JonathanM

If the error checking and correcting algorithm built into the failing hard drive was working properly (and it should be, the probability of a media error being reported correctly but handled incorrectly is exceedingly small) then when data was sent to the drive, it either wrote properly and returned a success, or it failed the write and returned a fail. When Unraid gets a message from the drive saying a write failed, the drive is immediately disabled, and all further writes will be only committed to the parity emulated drive. Drives are smart though, and when a write to a bad sector fails, it's not immediately tossed, it marks the sector as pending and writes to another sector instead. It's only if the drive is incapable of writing the data that it fails a write. If in the future a sector fails a read, first the drive tries again a BUNCH of times, and if it fails, it sends that to the OS. Unraid records that failure, calculates what should be there from the parity equations, and writes the data back to the drive. If the drive is successful in writing the data, Unraid continues as normal, and the only indication you get is an extreme slowdown and the error counter incremented in the Main GUI. If the write fails, see above. So, the tldr is, your data is fine. The drives themselves provide that protection by reporting back when data is unable to be read or written correctly. Parity provides the safety net when the drive gives up.

If you select GUI mode there will be no CLI login prompt.

IF your signature listed motherboard is still correct, or your new board has the equivalent, you can use your motherboard's IPMI which can be accessed with a browser.

Keep in mind that Unraid doesn't move files from one array slot to another automatically, so the drive that is removed instead of rebuilt will need to have its content copied elsewhere, and when you remove the drive parity will need to be recalculated for just the remaining drives. Ask if any of that doesn't make sense, I don't want you to lose data by making bad assumptions.

click mount

navigate to /mnt/remotes?

Have you manually copied the entire config folder from the original to the new one that is working? All I see is references to USB creator.

Unassigned Devices plugin will allow you to mount shares from one to the other, then you can use mc. I wouldn't bother trying to get Krusader involved, too much complication.

Do you already have a switch that supports LAG? Typical real world experience doesn't show much benefit though, probably not worth the expense and effort.

Make sure Unraid has a firewall between it and the WAN. If the IP Unraid has is a 192.168.X.X you are good, if it's a 47.209.X.X you need to fix that. Unraid isn't hardened against attacks, it needs an external firewall of some sort. Having a private IP with no ports forwarded is the safest, any ports opened need to be evaluated for risk based on the answering service. You can definitely host internet facing applications on Unraid, that's common practice, just don't put the server in a DMZ.

This is not a solution, but a check to possibly confirm what is going on. Unplug all the devices except your laptop's ethernet and the server from the switch and modem. That means all your wifi connections will be dropped as well, so don't do this if you have devices that can't go without internet for the duration of the test. Power cycle your modem. Restart the server and your laptop. If it's what I suspect, when you do that the modem will hand out IP's to the laptop and server, and you can talk to the server with the IP that shows on the console just above the login prompt. It's likely if the modem has multiple ports they may be set up specifically for the WAPs, so make a note of what plugged in where so you can put it back. If it's like I think, then you may need to either get your own router, or get the ISP to program the modem to hand out private IP's to ethernet as well as the WAP's.

Is the modem configured as a router? It may only be handing out a few IP's with valid addresses until it refuses any more assignments. Maybe your WAP system has one or more of those routable IP's and is handing out the private 192.168.86.X addresses to the WAP connected clients. Do your WAP's have multiple ethernet ports? Maybe try plugging an ethernet cable there?

What piece of equipment is on the other end of the ethernet cable you are using?

Why is your router handing out publicly routable IP's inside your LAN?

169 addresses aren't normally assigned, they are a fallback that the device uses as a last ditch effort to talk to anything that will listen. https://www.whatismyip.com/169-254-ip-address/ Can you plug a different PC into the same cable the server is using to see if it gets an address that is valid on your network?

Wait for the update to be built into this container, or move to a different container.

It's plenty. That rating is for continuous use, and spin-up is only a very short period of time. Good quality power supplies handle peak loads even above their continuous rating, so that supply will give you a margin of safety even as it ages.

Sounds like you have a plan! 4 pin splitters and extenders are ok, but not ideal. The fewer connections to get loose or corroded the better. If you can power everything with just the PSU connected cables that is the best option.

NO!!!!! You need to reach out to Adata tech support (or whichever PSU you land on) and get the correct cables. You need 2 more identical to the single molex that comes with the PSU. If that isn't possible or otherwise unlikely, then you will need to educate yourself further on WHY I said what I said earlier, so you can use third party cables safely. It involves checking pinouts, verifying voltages, testing for correct function BEFORE plugging any hard drives into the suspect cables.

Molex / SATA cables should all plug in to the same form connector on the PSU, typically they are just labeled HDD or something like that. Since the PSU you posted comes with 3 total hdd cables, 2 sata and 1 molex, you should be able to order a couple more molex cables and plug them in to the PSU, giving you 3 sets of power feeding your drives. More connections to the PSU = less electrical stress on each connection. You want to divide the load up as evenly as you can across as many wires as you can. Imagine you need to power 6 space heaters. If you plug all 6 into a single power strip plugged into 1 wall outlet, at best you will trip a breaker, worst case you burn the outlet inside the wall. You would want each heater to have its own outlet, with as short a cord as possible. That scenario is an extreme version of what's going on with your drives.

Direct quote from the diagnostics link.

exactly, but the spacing isn't likely to match up perfectly with your case. 3 connectors for 24 drives isn't likely. Probably 6 that need connections. Looks like that specific PSU has 3 connection points for hard drives, so you would want to source 2 more molex style cables for that specific unit and plug each cable into both 12 way bays, so both bays would have power from all three PSU hard drive power connections. Don't even think about trying to run all 24 drives from a single drive power cable.

Sure, I'll just throw one more wrench at you. Look for the number and layout of hard drive cables that come stock, along with the ability to source more for the specific model if needed. Modular power supply cables are NOT always interchangeable, sometimes they are, but the consequences if they are not is catastrophic to any attached hard drives. Make sure the power supply you choose either comes with enough connections out of the box, or has availability of the appropriate cables. I'm assuming (perhaps wrongly, but you didn't say) that for 24 drives you are using some sort of backplane arraignment that should make cabling easier, but it would suck to buy a PSU and find out that you can't connect your drives because of a lack of proper cables.

Since it's single rail, the only number that matters is the 183.33A. 24 drives X ~3A is around 75A worst case on the 12V hard drive feed. 75A x 12V = 900W. So, that PSU would have around 150W left available for motherboard, CPU, video and fans during a worst case power demand event. Realistically, that 900W hard drive number will probably never happen in real world, so it should provide plenty of power even when it's a few years old and has some wear on it. So, using back of the napkin rough math, that PSU is a very good candidate.