Christopher (Drashna)

Piotr, Thanks for adding the files. Alex should get to the file soon, and will let you know. Worst case here, if you want to do reset DrivePool's settings: http://wiki.covecube.com/StableBit_DrivePool_Q2299585B Or do a clean install of DrivePool: http://wiki.covecube.com/StableBit_DrivePool_Q3017479 These may help resolve the issue as well.

If it's getting unstable with StableBit Scanner, could you see about downloading the latest internal beta? http://wiki.covecube.com/Downloads

Definitely do this, It's been forever, but this setting was a HUGE issue in WHSv1. Make sure it's disabled, reboot and see if it helps.

To make sure, the disk is not a "dynamic disk", right? Also, could you update to the newest build of DrivePool and see if that helps? http://dl.covecube.com/DrivePoolWindows/beta/download/

Sorry for the delay in reply. I've been out of town this weekend (well, just got back in last night) As for your issue, that's very unusual, and definitely shouldn't be happening. Thank you for the service dump. I've flagged it for Alex and he'll take a look at it. In the meanwhile, could you get the error reports from the system as well? http://wiki.covecube.com/StableBit_DrivePool_2.x_Error_Reports As well as get the logs from "C:\ProgramData\Microsoft\Windows Server\Logs". Specifically the "Dashboard" logs? Additionally, have you tried the latest internal beta build? It may work better. http://dl.covecube.com/DrivePoolWindows/beta/download/ IAR/bug: https://stablebit.com/Admin/IssueAnalysis/2159 (will be empty til Alex updates it.

Unfortunately, not at the moment. But: http://community.covecube.com/index.php?/topic/208-stablebit-drivepool-controlling-folder-placement/ However, if the controller card is just passing the disk through and not using RAID, then you should be able to just replace the card and regain access to the data.

If it's corrupting files on that ONE drive, swap out the cable, and check for loose connections. Also, is this drive on a different controller? or in a dock (if so, reseat the disk)? But if it is consistantly corrupting data... you definitely have an issue.

I'm glad to hear it!

Just to confirm. Yes, exactly like that. At worst, you may want to reset the NTFS permisssions, but that's it. Well, except for "hoping". It will see the "old" pool, and recreate it.

If you're the user that just posted a ticket about this as well... I've answered there already. If not: You will need to open the disk in Scanner, and "update" the definition for the drive. That, or restart the system. Once that's done, it should stop crashing the enclosure. If the issue persists: Can you please download the latest version of Direct I/O test (our internal testing tool) and run it on your server. Select a disk that's located in that enclosure from the drop down and check the "USB" box. Take screenshot of the output and post it here. Download: http://dl.covecube.com/DirectIoTest That'll tell me which USB chip that enclosure uses and the pass-through methods that it supports.

Lurifax, Thank you for uploading that. As for the issue.... I really can't tell as I don't have access to any of the diagnostic tools.... so Alex will have to look into this. (I usually do a brief check to make sure it's not something else, as to help prevent Alex from getting overloaded, but I'm not at home, so....) For reference: https://stablebit.com/Admin/IssueAnalysis/2157

First, BSODs are never a good thing. Ever. So, could you upload the dump from the system to us, so we can take a look at it? http://wiki.covecube.com/StableBit_DrivePool_System_Crashes Also, during the BSOD, it will sometimes indicate the file that caused the BSOD. If it's CoveFS.sys, then that's definitely us. And we will definitely want to get that fixed. Also, what version of DrivePool are you using?

It's on our to-do list: https://stablebit.com/Admin/IssueAnalysis/1133 However, there is a lot on that. Alex has been busy away hammering out issue, and we hope to get to it soon. Though we have a pretty good list of issues to fix and features (such as this) to add. We want to get to all of them, but there is only so many hours in the day. But if you want to see a list of what we're working on: http://wiki.covecube.com/Development_Status But I'll bring it up to Alex, as Quota's are a pretty big part of NTFS.

Ah, okay. And yeah, I agree. Definitely not as easy as it should have been. As for the dashboard, it's connecting to Azure's background service. Which you access by using 'wbadmin.msc' on the server as well. But I have checked and it definitely errors out if it has more than 800GBs to backup. However, the integration and ease of use is nice. Just needs to be able to backup more from a volume (otherwise .... lots of 800GB partitions just to use it....)

Glad to hear it!

If it's identical hardware or if the hardware changed slightly, then this is expected. However, I recommend deactivating just in case. That way, you don't have to contact us and wait for our response to be able to activate your license (even if you need to, you should be able to activate the trial period while waiting).

Also, Azure seems to be limited to 834GB per volume (or something very close to this). That also kind of makes it impractical to use on DrivePool itself.

Yes, it will recognize the drives and "just go". Specifically, the duplication information is stored "locally", on the pooled drives, so this data is ALWAYS maintained. Even between systems. Everything else will be lost though (balancing settings, Network IO Boost, etc) But for the most part, you just need to deactivate your licenses, reinstall Windows, reinstall our apps, and reactivate. And that's it.

Wow... setting this up isn't exactly straight forward (or i went about it totally wrong). And definitely not cheap, either. Also, just FYI, if you load up the Windows Sever Backup console (wbadmin.msc, or ... the long name in Administrative Tools), it specifically mentions that it fails because it failed to take a snapshot of the volume. That means that it definitely depends on the VSS service, and therefore is not supported on the pool. Sorry. However, like the Server Backup feature, you CAN backup the drives that make up the pool. (or specifically the "PoolPart" folders on each drive). I know this isn't the best solution, but it WILL work. Just use the console that I mentioned above to configure it.

Have you ran the "duplicate now" option after removing the disk that the files/folder that these files were on? Also, if these were duplicated, have you checked the duplicated copies to verify that they are intact as well? Either way, hit the "Duplicate Now" button. If it comes back, please enable tracing, restart the server and try it again. http://wiki.covecube.com/StableBit_DrivePool_2.x_Log_Collection

You are very welcome. I can't comment for the others, but I enjoy sharing my experiences and knowledge with others. Oh, and it looks like the Intel Xeon E3-1245v3 is probably the best CPU to purchase. It's a bit cheaper and faster than the 1225, and runs at a lower max TDP than the 1225 or the 1275 (the next higher CPU, higher clock speed and higher price). So it's basically the best balance of everything. And sold out on newegg... probably for a good reason. Newegg comparison: link Amazon link for it (~$290, instead of the 1225's ~$315) http://www.amazon.com/dp/B00D697QGI/?tag=extension-kb-20

Also, looking closely at the CPU... I'd recommend a Xeon E3 12x5 v3 CPU instead. The reason being, that these ones have onboard (on chip?) graphics. The board you linked doesn't have IPMI, so you'd need some way to hook up a monitor ... or do it 100% blind. Either way, it would make things easier. Such as the 1225: http://www.newegg.com/Product/Product.aspx?Item=N82E16819116910 Or the 1245: http://www.newegg.com/Product/Product.aspx?Item=N82E16819116909 (Amazon has it in stock .... and cheaper than the 1225) But this is definitely your choice. I count data integrity as part of stability, sorry. And either way, I think it's worth spending the money on it (and I'm a cheapskate)

Also, that case that you have (the Rosewell 4U), you'd want to get normal breakout cables: http://www.newegg.com/Product/Product.aspx?Item=N82E16816116099

For the LSI card, yes most likely. I got a IBM ServRAID m1115 (one that you need to flash), when it was operating in RAID mode, I had to enable UnsafeDirectIo. Once I flashed it to IT mode (HBA mode basically) with NO management firmware, I didn't need to. So, I'm not 100% sure there. But the HighPoint card? Yes, you will. As for the link on how to enable this option: http://wiki.covecube.com/StableBit_Scanner_Advanced_Settings As for the Pros and Cons: The "UnsafeDirectIo" option tells Scanner to use command methods that are in our "unsafe" list. This means that these command can cause issues. It can cause the system to become unstable, the controller card to reboot drives (offline -> online), crash the controller, or even cause BSODs. We maintain the list, because stability is the #1 most important thing to us. Period. End of story. If this means that we don't always display SMART data unless you turn on a setting, well, we feel that's more important. Not to take a pot shot at another product, but I've actually seen a few people have issues caused by HD Sentinel because it was using one of these methods that we've marked unsafe. That said, most higher end RAID cards (HighPoint, 3ware, LSI, Dell PERC, etc) tend to require the unsafe methods, but don't seem to suffer any ill affects. I've used a HighPoint (lower end of the higher end cards) RocketRAID 2720SGL for more than a year. I've never experienced issues with it (well, one, but I don't count it because it was.... a tcpip.sys BSOD caused by the web management app for the HighPoint card, because of a drive that was catastrophically failing, and causing all sorts of issues, even crashed by USB enclosure) And yeah, I run a bunch of roles and services on my server, as well as share it with friends and family, so stability is important too. And I'm tired of stability issues (ASRock 990FX Extreme3, AMD CPU... only because it was cheap from a friend). And yes, ECC would be worth it if stability is important. However, I have two "8i" cards, so I only need 5 ports on the board (system and one "rail"). 10 is too many for me.. .but better too many than not enough. And you are very welcome for the links. Though make sure you get REVERSE breakout cables to convert the SATA ports on the motherboard to a SAS port. it's not reversible. It is one way or the other. So it's important to get the right ones.

That... build looks suspiciously familiar. Like almost exactly what I'm going to be upgrading to soon .... As for compatibility with StableBit Scanner and SMART... the onboard chipsets usually aren't an issue, but I'm not sure. Only one way to be certain. Worst case, you man need to enable "UnsafeDirectIo" to get SMART data (which you most likely will need to do for the SAS card anyways). As for the Norco... I have a RCP 4220 (the 20 bay version of what you're talking about). If the SAS card you have uses mini-SFF 8087 connectors, then you need to get a couple of cables and connect it directly to the drive "rails". Simple, and no need for cable management. Each one supports four lanes (aka drives). http://www.newegg.com/Product/Product.aspx?Item=N82E16816116093 Also, you may want to get a couple of reverse breakout cables (4 SATA to a mini-SFF 8087) to plug the onboard into said "rails" as well (is what I do). http://www.newegg.com/Product/Product.aspx?Item=N82E16816133033 And yeah the LSI cards are very very good buys. That, or get a HighPoint Rocket 2720SGL card or two (ROCKET, not ROCKETRAID) http://www.newegg.com/Product/Product.aspx?Item=N82E16816115111 Additionally, you don't absolutely need to get ECC RAM. if you have spare "desktop" sticks, you can use that. However, ECC has built in "Error Checking and Correction" built into it. So it would make the system more stable.