BVD

There's a pretty big reason that there can't (well... "won't" would probably be more accurate) be a single all-inclusive backup of everything in unraid - different applications/files have differing requirements for ensuring file consistency. A good starting point for your research would be looking for information regarding application consistent vs crash consistent snapshots/backups. In short, a crash consistent backup will just ensure that the filesystem of the OS itself is intact. An application consistent backup is one from which you can restore the application's specific data in a manner it can utilize, which means that your backup has to take into account that application's specific requirements. Each application's consistency requirements are unique to that application. The only way to get around it easily would be to, at the start of the backup, ensure all applications, docker containers, and VMs are shut down, and remain shut down until the backup completes. As one example: Lets take a relational database, as these are fairly commonly used on unraid; at a basic level, this type of database has two components, the tables (db file), and the transaction log, both of which are required in order to use the db, and are interdependent upon eachother... If you were to take a snapshot of the filesystem that the db resides on without first ensuring that the data between the db and transaction log is consistent on the disk (meaning no pending actions in memory, which there pretty much always are unless you have either a very low resource use case, or have stopped the application entirely), you'll end up losing data. Usually application downtime isn't an option, so backup tools instead use other mechanisms to quiesce IO to the DB to first create a snapshot, then back it up. While the filesystem itself would be consistent if you were to use just any backup tool, the application may well not as the application depends on the OS for it's own consistency, and that backup doesn't include anything in memory, and as far as the OS is concerned, 'the stuff that's on the disk is good'. The level of complexity involved in backing up everything in a virtualized environment is pretty significant... That's why companies like cohesity, rubrik, veeam, etc charge such big bucks, because they have to integrate with a million different things (VSS for microsoft stuff, inc server OS, SQL, exchange, vix api to backup vmware virtual machines, various oracle components in order to understand RMAN, nutanix API to backup AHV snapshots, and so on) to ensure that each one is backed up in a way that, when you go to restore, you get what you expected.

There are already 3-4 (I think?) onlyoffice templates available, I'm not sure another would get approved... But that's almost certainly something that the Pi community would have interest in if they dont already have a go-to alpine onlyoffice, I'd hit up that community and gauge interest!

I'm going to assume you've probably not taken a backup prior to the upgrade then, as it's not something that's automatically done... I'll try to take a look in the morning to see where the plugin/hub configuration is loaded from - I've an idea that you could simply remove that file from the directory and restart the container in order for it to be regenerated, but I'm not certain and it's about 0130 here. Will try to remember to check it out once I wake up

You'll just have to restore your backup is all, if you find no way to manually remove the plugin - though I'd assume theres a php file somewhere that you could edit out the reference to it manually to address. Youd have to do a little poking around in that one though, restoring your backup would definitely be far simpler and quicker

Right after an update, the cache will usually need to be rebuilt, as the DB indices are updated and the existing cache is essentially no longer 'valid' (the cache doesn't get it's existing contents updated). I'd give it some time and re-evaluate. As to the phone details, you can configure them manually as well within config.php (same place you put the DB connection information etc) 👍

That's calling out the SMB PHP module, which I'd assume is an installed plugin - if you uninstall the plugin prior to your upgrade, do you still have issues?

/config/nginx/site-confs/default

At first, it almost certainly must feel that way to most, I'm sure. But let's take a quick look at what all one can do with nextcloud (both by default, and by adding plugins from the hub) by thinking about what other containers we'd need in order to accomplish similar tasks: Syncthing (and various others) - Sync files between multiple devices automatically, retain backups of files/file versions, and so on Youtransfer - web sharing of files to guests/etc MakeMKV (ffmpeg) - Convert video files (seems ridiculous to have this in 'the cloud', but it's surprised me by being ridiculously handy) Wiznote - Collaborative notes co-editing/sharing system GooglePhotoSync- Except not just photos, and not limited to just you; anyone on your cloud can backup/sync all their google data (drive/photos/calendar/contacts) with a single server setup Jitsi - Video conferencing, instant messaging, and screensharing (I'm really surprised at how well this works... I'd honestly expected it to be a crippling amount of load to run, but the user machines actually handle the vast majority of it. Very well implemented!) Kanban - Project/task management/tracking playSMS / Google Messages - Send text messages from your laptop/desktop to all those contacts you're automatically backup up each night to nextcloud from your iOS / Android device that has the mobile app installed CherryMusic / Airsonic (etc) - Fairly decent music streaming app; allows you to use the media keys on your keyboard for control, and so on The list goes on - frankly, nextcloud's power sometimes means danger for those who aren't willing to invest the time into configuration and tuning, as simply installing all the plugins to do all-the-things while doing nothing to tune them to their environment will usually result in an atrociously performing cloud mess. The amount of effort one has to go through to get a full nextcloud+plugins deployment (as opposed to just the base file sharing/sync functionality) operating smoothly and efficiently shouldn't be underestimated. However, if you're willing to invest the time into it, tuning and tweaking the various components involved, you'll have a cloud that can rival almost anything out there from billion+ dollar companies. As long as one goes into it with their eyes open, understanding that they'll have a little homework in tuning php, postgres (it's far more performant for me than maria in my exp), nginx, and occasionally storage and network stack, they'll be mightily pleased with the outcome once it's finished.

@skois idk if you ever got this answered, but here's the format in case you still need it: root@048509f9b924:/# sudo -u abc /usr/bin/php -f ./config/www/nextcloud/occ config:app:set previewgenerator squareSizes --va lue="32 256" Just open the console, run that, and you're set (modify to suit) - the issue with yours above is: * Since their shell defaults to bash, you must first call php - bash doesn't understand the command context as it's php input (the application itself is being modified, so we're in effect 'writing in php' here - pretty much everything in this docker is web code, written in php) * all executions of code by the LSIO docker are done by user 'abc' - as this is the user running the web apps, we use the same user to make changes to those apps, and - * when executing under that user, it doesn't have 'occ' as an available command in /usr/bin - abc is a restricted/non-root account, and because that user doesn't have the command linked in bin, you have to call the source path for the command. If you're more familiar with Windows, you could think of this like running an application that's only installed under another users Windows profile (sorta...) If you didn't already have it sorted, hope this helps!

The error 43 is the exact driver shenanigans they pull with video cards, but with virtual functions its unfortunately even more difficult to get around. A couple things you can try: * instead of running the full installer, open the executable (.exe installer) with 7zip or the like, and manually install the drivers for your OS (using the .inf files) * test against a different version of windows 10 to verify whether its locked to a specific build type (education, pro, pro for workstation, or an enterprise LTS build) * read through the release notes for earlier driver versions in order to find one that has the best chances for success; 2017 or earlier is likely your best bet in my experience - note, this may require a firmware downgrade of your card, which I've very limited personal experience with

What version of windows 10? It's like I said with mellanox drivers, they're all kinda tied up in licensing crap... My suggestion would be to use something like 7zip to manually extract an older version of the drivers from the executable and then attempt manual installation. For windows and virtual functions, intel (and some chelsio) are really the only sure fire way to ensure compatibility thanks to the crap nvidia and mellanox have pulled with their drivers.

And you're only just scratching the surface! Wait till you start looking at latency numbers; I think you'll be pleasantly surprised when you start looking at nano instead of milliseconds My work life has kind of run over all my other projects the last few weeks so I've not been able to really dig back in as of yet - but I'm certainly glad to see you experiencing some of the benefits already! Let me know if you encounter any issues or have any questions as you're working on it - I'm happy to help 👍 @trott there's no direct way to do so via the unraid UI - I'm not sure why one would want to do so though honestly. If it's for security purposes, then it'd be best suited to being in a VM anyway, at least as far as overall security scope is concerned. A VM, at least IMO, is inherently more 'secure' than a docker container as there's far less attack surface area from the VM to the hypervisor than from a docker container to the hypervisor.

Totally understood! Moreso posting to try to help any others that might end up seeing issues and come in with such a question 👍 Would it be worthwhile to have an open access GitHub page or something where all of the 'known' scripts could be stored, then folks can just edit to update them as/when needed? Or is that something that'd be discouraged at this point?

Hey @Squid - I couldn't find if this had already been addressed elsewhere or not (if so, my google-fu sucks): As of at least 6.9, the top of the script needs a tiny edit, from: require_once("/usr/local/emhttp/plugins/community.applications/include/xmlHelpers.php"); To: require_once("/usr/local/emhttp/plugins/community.applications/include/helpers.php"); For anyone getting something about an error at line three, calling blah/blah/xmlHelpers.php when they run the plugin integrity check script, the above is the quick fix (hoping to catch some search keywords with this sentence to avoid repeat requests where possible 👍

I've seen various requests for it in all number of threads, but never explicitly as it's own topic... I might've missed it though, for sure. My hope is that they'll feel the time is right at this point to actually implement one - the team's grown significantly since version 6 was released, and the development community has undergone significant change and growth as well. I feel like having a roadmap available to us could only help accelerate adoption at this point, now that they've the additional headcount as well as the expanding number of those in the community who are willing to volunteer their time to contribute via community applications. Without a roadmap of any sort, it's possible (likely, in my opinion) that efforts will be duplicated, and that someone's work's value will be short-lived and lost upon a later release when whatever project/feature they created a tool (plugin/script/etc) for is implemented in the base OS. More features across a wider pool of developers, I'd think at least, would be more than enough to justify it. To me, it's not about 'making promises for deadlines', so much as 'when this release is done and made GA, it'll have this thing'

I wasn't sure where to post this kind of thing - it's not exactly 'forum feedback', nor is it really a 'feature request for UnRAID' (if the capitalization is wrong here, please do let me know - I may've been writing it wrong all this time 😕 ), but more a 'business feature request' I guess... An ask of the team as far as planning goes I guess? The request: For Limetech to publish a Feature Roadmap which gives a generalized direction for wherever Limetech is planning to take the core OS over the next (however long you feel is realistic). What would be super helpful (and hopefully more realistically possible now that the team's grown a bit!) is to publish a Feature Roadmap, updated perhaps once per every major release (6.8, 6.9, 7.0, etc), or every 3-6ish months (whatever makes sense based on your internal structure and cadence), which outlines the 'big new things' planned for the next (however many releases can be semi-accurately forecast). _____ Some background: A Little Justification: Ideas for what it looks like: <To be filled out once I get back from the day out with the fam>

Now I can't get past the initial scan screen:

It recognizes the drives and displays them in bold - it just doesn't pull the images. I can also confirm that I can pull the images directly from the server with wget DiskSpeed - Disk Diagnostics & Reporting tool Version: 2.8.1 Scanning Hardware 13:56:19 Spinning up hard drives 13:56:19 Scanning system storage 13:56:23 Scanning USB Bus 13:56:31 Scanning hard drives 13:56:35 Scanning storage controllers 13:56:38 Scanning USB hubs & devices 13:56:38 Scanning motherboard information 13:56:38 Fetching known drive vendors from the Hard Drive Database 13:56:41 Found controller WD Black SN750 / PC SN730 NVMe SSD 13:56:41 Found drive WDS500G3X0C-00SJG0 Rev: 111110WD Serial: 20359E805968 (nvme1n1) 13:56:41 Found controller WD Black SN750 / PC SN730 NVMe SSD 13:56:41 Found drive WDS500G3X0C-00SJG0 Rev: 111110WD Serial: 203686802824 (nvme2n1) 13:56:41 Found controller C620 Series Chipset Family SSATA Controller [AHCI mode] 13:56:41 Found drive Samsung SSD 860 EVO Rev: RVT01B6Q Serial: S3Z6NB0K301833Y (sdb) 13:56:41 Found controller C620 Series Chipset Family USB 3.0 xHCI Controller 13:56:41 Found drive SanDisk' Cruzer Fit Rev: 0 Serial: 4C530000030627220171 (sda) 13:56:41 Found controller WD Black SN750 / PC SN730 NVMe SSD 13:56:41 Found drive WDS500G3X0C-00SJG0 Rev: 111110WD Serial: 20359E804913 (nvme0n1) 13:56:41 Found controller C620 Series Chipset Family SATA Controller [AHCI mode] 13:56:41 Found drive Seagate ST16000NM001G Rev: SN03 Serial: ZL28N8NB (sdc) 13:56:41 Found drive Seagate ST16000NM001G Rev: SN03 Serial: ZL28BYR3 (sdd) 13:56:41 Found drive Western Digital WD6003FFBX Rev: 83.00A83 Serial: V8K0NJVH (sde) 13:56:41 Found drive Western Digital WD6003FFBX Rev: 83.00A83 Serial: V8K0EHLH (sdf) 13:56:41 Found drive Western Digital WD6003FFBX Rev: 83.00A83 Serial: V8K0MVMH (sdg) 13:56:41 Found drive Western Digital WD6003FFBX Rev: 83.00A83 Serial: V8K0EJKH (sdh) 13:56:41 Found drive Western Digital WD6003FFBX Rev: 83.00A83 Serial: V8H3PZSR (sdi) 13:56:41 Found drive Western Digital WD6003FFBX Rev: 83.00A83 Serial: V9H7HG6L (sdj) 13:56:41 Found controller WD Black SN750 / PC SN730 NVMe SSD 13:56:41 Found drive WDS500G3X0C-00SJG0 Rev: 111110WD Serial: 205115802117 (nvme3n1) 13:56:42 Fetching drive images 13:56:42 Fetching Drive Platter Information 13:56:42 Checking Hard Drive Database for drives 13:56:42 Configuration saved I can also confirm that I can pull the images from the command line without issue on the host: root@milkyway:/# wget "http://strangejourney.net/diskspeed/Drives/ADATA/ASP580SS3-120GM-C/Johnathan_Bartlett_154B_007A_D97H_E03000000471.png" --2021-03-10 13:58:45-- http://strangejourney.net/diskspeed/Drives/ADATA/ASP580SS3-120GM-C/Johnathan_Bartlett_154B_007A_D97H_E03000000471.png Resolving strangejourney.net (strangejourney.net)... 205.210.191.162 Connecting to strangejourney.net (strangejourney.net)|205.210.191.162|:80... connected. HTTP request sent, awaiting response... 200 OK Length: 41097 (40K) [image/png] Saving to: ‘Johnathan_Bartlett_154B_007A_D97H_E03000000471.png’ Johnathan_Bartlett_154B_007A_D97H_E03000000471.png 100%[================================================================================================================================================================>] 40.13K --.-KB/s in 0.09s 2021-03-10 13:58:45 (456 KB/s) - ‘Johnathan_Bartlett_154B_007A_D97H_E03000000471.png’ saved [41097/41097]

Pulled and attached Found something interesting in the host side logs - seems to hit this once for each disk attached: at lucee.runtime.extension.RHExtension.toResource(RHExtension.java:677) at lucee.runtime.extension.RHExtension.<init>(RHExtension.java:194) at lucee.runtime.config.XMLConfigWebFactory.loadExtensionBundles(XMLConfigWebFactory.java:4369) at lucee.runtime.config.XMLConfigWebFactory.load(XMLConfigWebFactory.java:450) at lucee.runtime.config.XMLConfigServerFactory.load(XMLConfigServerFactory.java:174) at lucee.runtime.config.XMLConfigServerFactory.newInstance(XMLConfigServerFactory.java:125) at lucee.runtime.engine.CFMLEngineImpl.getConfigServerImpl(CFMLEngineImpl.java:788) at lucee.runtime.engine.CFMLEngineImpl.<init>(CFMLEngineImpl.java:270) at lucee.runtime.engine.CFMLEngineImpl.getInstance(CFMLEngineImpl.java:623) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:498) at lucee.loader.engine.CFMLEngineFactory.getEngine(CFMLEngineFactory.java:1589) at lucee.loader.engine.CFMLEngineFactory.initEngine(CFMLEngineFactory.java:395) at lucee.loader.engine.CFMLEngineFactory.initEngineIfNecessary(CFMLEngineFactory.java:278) at lucee.loader.engine.CFMLEngineFactory.getInstance(CFMLEngineFactory.java:174) at lucee.loader.engine.CFMLEngineFactory.getInstance(CFMLEngineFactory.java:212) at lucee.loader.servlet.CFMLServlet.init(CFMLServlet.java:42) at org.apache.catalina.core.StandardWrapper.initServlet(StandardWrapper.java:1188) at org.apache.catalina.core.StandardWrapper.loadServlet(StandardWrapper.java:1132) at org.apache.catalina.core.StandardWrapper.load(StandardWrapper.java:1021) at org.apache.catalina.core.StandardContext.loadOnStartup(StandardContext.java:5085) at org.apache.catalina.core.StandardContext.startInternal(StandardContext.java:5397) at org.apache.catalina.util.LifecycleBase.start(LifecycleBase.java:145) at org.apache.catalina.core.ContainerBase$StartChild.call(ContainerBase.java:1410) at org.apache.catalina.core.ContainerBase$StartChild.call(ContainerBase.java:1400) at java.util.concurrent.FutureTask.run(FutureTask.java:266) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) at java.lang.Thread.run(Thread.java:748) TrialUser_20210310_130033.tar.gz

Yeah... if collecting diagnostics hangs, it's almost definitely not one of the components of nerdpack causing it. More likely that sometime between the last boot and now, the thumb drive got bumped, or as Squid noted, it's on it's last legs.

I've seen some weirdness as well with seemingly random items that use libvirt... Some of the nerd/dev pack tools are older than the versions built in to UnRAID, and can cause some pretty odd behavior. When you removed and re-installed nerd-pack, did you reboot in between? Since the OS is all in-memory, you can uninstall the plugin and reboot to load bzroot and bzimage fresh to validate whether or not nerd-pack is related in some way. I had to do so with a couple driver hacks that required dev-pack tools at one point, as the version in the dev-pack repository didn't jive with the system well. If this 'fixes' it, then when you to to re-install the individual components inside of nerdpack that you need, you can incrementally add them to find which specific component is hanging you up.

We can take this to DMs - I shouldve done so in the first place, come to think of it. In short though - complicated might be an understatement lol.

It does have internet access - in the event it was related to some funky docker network issue, I've tried setting it to Host, setting the port to one I've verified isn't in use anywhere on my network per pfSense, etc... However, one weird thing is, the downloading of the images is actually working fine; the .png files exist. This led me to look at permissions, and I found something odd: drwxrwxrwx 1 65534 users 0 Mar 9 22:38 DiskSpeed/ drwxrwxrwx 1 nobody users 34 Feb 2 14:25 Authelia/ It's showing the 'nobody' user differently than any other folder or share displays it's permissions. Changing the permissions fixes nothing as it just recreates them back to the same thing. The parent directory (appdata) shows up as everything other than the diskspeed folder does. Tried the beta tag (it was the only other one available, but I figured it was worth a shot), no dice. Finally, I disabled all data collection from other areas (disabled my syslog server, disabled telegraf), but again, no change. With each change, I've wiped and allowed it to recreate the DiskSpeed folder contents so there wasn't any interference between one run and the next. I've added the debug file here. TrialUser_20210309_225359.tar.gz

Hey @jbartlett - handy tool, thanks! Small issue for mine I hope you can help with - None of the images are showing up, though every drive is in the DB: Looking at the drive details, I can't find any reason for this, as the model, serial, vendor, and fw rev all line up: The logs show nothing of note that I could find (all just loading module stuff), but here's the startup info: 2021-03-09 13:42:29.622 Start CFML Controller 2021-03-09 13:42:29.624 touched monitors =================================================================== WEB CONTEXT (cbe856ff790c9ba5208811309bdf168b) ------------------------------------------------------------------- - config:/opt/lucee/web (custom setting) - webroot:/var/www/ - hash:cbe856ff790c9ba5208811309bdf168b - label:cbe856ff790c9ba5208811309bdf168b I had the same thing on 6.8.3, and just never really dug into it, but now have had the same on 6.9 and 6.9.1... Any idea what's going on here? I'd think the controller wouldn't matter since it's using the onboard chipset and smart is accessible, but in the event it matters, this particular host is using the intel C621 onboard. I'll try and get the diagnostic output either tonight or tomorrow in the event there isn't anything glaringly obvious in the above that I'm just missing. Thanks!

@ConnectivIT looking for some feedback whenever you have a moment - - VF driver installation in MS OS's pretty much always possible, the process just varies a bit depending on which OS (8, 10, server 2019, etc), which version (home, pro / pro for workstation / pro education, IOT, enterprise enterprise IOT), and which release (1809, 1903, etc). I was planning on doing a separate guide for this, but then also figured most people using this one probably would be doing windows drivers as well... what're your thoughts on putting it in a separate guide and linking to in from here as an "optional next step" vs just adding it to this one directly? - I'd intentionally left off MLNX from the initial list here for the reasons I'd noted above, but if you're comfortable supporting it / providing guidance on it, I can add it to the main doc and reference you in it (as both credit to the additions and a contact for issues)? Idk how familiar you are with SR-IOV in the field, but if not quite there yet, I could always just reference your comments and link to them instead for now? - Performance is a HUGE subject with all of this, and I'm working on another topic related to performance/tuning with SR-IOV. For instance, RDMA (SMB Direct in MS terminology), which provides wire level latency for storage over the network, is only available to Pro for Workstations and Enterprise builds; many other nuances. But unlike VF creation, the parameters used to implement any filters/tuning are completely different between Intel and MLNX drivers... you've any interest in creating/supporting the MLNX side of that? Lemme know 👍