McFaul

So. With Scanner uninstalled, but with Drivepool beta installed. It did crash overnight. Whats interesting, is that it is did it later than i was expecting. this time it was between 5-6am rather than 1-2am. Which makes me realise its not crashing at a certain time, its crashing after its been on for about 19 hours (since it was a sunday yesterday i got up and installed drivepool much later, at about 10-11 am rather than the 6:30-7am than i've been rebooting it on a weekday when i get up for work. its lasted 19 hours 22 minutes, and 19 hours 27 minutes, before crashing. I'm taking the drivepool beta off now, and i'll reboot to check and see if it survives 24 hours (since im no no longer sure about it living through saturday night since it wasnt running for a full 24 hours, i had assumed it was crashing at a certain time rather than after a certain up-time, now i need to redo the test and wait for a full 24 hour) Chris Ok, all Drivepool and scanner stuff is uninstalled and the server is up at 9am Monday morning. So if it is still alive tomorrow morning, then its got to be drivepool. If it crashes at around 4:30am tomorrow, then i need to look elsewhere! I'll keep you posted

Hi, Scanner i told it that it could scan at any time - since its a media server and its got a beefy enough processor it doesn't matter what its doing or when from my point of view. it was set to scan file system on the disks daily though, but i still wouldn't have thought that could trigger the driver to crash, or indeed at what time it would do that. if it turns out to be scanner causing the crashes that'll be the first thing i'd toggle. On Drivepool, some folders are duplicated, some are not. but the status bar is green - duplication is all finished, and its all balanced, so in theory drive pool shouldn't actually be doing anything? (unless it surreptitiously doublechecks the duplication in the middle of the night?) and my understanding is that scanner is more likely to BSOD than Drivepool? Server has now been up for about 24 hours. I've left that beta of drivepool , which fixes the renaming issue (which is already being a big help!), installed, and for tonight im leaving scanner uninstalled. It should be about 4 hours until it crashes if its going to....

Ok, Having totally uninstalled DP and Scanner, my server did not BSOD between 1-2am last night, the first time in five days its made it through the night. hopefully the dump file i sent you will give some more clues. I've just put the latest beta you linked to on, and that did indeed fix the renaming folder issue in metabrowser, so well done Alex! Chris

Hi, Yeah it just seems like a strange coincidence that it was been stable for 9 days and the same day i installed the drivepool beta has it started crashing on a daily basis, for 5 days, and always at the same time. it's 11pm here , so as a test I've uninstalled both scanner and drivepool and lets see if the server survives until morning - since its crashed consistently every single night between 1-2am, hopefully this will rule in / out anything to do with stablebit! if it still crashes even without scanner/drivepool i do actually have a second PSU i can add to the system tomorrow to try and rule out power issues. I'll test the new beta tomorrow morning and let you know how i get on with the renaming! Thanks for your help Chris

Ok, wiki working again now, dump file is uploading (mcfaul.rar) There's scanner logs, the mini dumps, and the memory dump, but thats the memory dump from when i logged in, not from when it crashed all by itself. ive now unchecked the windows option to overwrite the memory dumps, so i'll be able to send a fuller one if it does it again.

Now the wiki page doenst load at all and just says "A database query error has occurred. This may indicate a bug in the software." and yes, i am running stablebit scanner, im including the scanner logs in the zip too

Hi, I just woke up to find it had crashed again. I've also realised that its crashing by itself between 1-2am; but the second crash just before 6am , is actually when i log in (i assume it was just frozen). So im uploading the files now, since its not reproducible manually and its crashing during the night, im including the minidump and memory.dmp file UPDATE: for some reason the page on your wiki wont load for me.. i've tried two differnet browsers, on two differnet computers. the side bar loads, and the title loads, but the actual page and upload widget wont show up

CUSTOMER_CRASH_COUNT: 1 DEFAULT_BUCKET_ID: WIN8_DRIVER_FAULT_SERVER BUGCHECK_STR: 0x133 PROCESS_NAME: System CURRENT_IRQL: d LAST_CONTROL_TRANSFER: from fffff8002e9f4a3b to fffff8002e874440 STACK_TEXT: fffff880`015011f8 fffff800`2e9f4a3b : 00000000`00000133 00000000`00000000 00000000`00000501 00000000`00000500 : nt!KeBugCheckEx fffff880`01501200 fffff800`2e8b8ea1 : fffff880`01501350 fffff880`014d8180 fffff880`01501360 fffff780`00000320 : nt! ?? ::FNODOBFM::`string'+0x142f2 fffff880`01501280 fffff800`2ef85e94 : ffffffff`ffd0d9a0 fffffa80`09800507 ffffffff`ffd01000 fffff800`2e86cfa8 : nt!KeUpdateRunTime+0x51 fffff880`015012b0 fffff800`2e86cf9e : fffffa80`09800580 fffb75a9`15da9ac1 fffffa80`09800580 00001f80`00da0701 : hal!HalpTimerClockInterrupt+0x50 fffff880`015012e0 fffff800`2ef810ba : 00000000`17dab439 fffffa80`09803008 ffff4c81`00000000 2029302b`00000000 : nt!KiInterruptDispatchLBControl+0x1ce fffff880`01501470 fffff880`00abd557 : fffffa80`09803008 fffffa80`09803008 00000000`17dab550 00000000`00000000 : hal!HalpTimerStallExecutionProcessor+0x10b fffff880`01501500 fffffa80`09803008 : fffffa80`09803008 00000000`17dab550 00000000`00000000 00000000`17dab439 : 3wareDrv+0xe557 fffff880`01501508 fffffa80`09803008 : 00000000`17dab550 00000000`00000000 00000000`17dab439 fffff880`00abbc7f : 0xfffffa80`09803008 fffff880`01501510 00000000`17dab550 : 00000000`00000000 00000000`17dab439 fffff880`00abbc7f fffffa80`09803008 : 0xfffffa80`09803008 fffff880`01501518 00000000`00000000 : 00000000`17dab439 fffff880`00abbc7f fffffa80`09803008 00000000`0001d4c0 : 0x17dab550 STACK_COMMAND: kb FOLLOWUP_IP: 3wareDrv+e557 fffff880`00abd557 ?? ??? SYMBOL_STACK_INDEX: 6 SYMBOL_NAME: 3wareDrv+e557 FOLLOWUP_NAME: MachineOwner MODULE_NAME: 3wareDrv IMAGE_NAME: 3wareDrv.sys DEBUG_FLR_IMAGE_TIMESTAMP: 5123ea59 FAILURE_BUCKET_ID: X64_0x133_3wareDrv+e557 BUCKET_ID: X64_0x133_3wareDrv+e557 Followup: MachineOwner So its the 3ware driver crashing... but it does seem strange that the machine has been dead stable for 9 days.. and after i put drivepool beta on there, i get four BSOD in two days... but on the other hand, I'm not really sure how that could cause a BSOD? if anything i would have thought that it would be the scanner which is capable of causing BSOD (although i do not have unsafe IO ticked) I've also checked that the 3ware card has staggered spinup - but i would image that may only be true for system startup, and not for if all the drives wake up from sleep? And yes, newest firmware, newest driver.

I have a follow-up that is maybe slightly off topic, or maybe not... So my server ran fine for over a week, but two days ago it BSOD'ed twice during the night, just after 1 am and again just after 5am. Last night, it has again BOSD'ed between 1-2am and again just before 6am. (its automatically restarted rather than just hanging on the BSOD) The only change i had made was to upgrade to the "beta" drivepool. Since that didnt help with the metabrowser issue i've rolled back to the latest non-beta drivepool. but now the server seems to have gone down again. I havent had a chance to look at the dumps properly, but it seems to be hal.sys causing ntoskrnl.exe to crash. which isnt terribly helpful info... What i'm now starting to wonder, is whether I'm overloading the PSU, specifically the 5V rail. I have a RM750 PSU, which is rated for 25A on the 5V rail. Adding up just the drives in the system , I get 17A on the 5V rail (12 4tb disks at 0.52A, 13 6TB disk at 0.6A, 3 8Tb disks at 0.35A, and a handful of non pooled disks at just over 2A total). (and apparently some MB & PSU combos can draw 10A on the 5V... and drawing too much on the 5V can cause the 12V rail to do funny things, so im wondering if its affecting the CPU? Those current ratings are from the labels on top of the drives, for the 5V rail only. so im not even sure if they are peak or average, but that fact that it seems to be crashing at fairly predictable times makes me wonder is something causing them all to spin up at once (since in the middle of the night i assume they would all be spun down). Its of particular interest since i am planning on adding another 10 drives over the next few months. I'll have a proper look in the dumps when i get home later and see if i can see whats causing the BSOD, but thought i'd float the idea here in the mean time, see if anyone else has experience of running a lot of HDD's and the relative power requirements. i could always put a second PSU in there and run 24 of the disks off that. but obviously i'll hold off buying new hardware until i can rule out any software issues. Thanks in advance for any thoughts!

Hi, This is either manual or automatic. If i make a change to a movie then press save, it tries to save/rename all the movie files, then the folder itself, and its the folder step which fails. it works fine on other drives, its just on the pool it fails. the "rename history" for metabrowser shows: Time Original New Name 06/18/2015 07:07:24.148 P:\ServerFolders\Movies\4.3.2.1. (2010) Bluray tt1514041\ P:\ServerFolders\Movies\4.3.2.1. (2010) DVD tt1514041\ File already exists 06/18/2015 07:07:24.148 P:\ServerFolders\Movies\4.3.2.1. (2010) Bluray tt1514041\4.3.2.1. (2010) Bluray tt1514041.nfo 4.3.2.1. (2010) DVD tt1514041.nfo 06/18/2015 07:07:24.148 P:\ServerFolders\Movies\4.3.2.1. (2010) Bluray tt1514041\4.3.2.1. (2010) Bluray tt1514041.mkv 4.3.2.1. (2010) DVD tt1514041.mkv

interesting, lets see what alex says

im definitely buying five

None that i know of the backup i use the built in, and i dont run AV on the server C:\Windows\system32>fltmc Filter Name Num Instances Altitude Frame ------------------------------ ------------- ------------ ----- FsDepends 47 407000 0 DfsDriver 0 405000 0 DfsrRo 0 261100 0 luafv 1 135000 0 npsvctrig 1 46000 0

Well I'm in europe... and the Euromillions lottery is up to £95m this week... and thats lump sum and tax free

Server 2012 essentials x64 (non-R2)

Hi, Under "disk details" it shows the various "SMART tests" available, i.e. short, long, conveyance etc.. Most of mine show "no error / no test run" Is it possible to trigger a SMART test to run from within StableBit Scanner? or is it too complicated because each manufacturer has a different way of doing this? I know scanner tests the whole surface once a month, but i'd quite like to run the short SMART test weekly (which is what i had my Synology NAS set up to do - short smart test weekly, long smart test monthly) Thoughts?

I actually have a follow-up, which i'll post here for the sake of not wnating to flood the forums. In StableBit Scanner there is a column "Name". For my Intel (in-built controller) connected disks the "Name" is the model "WDC WD60EFRX-68MYMN1" for example For the drives connected to my new (LSI9650SE-24M8) all of the disks "Name" is "LSI 9650SE-24M8 SCSI Device" I was just wondering where the "Name" field comes from? If i right click on a disk and view details it can correctly identify both the model and serial number. Perhaps, since you can choose what columns to display in the main Scanner screen, could a column be added for "Model" so for when, like in this case, the controller isn't passing the "Name" (wherever that comes from) we could un-tick showing the "Name" column and instead tick showing the "Model" column. It would definitely be helpful for those of us with a lot of disks and a controller which doesnt pass the right info (i've just ordered another 24-port card.. so if i fully populate it then i'll have 48 disks all with the same "Name"! Is that feasible?

Actually no, thats with the 16 port LSi card... I took this photo as a moment as i was dismantling that build at the weekend, prior to putting it all back together with the 3ware card in my new D8000 i did look (briefly ) at that case i more seriously looks at the "45 drives" cases ( but they are still too expansive for what they are.

that didn't fix it. "file already exists" is all that metabrowser says zip file is uploading now... (incidentally the page says to "upload server.zip" and it should say service.zip thanks for your help!

Yeah i did look at expanders, but most of the ones i could find (in the uk) weren't priced too differently to just buying another card - and that gives me a level of redundancy in that at least if one dies, i still have a "pspare" card

Yeah, its a media server, the twelve 4tb drives are WD Se 7200, most of the rest at 6TB reds, and a few 8th seagate archive drives, so speed doenst matter to me, its just how many i can plug in

yeah i already spotted the boost - that upped it from 30mb/sec to 150mb/sec.. but it still seems such a waste to only be using 2 out of my 28 disks perhaps its non-trivial to make it parallel - since the UI tells you which specific directory it is working on, but its definitely something worth considering for Drivepool v3 as it would certainly speed up my duplication (which has been running since friday night and is still going...)

even tho that folder is non-duplicated? its on a per folder level.. so you are suggesting i un-duplicate my TV folder to fix an issue with the movies folder? (its worth being clear as my TV folder is 50TB so re-duplicating it will take a week!)

the pool has two root folders, movies and tv tv is duplicated movies is not

I would imagine because he said that there isnt enough room to do any duplication so while he wants pooling, he wants individual folders to stay together so if a drive dies.. he will only loose whatever was in the folder on that dead drive (which is not ideal.....) rather than simply losing a fraction of the files in every single folder (which would be a huge pain in the ass)