Christopher (Drashna)

Christopher,
Alex found that software (not sure what) was overwriting the GPT.  Alex ultimately made the physical disk read only so that it will not be overwritten.  Currently, the pool is functioning as designed.  Therefore, the issue appears to be resolved.
 
 
Thanks,
Todd

I ordered the motherboard and CPU yesterday. I decided to bump up the CPU to a Xeon E5-2620V3 2.4ghz processor $430.00. It has hyper threading and a few other goodies that the 1.9ghz Xeon does not have. Stupid CPU costs more than the motherboard! Newegg was out of memory, so I'll have to get it elsewhere. While I'm at it I might as well get a SATA III SSD because I have a 256gig SATA II SSD for the OS drive now. What I paid for my current SSD I can get one with double the capacity and faster for a little less money then what I originally paid!
 
Sigh... :-)

Honestly, it depends on how you want to do it.
 
If you have the space available, you can remove the drive you want to replace first. 
 
But you could absolutely add the new drive, and then remove the smaller drive from the pool afterwards.
 
 
 
However, what I recommend here:
Remove the drive from the pool. Use the "Duplicate files later" option. This will remove the drive quickly, as it's leaving the duplicates on the drive. After the disk is removed, it will recheck the duplication status of the pool, and reduplicate files as needed.
Before it starts doing that, physically remove the "old" drive, and connect the larger drive.
This way, when it starts duplicating the data, it will use the new drive most likely. This will minimize the reshuffling of the files on the pool.

Well, PCI-X is still pretty darn fast, so either way, you should be fine!
 
And to make sure, I recommend the IR (RAID) firmware over the IT (passthrough/HBA) firmware. Both can do passthrough, but the IR seems to behave better. Besides, extra options are great.
 
 
As for Intel SAS Expander card, great catch! And yeah, I've only been able to find them for $200 or so. So finding one for $60 is fantastic! (if you see another, link me!)
 
 
As for the BSOD, it indicates a CLOCK_WATCHDOG_TIMEOUT bug check. This indicates that an expected clock interrupt on a secondary processor, in a multi-processor system, was not received within the allocated interval.
https://msdn.microsoft.com/en-us/library/ff557211(VS.85).aspx
As for the hotfix, it may be OS specific. And hotfixes aren't usually pushed through Windows Update. They're usually option installs offered to fix specific issues.
This is the hotfix you're referring to:
https://support2.microsoft.com/default.aspx?scid=kb;EN-US;975530
And if you check lower down on the page, there are a couple of registry tweaks that you can try, and see if it works.
 
 
 
If it happens again, set the dump to us:
http://wiki.covecube.com/StableBit_DrivePool_System_Crashes

Well, if it is an educational excersize only then I would opt for the WS10 Tech release. I have that installed on an old machine for that purpose as preperation for the migration from WHS2011 sometime in the early 2020's

Yes it is for WHS 2011. Thanks for the link. I'll save that for future refernce. Everything is back up and running now. I was able to recreate the shares through the dash board. All the shares are where they where before WHS freaked out. I also ordered some AOC-SAT2-MV8 cards for spares. I found a great deal on them on EBay. $35 each for two new cards.
 
Well I'm happy that things are back to normal. I need to pick up some more 6TB drives over the next couple of weeks. Gotta replace one 2TB drive that has a some bad blocks and a couple other 2TB drives that Scanner says I should consider replacing. Funny thing is I've had the most problems with WD 2TB drives. The WD 1TB drives I have installed in the pool I've never had a problem with and they are older drives.
 
At least I have access to my music and movies and performance is snappy.

Yep, it really is magical.

I believe my wife thinks we are running a data center sometimes. I've made a lot of changes recently and everything is a mess, but I have it on my list of things to do to clean up the closet this weekend. When I do, I'll get some pictures and put them up. 

pclausen, would you be willing to detail out your hardware?  I'm just curious about how you have that many drives connected, cards/connectivity type used, if you'd do things differently a second time around, etc.
 
Thanks in advance
 

Which would not help you against a number of things. I would recommend offsite backups for important data (all data actually to the extent possible).

For what it's worth, I run BI on the same "server" as the rest of my stuff. But, what I've done is set up ESXi, and BI is on a separate VM, and each of my VMs have a dedicated NIC. I've only given it two cores and 4GB of RAM, and works just fine at about 50% utilization under full load. Also, it stores to a separate pair of drives than the rest of my machines. This way, no matter what BI does, it never affects the rest of my system. This will keep you from needing to have a separate system just for BI. I'm sure it work just fine under HyperV if you wanted to go that way.
 
My BI implementation utilizes 7 3MP cameras and two 1.2MP cameras. The 7 3MP are hardwired using PoE, the two 1.2's are wifi, all configured for direct to disk. All cameras come into a different switch on a different IP range, and as I said a separate dedicated NIC. The other VMs I have truly have no clue that the BI machine is even there. The same server is running WSE2012, so it backs up my workstations on the local LAN NIC, and it is also my IIS server for a few websites.
 
I just checked power consumption on my server - 77 watts. 6 hard drives, Xeon E3-1246v3, Intel server motherboard, etc. I am using a Seasonic gold plus power supply, that may help a little bit.

Another advantage of Drivepool is that if you have a specific folder that you want extra protection for, like pictures, you can set it to duplicate across more than just two drives. So you could have three copies if you wanted.

Using 'dir /o:n' resolves the issue in cmd.exe and doesn't rely on any default behaviour (which has changed between releases of Windows). Of course it means changing your batch files but that is a one time hit.

I'm going to jump the gun and post my conclusion here: my issues were most likely caused by my add-in controller, even after installing its driver. I haven't yet run the burst tests, but here is a list of events that point to it:
When I first built the pool and relocated some pooled drives onto the new controller, those exact drives began to exhibit file access issues. I didn't mention this when I originally posted because I didn't notice issues immediately after doing so. Removing those same drives and reading them with my USB HDD dock exhibited no access issues (nor corruption). When formatting a DIFFERENT set of drives on the suspect controller, the formats would freeze. This was attempted three times. After flashing the firmware, the formats appear to be progressing normally and are further along. (To be clear, updating the DRIVER seemed to make no difference.) Currently, my pool is rebuilt from drives using the onboard controller only. There are no I/O issues. I could have done more to isolate the controller as the point of failure, but at this point, noticing how everything works perfectly fine until it's connected to the Rocket 640L is good enough for me. I guess by shopping for a cheaper card, I paid the price. So far, the fix appears to be to flash the stock firmware. For anyone looking, the Rocket 640L (not to be confused with the RocketRAID 640L) uses the Marvell 88SE9230 controller.
 
I've started a separate pool for the controller to verify controller stability before I let that controller contribute to my main pool. Also, burst tests will be underway soon.
 
My thanks to Christopher for his patience and guidance. To show my appreciation, I've tried to be thorough in this thread in hopes that anyone who runs into a similar problem with the same card finds this thread.
 
- AJ

Thanks to eBay I was able to assemble the following parts for surprisingly little.
 
1. Supermicro SC846E16-R1200B 24 Bay 4U Chassis with  2x 1200Watt 80plus Gold PSU.
 
2. Supermicro X8DT6-F Motherboard
 
3. 2x Intel Xeon x5570 cpu's.
 
4. 32gb Hynix DDR3 Server Ram (4 x 8gb)
 
5. Supermicro passive heatsinks (brand new in the box)
 
I loaded Win Server 2012 R2 Datacenter (already had) and installed 17.5TB worth of HD's (already had). More to be added later.
 
Everything I got from eBay could have passed for new and it all works flawlessly.
 
I decided to go with Drivepool versus RAID / SS / other solutions for 2 reasons 1. Worst case senario all I can loose is a what's on the failed HD. The remainder of the pool leaves my data unmolested and accessible in any Windows computer (this is important to me). 2. I've been using it for quite some time now and it's proven to be totally reliable and simple (too simple but very good at what it does).
 
I'll be using this server for the following -
 
1. Data Storage
 
2. Media Server
 
3. Backups.
 
4. Blue Iris Server
 
Given the expandability of this system, it will probably outlive me.
 
Yes, it's noisy and sucks up the juice but it won't be running 24/7 and it's not located in a people area so neither matters.
 
Excluding the HD's and SW which I already had, The system cost me well under $1000. I'm happy.
 
 
 

Totally agreed.  I had a couple of USB enclosures.   Eliminated all of them due those very problems...   I do have two TR5M enclosures connected via eSATA.   Fortunately my server lives in the crawl space basement, and have never had a problem with disconnecting, especially since no one ever comes near the thing except me!

Hey, thanks for your perspective on this!   Just for good measure, I moved some drives around so that this big drive was inside the main box and not in one of the external enclosures.   Everything worked just fine!   Thanks for the tip!!!

Thanks for the suggestions, I just needed to know I wasn't going to screw things up too bad :-) Is there anything I need to know about restoring the client backups if I lose the data in the pool and need to put it back from the external backup?
 
On a side note... I noticed from another thread that you started using Emby. What are you using for the front end. I've set up the server on my windows server box and also set up kodi on the same box to feed a mysql DB and i'm running kodi on RPi 2's and PCs for clients. I love the combination and the ability to use Emby for remote access and transcoding.
Emby is like Drive Pool and Flirc in my opinion... these are elegantly designed programs that just plain work and the simplicity of the design is what makes the products great. Like the Flirc just simply showing up as a USB keyboard to the computer makes everything else behind it trivial.
 
Sorry for getting side tracked there.:-)

FYI - I just emailed it directly.  I'm hoping that the screen capture of the error message "Dashboard.exe - Application Error" message will help as it says the exception # and application location of error.

To complete the thread for my part I can report that BETA 612 now has the dashboard working correctly.  NOTE: Reboot client computers too before using dashboard after updating Drivepool.

Have you cleared the status in StableBit Scanner to see if the issue comes back?
 
To do so:
double click on the disk, click on the arrow next to "file system scan" and click on the button with the green circle on it, and select the "mark all damaged file systems as unchecked" option. This will clear the status and cause it to rescan the disk at the next opportunity.
 
Also, a surface scan may not detect file system errors. 
The surface scan detects areas of the disk that are unreadable. Not that are corrupted. The file system scan uses chkdsk actually, to find corrected file system entries (and to fix them).

Unfortunately, properly not.
 
The reason for that is that hard links only link files in between locations/files on the same "physical" disk. Because DrivePool is using a bunch of disks, it makes implementing this ... well, a nightmare.
 
However, we have a workaround to this problem in the works, actually.

With default settings, DP allocates to HDDs with most free space and it seems as if you have added a 1TB drive in a po with rather larger ones that are not filled halfway. It will be quite some time/data before that drive is used.

Hi scanner does just that if the settings are
Correct upon a falling drive it will start to evacuate data providing you have enough space on other drive, good practice would be to leave enough free space to cover the largest drive you have installed. I have mine set to only remove non duplicated data this to me is the quickest and best option then drivepool will start to reduplicate the files once you remove the failing drive.

Windows 10 build 10041 is working fine for me.  Got 44 drives in a pool.