denywinarto
-
Posts
104 -
Joined
-
Last visited
-
Days Won
1
Posts posted by denywinarto
-
-
4 minutes ago, Shane said:
In the DrivePool GUI, at the very top where it lists your pool(s) - NOT the Disks section below that - what is the string inside the first set of parentheses?
E.g. on mine: "DrivePool (P:\) - (39.1 TB)"
Ah it's DrivePool (N:\), So this is what i should use? don't wanna screw my installation
-
5 minutes ago, Shane said:
Hmm. Just checking, you are using the pool's drive letter and not the bad drive's letter, quiaff? If the pool itself is mounted as a path (e.g. "c:\mounts\poolP") instead of a drive letter (e.g. "p:"), try the path. If it's not mounted at as either a path or a letter, try mounting it as either first and use whatever you've assigned.
Complete path doesn't work either, i have edited my post above, the complete path to pool is like this :
I'm mounting it as folder
C:\Drivepool\HGST_41_12TB_8DH24RYH\PoolPart.654e1b5c-05b8-44a2-8b6b-a0251f2ec7d6
-
7 minutes ago, Shane said:
Try the following command from a command prompt run as an administrator where pooldriveletter is the drive letter of the pool itself and poolpartname is the hidden folder beginning with "PoolPart." on the bad drive:
dpcmd ignore-poolpart pooldriveletter: poolpartname
This should tag the relevant poolpart to be ignored by DrivePool at the system level, which will effectively make it "missing" as far as the pool is concerned; the pool will become read-only until this is reversed or the "missing" drive is removed from the pool - which you should be able to immediately do as DrivePool will treat it just like removing any actually missing drive and drop it without attempting to retrieve anything from it.
C:\Users\Administrator>dpcmd ignore-poolpart HGST_41_12TB_8DH24RYH: PoolPart.654e1b5c-05b8-44a2-8b6b-a0251f2ec7d6 dpcmd - StableBit DrivePool command line interface Version 2.2.4.1162 Error: Can't open handle to volume.I am not using alphabet as volume though, do i need different syntax? Complete path to folder doesn't work either
C:\Users\Administrator>dpcmd ignore-poolpart C:\Drivepool\HGST_41_12TB_8DH24RYH: PoolPart.654e1b5c-05b8-44a2-8b6b-a0251f2ec7d6 dpcmd - StableBit DrivePool command line interface Version 2.2.4.1162 Error: Can't open handle to volume. C:\Users\Administrator> -
43 minutes ago, Shane said:
Updating shouldn't make things worse but if the drive's at the point where it's generating BSODs then I doubt updating would make things any better either.
Personally I'd physically remove the drive from the server altogether (it'd then show up as a "missing" drive in DrivePool and can be safely removed from there); if I still needed to (try to) get any remaining data off the bad drive I'd use a separate test machine so any further BSODs wouldn't interfere with the server.
Is there any software method to temporarily prevent DP from writing to the bad drive?
I'd need to prepare a maintenance time but in the meantime I dont want anything gets written to it, maybe renaming the poolpart or something?But if it's not safe and causes BSOD then it makes no difference i guess
-
OS : Windows server 2019
DP version : 2.21162
Hi, just wondering, since my version, would there be a fix that mitigates this issue?
So I have this 12tb with bad sectors, i removed it overnight then it seems to stuck at 90%
Tried stopping removal process but after 20 min it doesn't stop
Then I searched for files in that drive and delete them manually, DP still stuck, so I stopped removal process, it finally stops.
Tried removing the drive again in DP.
But not long after that BSOD kicks in. Which is making me anxious cause I have alot of drives in my machine.
Should i update to the latest version to prevent this?
All 3 options are checked before removing the drive.
-
Topic, my drivepool is also pending update.
Any idea why?
-
17 hours ago, gtaus said:
I had some problems with a HDD that Stablebit Scanner did not detect. I found a very useful program called Hard Disk Sentinel which is like Scanner, but also different. Hard Disk Sentinel comes in both a free version, which is what I am using, but also a paid version with extra features. The free version of Hard Disk Sentinel will check/test your HDDs and give you an estimate of their Health Status, which is a %. For example, most of my HDDs have a Health Status of 95-100%, but the failing drive tested out and reported an estimated 3% Heath Status with a warning to immediately move all data to another drive. It correctly reported that the drive was failing, whereas, Stablebit Scanner, for whatever reason, did not flag it. The paid version of Hard Disk Sentinel has extra features which allow your to "fix" some HDD problems and restore the health of the drive towards 100%. Essentially, from what I understand, it scans the HDD for bad sectors, blocks them off from use, so they no longer are used to store files and corrupt your data. In my case, the bad HDD was totally failing and it could not have been fixed even with the paid version.
Thanks for sharing your experience, thats also what i've been using together with stablebit scanner.
My only concern is that it lacks daily scanning scanner has, or at least i havent figured out how to do it since i'm used to rely to scanner.
Sentinel also has this estimated lifetime, but i found out it wasn't always accurate as well. One of my 6tb red wd died before the estimated time.
But then again my experience with scanner wasn't much better, iirc it only warned me a smart error less than 24h before the drive died.
For me now whenever a drive shows filesystem error multiple times im replacing it right away.
Edit : i forgot to add, some of my drives were failing even at 100%, so its not a guarantee.
-
Just windows defender.
I have ticked all options and i'm pretty sure that one as well.
You're right about SS it could be really a PITA.
But i dont see any other options at this point that could stripe 4x3tb drives. Drivepool doesn't seem to capable.
Even 8TB ssd still costs $800, 3x 4tb wd black is not even half the price, not to mention i have to sell these at less price.
I suspect i will have to rebuild the Storage space pool from scratch,
makes sense somehow because these drives have been migrated from a different motherboard, (but it was working fine for 3 months after migration)
I haven't actually tested storage spaces running without Sb scanner on my main server,
just on my spare server with exact same motherboard and OS (it was fine for 3 days with scanner service stopped)
It's too risky since i'm definitely sure it messes up the OS.
Just bought another HDD for the backup, i'l ltry recreating storage spaces from scratch once it arrives.
-
Bad news, this is what i did.
Made a backup
Installed 3x4 TB wd in the rack, (In my other spare machine with same motherboard and cloned OS it had no issue)
The SS Disks were offline. So i thought i'd exclude it from scanner's surface test lists first.
Install stablebit scanner, exclude it from scanner
It does appear on scanner as "Microsoft storage space drive" IIRC.
All of a sudden BSOD, haven't even re-initiate the disks yet.
So i unplug the 3x4 TB.
ill got BSOD.
So i restored the backup before scanner installation, thankfully it appears to be stable until now.
So conclusion is, excluding it from scanner's list doesn't work..
I'm gonna check the SS drives on my spare machine again, but last time it was repaired and there were no issues
-
3 hours ago, Shane said:
Yes, I read the post.
If your drives are now "fine", they should be able to cope with being scanned. If they still can't cope, then you have bigger problems than a damaged MBR (and based on my own experiences, a damaged MBR is a symptom of something else having gone wrong).
I'd also be changing away from Storage Spaces anyway, because collapsing into RAW because its drives got poked by a diagnostic utility seems a terrible failure mode for a storage system.
"3 days and not a single response? Not even my ticket gets responded"
3 days isn't a long time for a small programming business with just a few employees (and two-ish volunteer forum mods). For comparison, Dell has hundreds (thousands?) of service reps and earlier this year Dell took four months to finally find and send me a backpack I'd paid for. However, since Alex and Christoper are (in my experience) usually quite prompt in responding to fault tickets, it may be that something has happened (e.g. COVID, though I hope not). If you haven't had any response to your ticket after a full week, let me know via a direct Message and I'll see what I can find out.
P.S. If this was happening to me, I'd be wanting to know if I had a bad drive and I'd be trying to find out exactly which drive(s) trigger the BSOD when scanned - that's why I suggested turning Scanner off for all but a single drive at a time.
Okay, i'm not expecting a fix in 3 days or something like that,
Just a word acknowledging this issue and a confirmation of future version with fix would be nice.
In my opinion this should be regarded as critical issue because its affecting OS stability.
About storage spaces,
Well, I need 12tb with ssd-like read speed, with the current price of SSD i can't find any other alternative which offers similar performance to size ratio,
and i'd rather stick with what i have now. Maybe once SSD price goes down i'll reconsider it.
I plan to make OS backup and run scanner and turn the scanner off for the storage spaces drives. If anything goes wrong i can just revert to the backup, but hopefully i wont need to do that.
-
10 hours ago, Shane said:
Keep in mind that Stablebit is not some huge company, it's a small business. Current circumstances (COVID et al) might not be helping either.
Can you avoid using Storage Spaces? I've never heard good things about that "feature" outside the marketing hype, only horror stories about people discovering that it seems great until if (when) something goes wrong, then it tends to go BADLY wrong.
Could be a bad drive (WD Blacks are usually great drives but I've had one go bad so don't rule them out), could be a bad controller. Try using Disk Settings in Scanner to turn off surface scanning for all drives, then turn it back only for a single drive each night to discover if it's a specific drive (or drive cable, or drive caddy, or controller port) at fault, then after that turn on an additional drive each night (1 drive, 2 drives, 3, 4, etc) to discover if it's a controller/driver/system load-related issue. Isolating a hardware fault can be a pain in the backside, I know from experience.
Edit: regarding "While i know the bad drives are the fault, i think scanner could do a better job preventing windows BSOD trigger", the short story is that's not possible - a BSOD is the OS kernel equivalent of an Emergency Stop button on factory machinery. If something is wrong with the machinery and someone (Windows) decides to press the button, by design nothing is supposed to stop the machinery from being halted.
Did you read my post about the chronological order of BSOD?
windows server 2019 + corrupted Storage spaces MBR= no BSOD, windows simply gives warning "this drive needs to be scanned blablabla"
windows server 2019 + corrupted Storage spaces MBR + stablebit scanner running = BSOD, right after inserting license code and scanner starts running. As you can see on the reliability pic on my last post.
The fix is for scanner is quite simple, the trigger is bad MBR sector from Storage spaces harddisk, so simply prevent it from scanning once it detects corrupted SS drive.
I know this is stablebit forum, but without customer critics a product won't get any better.
the 3 drives are now fine in SS after chkdsk, without stablebit scanner of course.
-
3 days and not a single response? Not even my ticket gets responded
Scanner is an abandoned product then,
Anyone know a decent alternative?
-
Update : thankfully CHKDSK saves the day. it says the MBR was corrupted but it was able to repair it
So now i can probably use my storage spaces again
But should i use scanner or not now? Any words from Chris / Alex or the devs?
Can't risk another BSOD
-
Well got another case now with stablebit scanner. This time its worse, its 3x4 tb wd black in storage spaces. It became RAW after BSOD
I'm pretty sure its scanner cause as soon as it starts scanning after installation BSOD occurs. And the reliability windows page back this up.
(I installed scanner on 6.36 and input the serial about half an hour later.)
Its not drivepool cause the drives arent in the pool. This is on newly installed Server 2019, i just reinstalled it yesterday
I immediately uninstall scanner.
What should i do now?
I need scanner to warn me of bad drives so i can replace them before they go bad.
But i have 2 dozens of drives if a single bad drive + scanner = BSOD then that means 2 dozens reliability holes.
Edit : I should have add, The reason i reinstalled my machine was because i'm having multiple BSOD's everyday.
Now i suspect my recent midnight BSOD's were caused by scanner scanning at midnight.
1. A new installation BSOD'ed right after i install scanner and it's scaning
2. The BSOD always happens at midnight, i check task scheduler and nothing is run at midnight. Seeing scanner's description of "always scans at midnight" reminds me of this.
While i know the bad drives are the fault, i think scanner could do a better job preventing windows BSOD trigger
-
3 hours ago, muaddib said:
Which service do I shutdown?
You dont need to disable any service if that's what you mean
-
2 hours ago, muaddib said:
I am having the same problem removing a drive. I am running it on Server 2016 OS. It blue screens about 45% through the removal process. I have a new drive that I want to replace it with. I wonder if I could copy the files from the existing drive to the new drive and swap them. Will that work?
1. Manually move all files to the outside of the poolpart.xxx folder on the bad disk so that drivepool can't see them any more.
2. Remove the old disk from DrivePool (it will be instantly complete since the poolpart folder is empty).3. Insert new disk (new different PoolPart.xxx is created).
4. Manually copy the files from the old disk to the new poolpart folder on the new disk.I found out by doing this i didn't get BSOD compared to moving the disk using DP
On 10/9/2020 at 10:27 PM, Shane said:Buying decent hardware (not always easy/obvious I admit), updating firmware and software (OS/drivers/apps) to the latest "stable" or "production" version (e.g. avoiding anything marked as "unstable" or "beta" or "alpha" where possible) and keeping regular backups.
Above is why i think the BSOD is related to drivepool, what i'm worried about if one my drives go bad then it would just throw random BSOD and screw up my OS
-
10 minutes ago, Shane said:
I'd say it's a "known" issue in the sense that drives can go bad in various ways, and some of those ways aren't necessarily handled well by the various hardware and software layers involved (e.g. drive controller, controller driver, kernel file system driver, etc). Any one of those throws a tantrum in response, there's only so much DrivePool can do about it.
So you're saying its more like OS issue?
Hmm i'm using server 2019 and LSI 9300 8e..
Any possible workaround that is recommended?
I mean eventually all my drives are gonna go bad someday, just wondering if there's a way to prevent it from screwing up my OS installation
-
About 2 days ago i instaled 1tb hdd which i thought was still in good condition because HDD sentinel and stablebit scanner didnt report anything strange.
Last night i think my server got rebooted couple of times and some drivers got corrupted.
Thankfully i made daily backup, so i restored it.
But then just few hours ago it rebooted again, and that 1tb HDD became "RAW" and offline.
I also remember when moving bad drive from the pool last year i got BSOD once or twice.
So Is this a known issue? I'm using the latest build.
-
5 hours ago, KingfisherUK said:
I believe you can have two GPU's, i.e. the esxi default svga and the iGPU passthrough but I'm not sure if/how you can specify which one it should use.
Any links for this? Have you seen people achieve it?
I could make windows detect them both in device manager, by still enabling the svga, but running gpu-z gives me svga as the detected gpu.
I'm also using the igpu for blueiris so im forced to use windows
-
29 minutes ago, KingfisherUK said:
Although my storage server is Windows based, I've gone with VMware ESXi for my hypervisor. PCI Passthrough is pretty straightforward to setup and I've got my Perc H310 passed through to my storage server VM. My other machines are all Linux based, accessing the drive pool over Samba.
I have a funny feeling Hyper-V doesn't do PCI passthrough at all, or if it does it's not very good.
Yeah, strangely the machine script says my lsi-9300 8e cant be passthrough'ed
So I tested with esxi as well. It worked but there is one issue that makes the setup not ideal, to passthrough the igpu, i have to disable svga, thus i lose console display and i have to rely on rdesktop alone. Not ideal for 24/ server IMO
Are you by any chance also pasthru your gpu / igpu?
-
Ah i thought you were using hyper-v lol..
Thanks for the license tips will keep that in mind.
I'm leaning towards hyper-v cause most of my environment is windows based..
-
Thanks chris, how long have you been doing it? Is it stable enough?
Also there is something i forgot to ask, were you able to passhthrough the igpu as well?
I will need it for emby and blue iris i think
-
On 1/31/2019 at 4:41 AM, Philmatic said:
Are you asking if DrivePool is compatible on a Hyper-V guest running on a Hyper-V host? If so, yes. I run it this way and it works fine, you just have to pass each disk individually and the guest OS (With DrivePool) sees it just fine. In fact, I went form a full physical setup on Server 2012 and Virtualized Server 2019 on Hyper-V, passed all the disks (Was a huge pain, not being able to pass thru controllers is a huge PITA, VMWare FTW here) and DrivePool picked up the entire pool (32 disks) within a minute or so and that was it.
Or are you asking if DrivePool is compatible with the Hyper-V Hypervisor itself (Either in Hyper-V Server or Server Core with the Hyper-V role) as in you want to pool drives inside the Hypervisor and present them as large disks to guests? I've tried this, and it DOES work, but it's not a good idea. DrivePool is not a traditional RAID technology can isn't well suited for large random writes, it's just not recommended. It's best to pass the disk individually and let DrivePool on the guest pool everything together.
Hi Philmatic, sorry for bumping an old thread,
I thought controller passthrough is possible already on server 2019?
https://www.reddit.com/r/HyperV/comments/bc3vvc/pcie_passthrough_possible_with_a_hba/
Have you tried it?
If it does passthrough the SMART status as well then it's probably the better method
-
Anyone ? Chris? Have you seen this implemented without issue?
Removing damaged drive for the second time triggers BSOD
in General
Posted
Thank you, i can finally remove it, will remember this trick next time removal gets stuck!