benyaki

I was updating a few things yesterday as well as installing a second NVMe cache drive for a cache pool, then noticed that a bunch of things stopped working correctly. Before I added the second drive to the pool (and moved data off to reformat the xfs cache drive back to btrfs) I noticed that I was unable to download 6.9 RC2 through the update OS tool. Didn't think much of it. So I did it manually. Then I rebooted and after about an hour or so of tinkering around, I was unable to access the webgui. Booted to the GUI on the server itself, no problems there, but still no access. Put 6.8.3 back on the flash drive, rebooted, and can access the webGUI from a laptop, BUT NOT a phone (iphone, tried a few). Tried to mount the flash drive to check a few things - it mounts via SMB, however only two files show: ldlinux.c32 and ldlinux.sys. Absolutely nothing else. A few plugins were not working correctly, so I uninstalled some, however when I go back to reinstall them, the download hangs (usually at 45%). I am on a gig connection, so it appears to be something server side as I can download from the link directly on any other computer just fine. On that note: I cannot download community applications again, nor can I run 'fix common problems' since it needs to download a few things to work. I am currently in the process of having mover put back appdata to cache, so I cant shutdown anything right now, but wondering if someone would be able to take a look. The ONLY thing that I can think of is perhaps I am using too many PCI lanes and that is screwing with the onboard NIC? It really seems to have just started with the addition of the second NVMe drive. Here is a sample of what trying to reinstall a plugin looks like: tower-diagnostics-20210201-2044.zip

So I let it run (moving some data around) and came home, could not access any dockers. So I stopped and restarted the array. Cache drive is missing (not a visible device). Restarted - same thing. Looks like the drive has disappeared ... Attached diagnostics tower-diagnostics-20210121-2011.zip

Isn’t the drive limit 30?

Any suggestions to fix or troubleshoot? Or is it just something that occurs randomly?

Just noticed that log was at 100% so I had a quick look through it and the last hour was riddled with BTRFS errors. I currently have a mix of xfs drives and reiserfs drives (eliminating the later right now, just moving things around). Any insight as to why these errors are popping up? Of note - unbalance is currently running, moving data off reiserfs drives onto xfs drives before they are either pulled or reformatted. The NVME cache has been in use for a few weeks. Attached diagnostics. tower-diagnostics-20210121-1140.zip

Thanks for the help and looking into this BTW. much appreciated!

Yes, I made sure to grab a screen capture of the drives, their slots etc while it was working. dual parity, I’ll probably cut that down to single as I am moving down to 4 data drives at 12tb each. Or 3 12tb drives, a few 4tb drives, and dual parity.

Booted it back up, everything was assigned properly, there were 2 unassigned devices (1 which I have in there for no reason, and the second which was the cache drive that was just removed). The SATA drives are all attached the PCI-X controller cards (Supermicro something-something, cant remember). Nothing on the motherboard. Anyways, because I was bored, I setup the bare minimum for the new hardware, booted it up (obviously without any drives attached) and it seems to have no problem with the plugins that were installed, whereas with the other (old) motherboard, it still has the problem. I am going to put it down to the new BIOS, and hope for the best on the new hardware?

It could be that I just ran diagnostics really quickly after it booted and they have not been assigned yet - noticed that on the GUI that they take a few seconds to show up. Just to add - I just tried removing CA, added in a couple other plugins (unbalance and unassigned devies) - still prevents me from accessing the WebUI or terminal. One thing I did notice earlier was that before when all the plugins were still installed - either in safe mode or normal, I could not ping any address. I just checked now on command line on the machine and can ping an IP address (again, unbalance and unassigned devices are installed, but still no WebUI or telnet access)

So I have an ageing server that I have new hardware here to replace (almost all parts, waiting on some HBAs so I can connect all the drives to transfer to new larger ones). Everything was running fine, then I had a problem with the bios being corrupted, replaced the BIOS chip. That finally arrived in the mail. Booted up the system last night, everything was OK, started going through and cleaning up some items (older dockers not in use and changed from 2 cache pool down to 1 cache drive. Tried rebooting - no WebUI. Can boot into safe mode, but still no webui access or telnet. Disable autostart. Removed ALL plugins, can access WebUI again. Great! Except - the first plugin that I add back in - Community Applications, installs properly, looks good, so I reboot to make sure everything works - again, no telnet or WebUI! Shutdown and remove the plugin, can access again. I have attached diagnostics from when a few plugins were installed but it is the same result. Anyone have any ideas? Would like to get this all running and before I switch over to new hardware. Thanks! tower-diagnostics-20201230-1531.zip

Thank you for your input, same way I was going. Also minimizes potential errors. Just wanted to be sure it would work. thanks!

After upgrading some hardware, I am reducing the number of drives on my system as I have setup other backup solutions and no longer need some redundency. CURRENTLY, I have a cache pool of 2 HDDs, 500GB each. What I would like to move to is a single cache drive, a 500GB NVME M.2 drive. There are a few ways to go about this, and I am wondering what is the best and safest way to do so. Worth noting that the cache is backed up every day. Here are the options that I see. 1. Remove one drive from the drive pool (as per the following) and then copy all the drive contents over to the new drive, stop the array and replace the old drive with the new one from the cache selection 2. Replace one the 500GB drives with the M.2 NVME drive, let it rebuild, then remove the other 500GB HDD as per the above instructions 3. Copy all the data from cache over to the new drive (as an unassigned drive). Stop the array and unassign both cache drives. Then assign the NVME to cache as a single drive. I have a feeling that 2 is possible and the fastest way to do it without having to worry about moving data around. Greatly appreciate any input/help. Thanks!

Hello! Just putting this in a new post as it looks like the old will get lost and I want to order parts ASAP. I am upgrading a 24 bay server and since my knowledge of hardware is about 8-10 years out of date, I need a little hand-holding here. BIOS chip on old one died, waiting on new chip to limp it along until I can transfer to new hardware. I do not have a set budget, but don't want to be wasteful. Wife said build it without cutting corners. Can someone please have a look over this to make sure everything plays well together and that I am not missing anything that might be important to know. It will be running Plex, the regular media stuff, a VM (MacOS), no gaming, about 8-10 dockers, and a medium around of transcoding (<5 stream max). I will be re-using the modified 24 bay case that has a SATA backplane, the Corsair TX650 PSU. Everything else is new (so not really an upgrade, a new build in an old case) Here is a list of the parts that I will be ordering (plus some 80mm and 120mm fans, haven to decided on any specific ones unless someone has recommendations. Noctua?) https://newegg.io/7fda6c6 Here are the components: - Intel i7-9000 (a little overkill, but would rather not upgrade again for awhile). Decided against the 9000k for energy use. - ASUS Prime Z390-A - any problems with this or other suggestions. - 32GB (2x16GB) CORSAIR Vengeance CMK32GX4M2B3200C16 (not sure if this is adequate, cheap, or?) - Western Digital WD 500GB nvme drive for cache (WDS500G3X0C) - Coolmaster 212 Black edition CPU cooler - LSI 9211 card (IT mode) with SFF-8087 to SATA cables (x2) I will use onboard SATA for parity Cache will be M.2 Data drives will sit on the LSI card. This leave 2 extra PCI slots for GPU and extra LSI card or expander if needed Ready to order (not necessarily all from newegg, there are some other places here that are quicker (same day) or cheaper.

Does anyone have any recommendations? Hoping to take advantage of some sales if possible. Thanks a bunch!

I dont mean to hijack this thread, but if anyone would mind giving me some help picking out some new hardware, that would be great. I am in the same boat, running basically the basic stuff, just older (though less dockers and less streams for plex) and I have been out of the hardware part of this for far too long to even know what to look for. I currently have a 4u supermicro server, ECC ram, 24 bay, 2 cache drives (HDD) and dual parity. Looking to upgrade the MOBO/CPU/RAM, controller card, and probably add a GPU in the near future. Would be looking at switching cache over to a dual 512GB M.2 pool. I could gut an old dell server from ebay, but wouldnt know what is going to be sufficient, and dont know what hardware would limit other (ie 'this motherboard doesnt support GPU passthrough' or something stupid). Very sorry for the ignorance.

I currently have a 24 bay SATA backplane server with a regular PSU and old internal hardware (10+ years) which I would like to upgrade. Expectations of the server: - Plex (including transcoding up to 3 streams at a time) - Virtualization (OSX) - photo, file, and music storage, backup - basic dockers for media handling, VPN, pfsens - NOT sure if I need a discreet GPU? wont be doing any gaming, only help would be with Plex transcoding? - Upgrade the HDDs, but have time to do this and can add drives at this time if needed. I am currently using 33TB and need room for more. The last time that I put any hardware together was well over 6+ years ago, so I am definitely 'out of it' with that is available. I am looking for recommendations for a motherboard, CPU, RAM and controller cards that would be sufficient. I would like some decent speed out of the server and realize that the old multiple smaller drives are a bit of a bottleneck at this time, so will be slowing moving to 8TB or 12TB drives. I would be happy to move the two cache drives that I have over to a set of M.2 drives on the motherboard to reduce a bottleneck (OR would 2 internal 2.5" SSDs be just as good - I dont see much in price difference. Budget right now is not really an issue, but I don't want to spend money on hardware that is going to go to waste (ie a GPU if I don't need one, a ridiculous motherboard if I dont need all the features etc). I am looking for parts mostly in Canada since I cannot get anything picked up in the US at this time with the border closed. Thanks for the help!

Please delete or close this thread, I have changed plans (see last line) and have started a new thread in motherboard/cpu section) I am looking for some new server recommendations (hardware). I looked into this last year, but never ended up getting around to it. I currently have a 24 bay case whose internals are going on 9+ years old. Uses a regular PSU, has SATA backplate on the drives. I am currently using 33TB and need room for more. There is obviously some inefficiencies in the hard drive capacities with a few of them being 1.5TB and 2TB drives (dual 4TB parity and dual 500GB cache). Open to upgrading a bunch of drives to 8TB to lower drive numbers. I currently use or plan to use the server for: - Plex (including transcoding up to 3 streams at a time) - Virtualization (OSX) - photo, file, and music storage - basic dockers for media handling, VPN etc - I would like some expandability for the future in terms of storage Currently this server is inefficient, slow for transcoding etc and uses around 160W at idle. EDIT: I have decided to upgrade the internals of this case (new MB, CPU, RAM, Controller card) - new post in motherboard/cpu section as I have no idea what is required.

Thanks Mmr8cmh - I might have to look into something like that and getting rid of the older smaller drives. At least with 12 bays I still have room for 8 data drives with dual parity and cache.

Just wondering if anyone had some ideas or suggestions for this. I would like to get a new server or build going soon.

Hello! Just looking for a little bit of guidance on selecting a new server. The one that I currently have is a 24 bay supermicro build from YEARS ago (10?) that was through tamssoltions if anyone remembers that. I was looking to buy something that is already built, either a 2u or 4u (though likely 4u, see further on). I use it primarily for plex, but will add on some other light docker containers for home automation, cloud storage etc. The problem right now is that I am still halfway through updating my HDDs and have been sing 4TB drives for awhile. I currently have a wide mix of drives (16 drives including dual parity and a cache pool of 2 drives) with about 36TB of sable storage. YES, I could move things over to 16TB drives or even 8TB drives, but I would rather not put that money into drives right now. I was looking at something along the lines of a R710 server with a fair amount of ram, dual CPU etc for a relatively low price on ebay, but obviously there are not enough drive bays. Just wondering if anyone has some suggestions on any prebuilt servers with more bays that is of decent performance. Honestly the last time that I looked at hardware was years ago, so I am very far behind on what is good, too much for what I need, or not enough. Like I said, mostly plex, some other light docker containers, and probably a fair amount of transcoding for remote or local steams. If anyone wishes to provide some guidance and needs additional information, please just ask! Thanks!

Thanks again for the help

One more question, just want to make sure I dont mess something up when transferring files back - should I temporarily disable user shares while I run copy command: cp -av /mnt/disks/UAdisk /mnt/diskX

Ok, thank you for your help. Tip to future self - don't make server changes after a 20 hour day and low blood sugar 🙄

Yes, sorry I see I used the wrong terminology. Should I let it finish the (incorrect from my point of view, correct from the servers) rebuild then move the files after?

Because when I replaced it, I stupidly stopped it during the rebuild to change the format to XFS. Now the array assumes the drive should be blank.