modification needed, and you'll probably break things if you touch them) There is no appdata for cAdvisor. Only thing is modify the host port (9243) if it happens to conflict with something else (pretty much I picked it out of a hat so you should be good). Guess I should add a note that says no modifications at all needed. I don't want to bypass dockerMan completely and install it manually, because I do want it to be able to handle cAdvisors GUI, etc through the dashboard and what-not. EDIT: You actually gave me an idea though. On 6.2, I can set those mappings to be hidden so that nobody will even know that they are there under normal circumstances

There's different blake2 algorithm's. File Integrity Plugin uses blake2 (full version) Corz uses blake2s which is slightly faster to generate and also much shorter in length. Unfortunately, they are incompatible, and IIRC corz when it sees the full blake hash thinks that its SHA256 Also, without the extra information within the hash files (namely the lines starting with #), you're probably going to be best off with using md5 as I'm not 100% sure if corz will recognize using SHA without the comment lines (but it will recognize md5)

Nice change log buddy lol

Fixed: Under certain specific circumstances, internal data structures could get corrupted Further enhancements to security

While I'm glad it's working it doesn't actually help me figure anything out as to what happened to a couple of you as that's basically the same thing as an upgrade

PM'd you the file. Thanks. There is some extraneous entries at the end of it, but it still looks like a valid json file, but while I'm at work, pretty much all I can do is VPN into my server and see if it works (which it is - on both my servers 6.2b20 and 6.1.9). Not going to attempt comparing the two files over my phone. If anyone else is capable of sending me the file, I'd appreciate it. Also, does the problem disappear after a reboot.