May 17, 20206 yr I had my Unraid in a tower with an AMD FM2+ CPU, but after over a year in that configuration, the CPU started to overheat and shutdown. (I'm pretty sure that the AIO water cooler that was on the CPU failed - I probably wouldn't have been using one, but it was given to me with the CPU). I had recently been given an old HP Server with a Xeon 5345 and a lot more PCI slots (the FM2 only had one). So I decided to move all my drives over to this new box. The Xeon is actually considerably slower, especially with a dual-parity setup, but not horrible. But then, after being online for just over a week, I got a message saying that one of my parity drives had been disabled with write errors (2048). I am hoping that the new machine is not causing the problems... Maybe it's just a coincidence? I should know better than to change something that is already working just fine. Sigh. Anyway, attached it my diagnostic in the ZIP file downloaded from the Tools. I read through some of them, but I'm afraid I really don't know what I'm looking for - I'm ok technically, but the world of UnRaid is kinda new to me... Thank you for any help! Dave w-server-diagnostics-20200517-1650.zip Edited May 17, 20206 yr by EternalFootman99
May 17, 20206 yr Community Expert No SMART for parity, apparently not connected. Check all disk connections, power and SATA, both ends, including power splitters. Then post new diagnostics.
May 17, 20206 yr Author So, I saw that drive was omitted from the SMART reports - So that's a hardware issue, not the fact that it has been disabled? I'll check power and SATA and see what I can see...
May 17, 20206 yr Author I couldn't find anything wrong with the cables, as far as I could see... Here is the drive log as it shows me in the GUI - I'm going to reboot and see what happens....
May 18, 20206 yr Community Expert 3 hours ago, EternalFootman99 said: So, I saw that drive was omitted from the SMART reports - So that's a hardware issue, not the fact that it has been disabled? SMART for all attached disks should be in diagnostics. So it isn't responding. 2 hours ago, EternalFootman99 said: I couldn't find anything wrong with the cables, as far as I could see... The connectors should sit square on the connection with no tension in the cable that might disturb that square connection. And don't bundle SATA cables. 2 hours ago, EternalFootman99 said: I'm going to reboot and see what happens.... Post new diagnostics.
May 18, 20206 yr Author Here are the new diagnostics. I checked all the cables - everything seems to be plugged in. I see smart data for this disk now, so things are looking better? w-server-diagnostics-20200517-2008.zip
May 18, 20206 yr Community Expert SMART for parity looks OK. To rebuild to same disk: Stop array Unassign disk Start array Stop array Reassign disk Start array to begin rebuild
May 18, 20206 yr Author That's good news - but what do you think happened? One thing I noticed is that this old server has SATA 3.0Gb/s, and then the PCI controller I got is running the other drives at 6.0Gb/s. I put the parity drives on the motherboard at 3.0Gb/s, so maybe, with Parity running slower than the data drives there were issues? Do you think there would be issues running my parity drives on a PCI controller vs off the main motherboard? Anyway, I guess my biggest concern is that this is a hardware problem related to moving the array into a new box. I kept everything the same that I could - PSU, cables, etc. Just the Mobo, CPU, and RAM are different. And speaking of RAM, I'm running with only 6GB of RAM, which is way too little. But I wasn't going to buy a bunch of DDR2 RAM, because I was thinking of upgrading the MoBo sometime very soon - this computer is very slow. Dual-parity is really slow with this Xeon CPU - Parity checks were happening at an average or 88Mb/s with the AMD FM2 CPU I had the array on before - now they're happening at 66Mb/s or so. With the cache drive the write performance isn't too bad, but I don't think I really like the overall setup. Thanks for the check-up on the SMART status. I need to learn more about reading through that data and figuring out what's what. I know of the other drives had SMART errors - but that was from way back when I was using a port-multiplier before I knew what I was doing. Thanks again!
Archived
This topic is now archived and is closed to further replies.