Jump to content

Jaga

Members
  • Posts

    413
  • Joined

  • Last visited

  • Days Won

    27

Posts posted by Jaga

  1.  

    56 minutes ago, Christopher (Drashna) said:

    Also, don't use A:\ or B:\.  Windows is hard coded to treating these drives as floppy drives, and may cause odd behavior. For instance, the system will ping these disks every 5 seconds or so, IIRC. 

    Sadly true - 25+ years later and the underlying drive system still wants to treat these differently.

  2. Just for my own edification Christopher - does that also hold true when you have a single SSD set as the optimizer drive, and real-time duplication is turned off?  Is Drivepool smart enough to hold the single deferred copy of the file on the SSD only long enough to duplicate to the regular pool drives?

  3. If you have it configured in Scanner it'll email/text/push-alert you when a drive has issues.  Drivepool also has email notifications, but they're a little hidden.  Click the gear icon in the upper right, choose Notifications, then "Notify by email..".

    Glad you were able to figure out what was going on!

  4. Honestly, I don't know where verification is concerned.  You'd think it *would* be verifying all packets/chunks on both ends, but who knows.  The fact that FTP disconnects prior to completing means something somewhere is going on.  I've seen a lot of wonkiness (my tech term for it) over the decades with Internet communications, and it'll definitely make it easier to rule out the hardware/communications level if/when Christopher/Alex ask for logs and take a look.

  5. If your friend has checked the FTP server/volumes, it's entirely possible the corruption is coming from the communications layer - specifically your ISP and dropped/re-transmitted packets to/from the server.  Have you run some reliability tests on your connection to try and rule that out?  I recommend PingPlotter - download their free 14 day trial of PingPlotter Pro (you can revert to the free edition after it's over).  You'll want to look for Packet Loss (PL%), hops with spikes in ICMP (ping) times, etc.  At least a 8-12 hour trace with an interval of 10 seconds or less would be good.  You will want to try two independent tests - one to/from his FTP server, and one to/from a known reliable Internet server farm (like google.com).  If the tests from the second are good and his are dirty, it may indicate his ISP is having issues - he'd want to run the same tests in that case.  During the tests try not to hammer your internet connection with other traffic, as it will affect the results.

    Christopher may ask for logs from Clouddrive.  I generally try to rule out lower level pieces of the equation first (hardware, communications, etc) so that I know they aren't complicating the issue.

  6. I looked at mine - I don't see any attachment management tools, only filter and sort options.  You can browse to the post the attachment was in and edit it, then remove the attachment, but it's sort of a dirty fix.

    A better option might be a free account at Imgur.com (I use them a -lot-) so you can have virtually unlimited storage and easy links (like for bbcode and such).  They also allow regular and hidden albums.  Never had them email me with spam or try to force an upgrade to my account.

  7. What Umfriend said - screens would be helpful.  :)   Sidenote:  if the Prevent Drive Overfill plugin isn't near (or at) the top of the plugin list, it might get trumped by a different balancing algorithm.

  8. Mostly correct, yep.  :) 

    If A: is already a pool, step 5 is redundant.  The files would already be inside the hidden Poolpart-xxxxx folder(s) for it's drives.  To add to that pool, you'd just move files into the existing drive letter assigned to it (A: I assume).

    You can -manually- move files that are already on a volume (but outside the Poolpart-xxxxx folder) to a Poolpart-xxxxx folder in the Pool, but you'll want to do a re-measure on the Pool after that so it's aware of the new additions.

    You're correct in creating a hierarchical Pool (E:) and adding A: and D: to it.  If you're duplicating the entire pool there's no need to enable folder duplication, just turn on Pool duplication on E: and set it to 2x.  Not sure what folders you want to duplicate locally, but I'm assuming they're inside Pool A:, so you could easily enable folder/file duplication for them from Pool A's management interface.  Duplication on a pool requires >1 volume in the pool, so A: will have to have multiple volumes assigned.

    Any files that you put on E: will get auto-duplicated to A: and D:.  You'll want to do a re-measure and then kick off a manual re-balance, so that your Clouddrive is fully populated with files from A:.  Since you're using a Clouddrive for one of the volumes, you can turn off real-time duplication (a performance setting).

    You can remove drive letters after the hierarchical pool is created, sure.  And if you want you can use Folder Mount Points to keep them easily available (like from a C: root folder).  You do that in the same place in Windows Disk Management where you'd normally assign it a letter.  Just make an empty folder there for it first, then go assign the drive to the folder in Disk Management.

  9. Using a defragmenter on the drives in the pool won't hurt anything - Drivepool doesn't care where on the disks the files reside.  But letting a drive get to 0 bytes available can be a real issue for Windows (it doesn't like it one bit).

    Due to your space issues and different drive sizes, you should probably enable and use the "Prevent Drive Overfill" plugin in Drivepool.  You can get there by clicking the up arrow next to Manage Pool, then "Balancing...", then the Balancers Tab.  Find the plugin in the list and enable it, and try to set a reasonable "max full percentage" on the drives based on how much free space you have left (total).  Then move it up the order with the up arrow on the right so it's near the top.  You'll have to kick off a manual re-balance for it to take effect (usually turning on the balance immediately switch works).  If it doesn't kick off straight away, do a re-measure on the pool and it should.

    The SSD plugin really only helps with speeding up pool write speeds, and it will migrate files that land on the SSD to the regular pool disks so they don't stay there.  It's not considered permanent pool storage, unless the rest of the pool is completely full, in which case it would (reluctantly) continue to hold the files it stored.

     

  10. Also try turning off things like "make my programs and text larger" in the graphics properties (125% 150% etc).  Return them to 100% and reboot.

    And...  what's that little "your desktop is currently shared with ..." message on the bottom?  If you're using graphics overlays and/or programs that intercept graphics and share with others, they too might be interfering with Drivepool's display.

  11. So basically it just failed during removal from the pool (or was interrupted during the reboot), and when the machine came back up Drivepool identified it differently, though still as part of the pool (likely due to the hidden Poolpart-xxxxx folder on it).

    Here's what I would do, since the rest of the pool is still intact:

    1. Stop the Drivepool service.
    2. Find the J: drive in Explorer and turn on Show hidden files and folders so you can see the Poolpart-xxxxx folder.
    3. Copy the files/folders that are inside that to another of the Pool's drives (I: might be a good candidate) - make sure they go into the root of the other drive's Poolpart-xxx folder.
    4. If the copy tells you any of them are overwriting existing files, double-check the size/date before confirming/denying so you're sure you aren't overwriting good data.  If file size/date stamps match, you can skip those.
    5. When the copy is done you'll want to then move the files/folders from inside the J: drive's hidden Poolpart-xxxxx folder to it's root folder (i.e. J:\), so they aren't inside the Poolpart folder anymore.  That ensures you still have them as a backup if you need them.
    6. Start the Drivepool service and do a re-measure on the Pool.  It may indicate that J: is now empty and that balancing is off (since you may have had some duplicated files on J:.  That's okay.
    7. See if you can then remove both J: drives from the Pool through the interface.  You can always do a Force Remove if needed.  Start with the 2nd one in the list as you see it now.
    8. If you can't remove or force remove J:, stop the Drivepool service again, navigate to J: in Explorer, and manually delete the now-empty Poolpart-xxxxx folder.
    9. Start up the Drivepool service and program again and see if J: is still listed (either of them).  Try removing/force removing again if needed.

    Balancing may be off when you're done removing J:, but that's easily fixed by re-measuring the pool and then doing a manual re-balance.  Only kick that off when you're happy that J: is gone and the rest of the Pool looks okay.

    This is just my opinion on steps you can take to correct the problem while preserving files - different people may have different ways to resolve it.

  12. Just a few questions since it's Sunday and you may or may not get an official response today:

    • Can you get info from the "Error details..." link next to the second J: drive?
    • What was the original drive letter of the first drive you removed from the pool (successfully)?
    • Is the J: letter of the drive that won't fully remove from the pool it's correct letter (in the original pool)?
  13. With most of the topics here targeting tech support questions when something isn't working right, I wanted to post a positive experience I had with Drivepool for others to benefit from..

     

    There was an issue on my server today where a USB drive went unresponsive and couldn't be dismounted.  I decided to bounce the server, and when it came back up Drivepool threw up error messages and the GUI for it wouldn't open.  I found the culprit - somehow the Drivepool service was unable to start, even though all it's dependencies were running.  The nice part is that even though the service wouldn't run, the Pool was still available.  "Okay" I thought, and did an install repair on Stablebit Drivepool through the Control Panel.  Well, that didn't seem to work either - the service just flat-out refused to start.

    So at that point I assumed something in the software was corrupted, and decided to 1) Uninstall Drivepool  2) bounce the server again  3) Run a cleaning utility and 4) Re-install.  I did just that, and Drivepool installed to the same location without complaint.  After starting the Drivepool GUI I was greeted with the same Pool I had before, running under the same drive letter, with all of the same performance settings, folder duplication settings, etc that it always had.  To check things I ran a re-measure on the pool, which came up showing everything normal.  It's almost as if it didn't care if it's service was terminal and it was uninstalled/reinstalled.  Plex Media Server was watching after the reboot, and as soon as it saw the Pool available the scanner and transcoders kicked off like nothing had happened.

    Total time to fix was about 30 minutes start to finish, and I didn't have to change/reset any settings for the Pool.  It's back up and running normally now after a very easy fix for what might seem to be an "uh oh!" moment.

    That's my positive story for the day, and why I continue to recommend Stablebit products.

  14. Yep, if it reset to 0%, it sounds like Scanner detected something wrong there, or the drive dismounted and Scanner had to restart the scan.

    At this point I'd recommend stopping Scanner, shutting the computer off, removing the drive physically, waiting for the dock and mounting it in there, and then re-connecting the dock only after Windows (& Scanner) have booted up.  Then you can restart a scan on the drive.

    If it's truly damaged, allowing it to keep running in a situation where the problematic part could be the controller, or the port, or the cable..  probably isn't advantageous to retrieving data off the drive.

  15. Yes it does Alexander.  You need to make sure that "Read Striping" is enabled in Drivepool.  Click on the up-arrow next to "Manage Pool", move your mouse over "Performance" so it pops out, then make sure there's a check mark next to Read Striping.  If there isn't, left-click it to enable it on the pool.

    Note that it only works for pools with duplication on, or for folders/files that are specifically duplicated.  The pool needs to have >1 physical drive in it, and the read performance with multiple drives also depends on their speeds compared to each other.  Example:  if you have a pool with 1 spinner drive and 1 SSD drive and duplication is turned on, most of the time Drivepool will choose to read from the SSD first since it knows the performance is higher there.

    Drivepool's read striping can also take advantage of >2 drives with Read Striping.  The more drives you have the higher a chance of getting higher throughput.  It works best on larger duplicated files where reads can be sustained.

  16. 11 hours ago, Cekkent said:

    Can you expand on this statement, specifically on the "pool them together"? Creating multiple accounts is not a problem, and the ability to upload more than 750 gb/day for not a large increase in my cost with more gsuite accounts is VERY appealing. 

    Pretty sure what Christopher meant was to get a copy of Stablebit Drivepool, so you can take multiple Clouddrive drives and combine them into a single larger volume.  It makes multiple disks look like one larger disk, and works seamlessly with Clouddrive.

  17. As far as repair goes, Scanner uses chkdsk to scan/repair the disk.  It's pretty reliable in that regard, so I'd let it do it's thing first and see what it reports, then determine if you want to take it to the next level (3rd party repair tools).

  18. 1 hour ago, EldonMcGuinness said:

    Just for posterity, there is not a way to have CloudDrive utilize a cloud provider directly, sans container if you will, correct?  This is a major point to me as I would rather be able to manipulate the files directly, instead of having them in a container which prevents me from accessing the data via another platform.

    Since Clouddrive works on a user-set block size for custom chunks, faster manipulation/transfer, encryption, etc. you'll need to change how this is approached.  CloudDrive isn't emulating space - it's emulating a physical drive, and all of the features it uses really require that it handles this in a structured format (by blocks) and not by files.

    What you'd want to do instead is use file duplication/synchronization software (like SyncToy) and have it mirror whatever folders/files from your system that you specify directly to your cloud space.  You'll want to do it by mounting the space as a Drive letter, using the Cloud access software.

  19. Umfriend's suggestion of mounting it in an external dock after Windows has started is a good one.  And empty USB docks are usually inexpensive - I keep a spare or two handy for work.  It helps isolate it from the rest of your PC, but still allows all the usual utilities to be run against the disk/controller.  And it can even rule out controller/port issues due to being separated.

  20. 34 minutes ago, Carlos68 said:

    ... i think its pretty bad that there isn't a warning when looking at that update box that pops up telling you that all the data may be lost.

    It's pretty much a one-time thing, since they changed the way settings were stored to the newer json file format.  If your newer jsons were lost/corrupted, the method I suggested could get them back.  If you used a much older version that didn't have jsons, then you probably got hit by the upgrade bug.

    Still worth looking at the drive to see if the folder that holds jsons for Scanner has any old versions in it.  If not, you'll be re-doing settings from scratch, yep.

×
×
  • Create New...