Herdie27

So I found my Unraid giving me a warning that the USB is corrupted. (At the time of this writing I haven't seen the warning in a while, I don't know if this means anything but I doubt anything fixed itself because I still have other problems. When I click on the USB on the main page I see a long list of this these errors... Jun 8 09:17:18 Tower kernel: I/O error, dev sdh, sector 36896 op 0x1:(WRITE) flags 0x0 phys_seg 1 prio class 2 Jun 8 09:17:18 Tower kernel: Buffer I/O error on dev sdh1, logical block 34848, lost async page write Jun 9 20:43:18 Tower kernel: FAT-fs (sdh1): error, fat_get_cluster: invalid start cluster (i_pos 0, start 6955717b) Jun 9 20:47:48 Tower kernel: FAT-fs (sdh1): error, fat_get_cluster: invalid start cluster (i_pos 513378305, start 16ce9dfe) I've attached my diagnostics to this post. As per my research I've backed up my USB, apparently it's still readable. Here is my course of action right now based on the research I've gathered. Check the current corrupted USB for with windows CheckDisk and see if it can be repaired, if yes, boot up Unraid and go from there. If no, go to step 2. Make another install USB of unraid and do a custom install from my backed up USB config file. ( I was lucky that the corrupted USB is still readable) I need to figure out how to transfer my license to the new USB. Boot/install? From new USB. Question: Do I even need to do this if I have a cache drive I can run Unraid on? My research tells me I can use my cache drive as the boot drive as well, so do I even need to do anything with a new USB? These are the lots of other issues that I have but I'm guessing that most of them are caused by my boot USB being corrupted. I know the UDMA crc errors are due to power connectors so I'm reseating them after I post this. They've also only gone up once and they haven't got any higher. Custom Name "Warlock" Disappeared and now it's named "Tower" API is offline and can't get it working again 8 Plugin File Install Errors gpustat.plg dynamix.file.manager.plg fix.common.problems.plg folder.view3.plg nvidia-driver.plg tailscale.plg unbalanced.plg dynamix.unraid.net.plg Parity 1: 2 UDMA CRC errors Parity 2: 1 UDMA CRC errors Disk 2: 411 UDMA CRC errors Disk 4: 1 UDMA CRC errors Disk 5: 3 UDMA CRC errors tower-diagnostics-20260609-2348.zip

Okay, 1) I have the other extended SMART report for my data drive attached. I'd appreciate a quick check just for my own sanity. 2) I replaced the two quad sata splitters with y-splitters. Two attached to sata, the other two to molex Assuming I've read enough about it I can enable the parity and the data drive at the same time without issue. 1) Stop the array 2) Un-assign drives 3) Start in maintenance mode 4) Stop the array again 5) Reassign drives to original slots 6) Start the array and unraid will rebuild both drives (I do this little summary for archival purposes) data drive SMART report.zip

So here's the crazy thing I didn't mention, at least for me. The cable that came with the unit splits 3 ways. 1) 3 molex connectors. 2) 3 Sata connections. 3) The same 10 pin that powers 8 Small Form Factor drive bays, I don't use these.

Yeah, I always have to remind myself of this because it works all the time and I don't open it up often. But without getting too wordy about my setup, I have two 15 Pin SATA to 4 SATA Power Splitters https://www.amazon.ca/dp/B012BPLW08?ref_=ppx_hzsearch_conn_dt_b_fed_asin_title_2 to power 7 total drives. One splitter per row of drives in my HP Proliant 350p Gen 8 and the two drivers that have errors on them are right next to each other. I also have mini sas cables for my drives as well. https://www.amazon.ca/dp/B07B9SBSVW?ref_=ppx_hzsearch_conn_dt_b_fed_asin_title_1 I've taken a look at things and nothing is loose or mis-connected, etc. Come to think of it, even though my server is connected to a UPS there's been some strange things going on with my power at my place. I have an Led light connected to the same outlet as my UPS and it the light flickers and the UPS will sometimes kick on and then off within a couple seconds. So there is some where power things going on. In this case, could my problem be solved my rebuilding my data drive?

So all of a sudden I've got read errors on one parity drive and one data drive. In total, I have 2 parity drives and 5 data drives. Being an arm chair sys admin the only thing I know how to do is change out the drives. Luckily they're both under warranty still. And correct me if I'm wrong, I can change out the parity drive and the data at the same time, yes? But before we get to that I've posted my extended SMART self test for my parity drive and diagnostics in this post. parity drive SMART test.zip warlock-diagnostics-20260218-1330.zip

Since I have 2 parity drives my data should be just fine and re-enabling both drives at the same time should be okay? Or should I do one at the time?

So how can I get my Parity disk and Data Disk back up? Everything I've read says the data disk needs to rebuild on itself and I can't find anything on getting my parity disk re-enabled. Additionally, I decided to do another SMART test on each disk and it's been a day and a half and they're still not done testing. They both say 20% complete

No power splitters. I have two power supplies each with what I would say are pretty old power cables hooked up to a UPS. Do you think the power cables themselves could be the problem? The errors aren't listened anymore and the drives are still disabled

warlock-diagnostics-20250821-0102.zip warlock-diagnostics-20250821-0102.zip

It's been a while. I finally got a new cable in. I've replace the cable for the parity drive because it started with the parity drive and then there were errors on my new 12TB data drive, so I'm hoping the parity drive errors had had the errors which then caused errors in my data drive. But anyway, I've replace the cables for my parity drive, now what? I errors are still there when I started the array, how can I clear them and "restart" so to speak.

So recently I had some issues with a data drive that I had some questions about here... But I recently saw there were read errors on one of my parity drive, I have 2 parity drives, about 2300 errors. Since I didn't have time to address it over the weekend, outside of a extended SMART test, attached in this post, I'm here now to try and fix things. All the sources I've read say to simply replace the parity drive. As if there are any errors on the drive, it's in decline and isn't worth keeping. This parity drive is also one of my newest ones and it also passed the extended SMART test. Everything also stopped a read-check at 2.7%, it's now paused. I'm not sure what the best course of action is, other than replacing drives. warlock-diagnostics-20250811-1711.zip parity disk warlock-smart-20250811-1709.zip data drive warlock-smart-20250811-1742.zip

No errors during rebuild. Checked the log as well and didn't see anything wrong. Everything looks to be back to normal.

😬 aaah okay. If any errors I'll let you know, thanks so much for your help!

Yes, after I pressed the FIX button it did it's thing and I decided to check it again. This check came up with no problems so I started a rebuild.

Every time I think I have a handle on things I seem to make a mistake. So I wanted to make sure. I pressed the FIX button and now the drive is rebuilding.