dev_guy

I'm having this issue as well and it shows unprotected on the Shares tab with the warning triangles. I have two mirrored "Raid 1" btrfs cache pools set up (one with a pair of SSDs and one with a pair of conventional drives) and Unraid considers some of the shares using them to be unprotected. The "System" share using the same cache is happy. Why? Please see the attached screen shots.

@JorgeB thanks for the answer and suggestions! I wasn't aware SMB is significantly better in 6.11. The disk share idea is also interesting. I believe the file I/O in Unraid shares is perhaps going through Fuse? If so, that would help explain the abysmal performance reading large numbers of small files. I know NFS offers much better performance with small files as well. So perhaps NFS plus a disk share might approach the speed of the drive itself.

But my Unraid machines also handle multiple VMs and dockers and they idle at 15 watts and 23 watts. Your server is using around 10 times more power doing nothing. It's not the fault of Unraid it's using the wrong hardware. It's like buying a V12 Lamborghini and complaining it uses way more fuel than a Toyota Prius.

Why does it often take 3 separate steps, having to apply each along the way to reveal the next step, to apply share permissions? With any other NAS OS I know of you can do it all in one step and only click "apply" once. Why not just put all the options initially on the same screen instead of slowly revealing them like some kind of shell game?

Two unraid licenses but only one server runs 7x24 currently. I have other servers running 7x24 including OpenMediaVault and commercial NAS units. There are just too many performance and other issues with Unraid for it to be my sole solution.

Well it's still part of the issue. I have a gaming rig with an RTX3060. It idles at around 60 watts and about 15% of that is the graphics card. But you're basically trying to use a gaming rig as a 24x7 Unraid server. That's just not going to be efficient no matter what you do.

In my experience AMD desktop/gaming/server CPUs are much worse than Intel for real world power consumption especially at idle. Adding two power hungry graphics cards just makes your problem worse. I don't think there's an easy fix besides moving to more power efficient hardware. Depending on electricity rates in your area you could spend more on power in a year or two than it would cost to switch to much more power efficient hardware. I have two Unraid systems. One idles at 15 watts with the drives sleeping and the other idles at 23 watts with the drives sleeping. That's power consumption as measured from the wall with a Kill-A-Watt meter. Both are recent generation Intel with no graphics cards. Neither of them has been CPU limited even with many dockers and a few VMs running. So, depending on your use cases 300 watts, is just a huge waste of energy. If you want to game then use a gaming rig that's only powered up when you're actually gaming. That's much more power efficient than using a power hungry gaming grade rig powered 7x24 for an Unraid server.

Thanks for your work on this and posting the PDF.

This is a bit odd but I wanted to see if I'm going to break anything (namely parity) if I try this? Can I shutdown my server, remove a XFS data drive from the array, mount it read only on a Linux system to work with the data on the drive, and then return it to the array? If it's mounted as read only while out of the array will everything be okay? Or am I missing something? The reason for doing this is strictly performance. It's 12TB of data, with half a million files, and accessing it over the LAN via Unraid is proving to be painfully slow and easily an order of magnitude slower than having it be a local SATA drive. For whatever reason Unraid is rather slow for a large number of small reads over the LAN. So will my array never miss the drive being borrowed for a while as read only? Or will some bit get flipped somewhere throwing off the block parity calculations?

Thanks! I'll give that a try although it's happening on multiple PCs/laptops. It's possible none of them have been re-booted, had cookies cleared, etc. Just as feedback for the Unraid team, that's a poor way to handle things. If a user selects "never check" for updates you shouldn't nag them under any circumstances. Or, worst case, only nag them once and never again after they dismiss the first update notice. It's something Unraid should fix especially given it's a paid product.

I have "Unraid OS Update Notification" set to "Never Check" yet I still get frequent annoying banners, pretty much every time I log on, telling me 6.11.1 is available. I'm running 6.10.3 and have no desire to update given the issues many users have reported with 6.11. Is there a way to stop these annoying upgrade banners? Why does Unraid not honor the "Never Check" option for those of us wanting to stay on our current version? This is more what I'd expect from Microsoft or Google trying to force updates not Unraid. I've searched the forums and have not found a solution. Any help would be appreciated if I'm missing a setting somewhere?

Thanks trurl. I love how responsive people are on this forum.

Thanks, but yeah, I'm starting with new disks and a completely different configuration. I've already moved most of my data to a NAS. I still don't know what happens if I use the USB Creator to install a new version onto the flash drive? Will it automatically license based on the serial number of the drive on a new system? Do I need to copy back the old .key file?

I'm wanting to do something similar. I want to retain my license and use the USB flash drive on entirely new hardware. It sounds like I can just delete everything except the .key file and pop the USB drive into the new server if I want to use the same version? What happens if I use the Unraid USB Creator on a licensed flash drive (which I assume will destroy the .key file) to start off with a newer version? Thanks!

Thank you Trurl! The system I was doing the preclear on DID have bad RAM! It has 32GB or RAM so it took a while for it to show up in the memory test. So thanks for that tip! And, likewise, the same drives precleared successfully on different hardware. The drive that had logged a SMART error for a pending sector needed to be precleared with pre-read disabled so it would not abort before it ever got to the zero phase. During the zero phase the pending sector was recovered. Thanks also to S80_UK and Gfjardim for their input. Indirectly preclear helped me find bad RAM rather than bad drives. A question to gfjardim or anyone else: Is the pre-read designed to cause a preclear to abort early if a drive has problems? I suspect this is documented elsewhere in this 109 page thread, but it seems to fail drives with "soft" errors unless you disable the pre-read and then the pending sector is recovered during the zero phase? Someone using the default settings might assume an expensive hard drive is defective when, in reality, it's not.

Thank you for the reply and also the suggestions from S80_UK and Trurl. I ran HTOP with columns for I/O enabled and noticed DD and CMP were being used. And I agree those by themselves should not be an issue. I am currently trying to preclear the drives again on different hardware. I ran HTOP because Unraid was reporting 100% CPU doing the preclear on 7 drives at the same time. But HTOP was only reporting around 50% total CPU usage so I'm assuming Unraid somehow has that wrong in the dashboard (at least for my configuration). But I would assume, even if the preclear process is limited by 100% CPU, it still should not cause false failures. I'll report what happens on the new hardware when the new batch of preclears complete. Thanks again!

I agree memory problems can be elusive. I've been running Memtest for the last 6 hours with no errors so far. I'll let it run overnight. I think a more likely explanation is there's some weird incompatibility with the current version of Preclear (or the script) that's not compatible with these drives at least on that hardware. They're 2.5" drives which I suspect not many Unraid users use. The target hardware is a 1U shallow rack Xeon Supermicro server with ECC RAM but I'm doing the preclear offline on a different system where all the supposed "failures" occurred.

I have not. That wouldn't explain why other drives have precleared on the same hardware but it's a good suggestion. This is a different motherboard and RAM than I had the earlier array running on. You would think, however, if I had bad RAM it would eventually crash Unraid and the entire machine. Still, it's worth a try.

I please need some help. I have a mix of brand new Seagate ST2000 drives and identical drives that were in a previous trial array where they worked fine. Now they're all failing preclear at the post read. All but one of them (S/N ending in 6381) have no SMART errors and I can't see anything specific in the log files that indicates an actual failure. The Post-Read process is simply terminating with "your drive is not zeroed." To make things more confusing other drives, but not this particular ST2000 model, have precleared just fine on the exact same hardware and Unraid USB configuration. I'm using the classic supported LSI/Avago 9211-8i controller running IT firmware. This is running the latest version of preclear with the default script on Unraid 6.8.2. I'm wondering if everything worked before because it was an older version of both? I've attached the report and log for the serial number ending in 1033. It shows the SMART status is fine. The reports/logs are identical for all the other "failing" drives (except S/N 6381 which may have a legit problem). I've also attached the Unraid diagnostics ZIP and included a few screenshots. Does anyone have any ideas or suggestions? Any help would be greatly appreciated! preclear log failure sn1033.txt preclear report failure sn1033.txt tower-diagnostics-20200305-0916.zip

Yes but it defeats the purpose of the new easier-to-use tool, and may well cause some to give up on unRaid, if the recommended download and tool fails. I've been using unRaid since before the tool was released and know about the old way. It also doesn't help that unRaid is significantly different than any other NAS OS I know of. The others generally use an image file and a tool like Rufus, Win32DiskImager, etc. to create a bootable USB drive.

I'm now having this issue as well on a Windows 7 machine with a Sandisk USB 2.0 FIT and also with a Silicon Power USB 2.0 drive. It just hangs at "Syncing File System" after showing about 98% complete. This is obviously a real problem and hopefully something that will get addressed. I know the USB creator is relatively new so a few bugs are likely to be expected. Sadly it also forces you to re-download the OS image each time you try something different. I've easily used up over a gigabyte of bandwidth trying different drives and ports. It would be nice if the tool had the option to use a local copy of the OS image and/or it saved the downloaded version.