Jump to content
  • 0

Damaged disks in drivepool


Kazooba

Question

What can I do when a damaged disk from drive scanner starts rebalancing in Drivepool?

 

I have recovered any files, but drivepool will not let me use any disks that have been "damaged" in the past. I have two 3-4tb drives with 332KB or so on each that was "damaged" that I recovered. When I add them back to drivepool it won'y copy to those drives.

 

THX,

Kurt

Link to comment
Share on other sites

14 answers to this question

Recommended Posts

  • 0

In Scanner click SMART details for the damaged drive and click IGNORE.  You will have two options ignore current or ignore permanently.  Ignore current flags the disk as ok and if it gets works you will get the SMART warning again.

Link to comment
Share on other sites

  • 0

@Doug, Damaged != SMART errors. Damaged means damaged sectors.

 

 

@Kazooba, you can change DrivePool's behavior by chaging the balancer settings for Scanner. But I wouldn't remcommend it.

 

As for allowing it to use the other disks... if you run "chkdsk /r" on these disks, it will repair the bad sectors (by remapping them).

We don't do this automatically, because it means that you won't be able to recover data in that case.

Link to comment
Share on other sites

  • 0

@Doug, Damaged != SMART errors. Damaged means damaged sectors.

 

 

 

Unless of course you are unfortunate enough to have a system full of Samsung HD204UI ( A few years old now). In which case smart errors mean the drive is (probably) healthy except that polling for smart data could  then damage them. But good old sammy even brought a firmware to fix the issue and the firmware upgrades the firmware number to.... Yep the exact same firmware number. 

I had lots of fun upgrading them all before realising this.. Still to this day I have no idea if I managed to do them all. Hence I turned smart reporting off for all the Sammy drives. Thank god they are all so close to being replaced from my system.  :lol: 

 

Link to comment
Share on other sites

  • 0

Wow, that's .... Wow. I was not actually aware of this.

 

To be honest, in that situation, I'd pull out ALL of the disks. Put them into a pile. Then one at a time, update the firmeware, remove the disk and put into a "finished" pile. Screw up time! I want WORKING drives.

 

The putting them in a pile has often crossed my mind, though to be honest the next stage has involved a sledge hammer. :)

But fortunately despite all this I'm pretty sure I haven't lost any data. As mentioned in another post I'm working my way through them and replacing them all. No idea why a 4TB external is cheaper than an internal drive in the U.K., but the Seagates that I am pulling out of their enclosures are giving me write speeds of 150-180M/bps bursting at mid 200M/bps. 

Probably not state of the art stuff but affordable and it will bring my data drives down from 9 (running on different speed conrollers) to 5 all running on a Sata III controller. This should make my case look tidier too. (Stablebit scanner likes them also, so full smart reporting on all my drives at last.) 

Then when funds permit, I will invest in a decent Sata III card to future proof expanding data. Although possibly I may be upgrading the Mobo and CPU first. 

Then once replaced, I'm wiping the Sammys, doing the firmware then possibly running spinrite on them. Then they're are all going back in the USB 3.0 cases from the Seagates.

 

As a side note on the occasions that I have reinstalled my server, the initial stage that scans the system and decides what drives are installed and then where to install the O/S etc.. It takes absolutely ages (15 minutes or more). I'm thinking that this may be the cheapo Sata I or Sata II cards installed that are slowing things down. Boot up times are also a good few minutes so I'm hoping that once I can pull the existing cards things will improve. I'm only guessing the sluggishness is hardware related because If I install with ONLY the SSD O/S drive attached its just as bad. 

Can an add on card (even if not in use) slow things down so much? If not then It's definitely time to look at upgrading the Mobo.

Link to comment
Share on other sites

  • 0

LOL. Yeah, destroying drives is oddly satisfying. Though I like collecting the magnets. I've always loved magnets! :)

 

As for the cheaper... it's the same in the US. However... the warranty is different. Usually 1 year instead of 2-3. Also, if you remove the disk from the enclosure, you void the warrranty technically.

 

Also, because the drives are in an enclosure... and usually without any cooling, the drives run much hotter (up to 30C hotter!), so the lifetime of the drive is shorted.

 

As for a SATA III card, I recommend anything running LSI really. Expensive ... but worth it. Or if you want to be cheap, get an IBM ServRAID m1015/m1115 card and "cross flash" it to IT mode (works as a HBA/controller card). You can usually find these for about $100-150USD (so roughly 70-100EUR, I believe?)

 

 

As for the long times, absolutely. If the controllers are slow (or you have a LOT of them), it can take a while to "ping" all of them and get a response. 

That's also the advantage of that "cross flashing", you can opt to not flash the management ROM on the card, which means no additional wait time. :)

And yes, the cards can slow down boot time. When each is loaded by the system, it has to load it's management routine, which can take a few seconds or more.

As for in the OS, a bad driver could potentially cause slowdowns with the Virtual Disk Service (how windows interacts with the drives).

Link to comment
Share on other sites

  • 0

OK thanks for this info.

I'm not too worried about the warranty, its a comprise that my *bank balance" is willing to make. :)

Time will tell with the controller cards... Hopefully the dodgy 2TB Samsung drives will be gone by the weekend and thus the number of drives. This will then enable me to get rid of the cheapo sata cards. If it isn't those that are causing the slow loading times then I'm pretty much at a loss. Even during and after a fresh install the issue is still there. There is no other hardware attached to the system! 

I will see how I go when my final 4TB drives arrive.

Thanks as always Chris. :ph34r:

Link to comment
Share on other sites

  • 0

Well, if the issue persists after removing all that hardware.... then please do open up a ticket.

 

 

And you are very welcome.

 

 

Also, which 4TB externals are you getting? 

Either you didn't mention it, or I skimmed over it, or .... I forgot.

 

Well i'm pretty sure the issues are nothing to do with Drivepool but thanks anyway.

 

The Externals are Seagate expansions. Price was definitely a factor for me especially needing 6 of them. They're only 5900 RPM but I'm getting aroung 150 M/bps write speeds bursting at over 200M/bps on the sata III ports. At the end of the day my server is more a Media server than anything so I couldn't really justify spending too much. As you probably guessed from the other post they have all arrived and the migration has begun.  Then hopefully after this set of shifts I can kick back and load up Plex. (Not you're favourite media server I know  :P . But I have a Samsung TV with it built in, so the obvious choice for me. So until I can eventually get round to building an HTPC it does me fine.  

Link to comment
Share on other sites

  • 0

Okay, I just wanted to make sure.

 

And yeah, the external Seagates are very nicely priced, and do performance pretty well.

 

And It's not that I don't like Plex. It's that I've never gotten good performance out of it. Works okay locally, but remotely? Lags or buffers frequently. Even with a good connection. :(

Link to comment
Share on other sites

  • 0

I tried it a MONTH ago. Locally, it works great (minus a hiccup every once in a while), but outside my network..... buffers every 5 minutes or so. Or just starts dropping frames like crazy. 

And I have decent upload (4mbps), so it shouldn't be an issue.

 

Fair enough..Its the same as everything I guess. There's no good or bad just whatever works for the individual.  :)

Anyway it's confession time... As a user of Drivepool since the very begining , I have to admit to having a MAJOR issue that I was almost willing to blame on Drivepool for the first time ever  :o

 

As mentioned on various threads I have being upgrading all my samsung 2TB drives with the dodgy firmware. Since last mentioning how responsive the system Is with the 4TB Seagate replacements, one of the Seagates was behaving badly and Stabebit scanner threw up an error. It became unresponsive and would crash at anything more than reading the drive. So copying data from it was fine, but if i selected move.. It was as if even deleting the data after it had copied would crash this one particular drive.It wouldn't even let me format it or run chkdsk. 

Anyway not ideal but one drive out of the 6 I purchased was no big deal and easy to RMA being less than a week old. In the meantime I had decided to purchase another to use as server backup.

It was fine until you helped me with the permissions issue on the system files that I posted about a couple of days ago. After that the new drive showed the same symptoms as the "fatally damaged" drive. 

 

Normally I am keen on really troubleshooting issues but this was all in the middle of a string of 8 night shifts. However yesterday I had a little time before work. 

 

It turned out the issue was nothing to do with Drivepool or even failing drives or even a failing sata port. It was simply a dodgy sata cable throwing up I/O issues. A replacement drive in the same sata port using the same cable was just too much of a coincidence. Both the drives are fine it seems. 

 

I'm not sure if there is a way to force Scanner to recheck the drives immediately but that would give me ultimate piece of mind. On the subject of which I think it would be great we were able to select an individual drive and let scanner ...erm... Well.. scan it basically.  :P

Link to comment
Share on other sites

  • 0

Wow, yeah, a bad cable can cause all sorts of issues. And it's a weird issue to troubleshoot. Glad you were able to identify it.

 

And I know what you mean about the "string of 8 night shifts".... though, I just keep weird hours (timestamp for verification :P).

 

As for Scanner, there is absolutely a way to force a new scan. If you click on the "+" button on the left hand entry of the disk, it opens the drive map. On the left hand side, there should be a button with a green circle with a circular arrow. Click on this, and select the "mark xxxx as unchecked". The "Unchecked" state should cause StableBit Scanner to scan these sectors immediately (depending on your settings).

 

And you can do the same for the "File system" section.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Answer this question...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...