Jump to content
Division-M Community

Davey126

Members
  • Content Count

    39
  • Joined

  • Last visited

  • Days Won

    3

Everything posted by Davey126

  1. Yeah - I also considered multiple pools but the high (and increasing) file counts aligned with the inaccessible pool and nearly identical drive flyouts suggest a single pool that is still being built. What I think may have happened is the "I:" drive links (loose term) got farkled and are pointing to an unpopulated pool. But DB doesn't know that and won't allow another MP to be created to the populated pool because it thinks I: is serving that purpose. To my knowledge there can only be one MP per pool. The remediation may be as simple as a pool repair or a little more involved process that involves carefully deleting (or renaming) some of unconventionally named folders DB generates for each pool. It's pretty easy to determine which folder contains pool data; just look at the size. As others have said - open a ticket. I don't think the fix will be difficult...just need to know the best procedure.
  2. I assume you have this solved or at least have been presented with additional 'clues' to what's happening under the covers. Before responding a quick disclaimer: My use of DB is rather simplistic (two drives) and I have done very little maintenance (add/remove drives) after my initial pool was stabilized. But I also experienced a rough patch getting DB set up and another with the migration to R2. I'll try to help but welcome comments from more experienced DB users. First and foremost I suspect your data is safe. I have yet to see DB permanently mangle/loose files either through personal experience or forum postings. Yes, some people report losses but there is usually another factor in the mix such as RAID or a flaky drive/connection. Given the number of drives and large numbers of files I suspect your pool is still under construction. Give it some time. I assume you have rebooted as this is a requirement to complete a drive conversion (as noted in the documentation). At some point you are probably going to have to remove your existing mount points to individual drives and runs some type of repair to make the pool visible to one (and only one) mount point. I suggest you open a ticket with support to determine the best path forward. I believe a drive conversion (vs add to pool) accomplishes its magic through operating system 'hard' links, junctions, etc. If true I'd be a little concerned with the huge number of these that were created during the conversion and the potential, however unlikely, of downstream corruption. Might want to discuss that with support. Sorry I could not be more helpful.
  3. I have had mixed results with Acronis. On some systems it is rock solid while other is either starts or becomes flaky over time for no apparent reason. Just to rule out Acronis try few quick images with the free version of Macrium Reflect. I am not endorsing this product but it has worked quite well for me across a wide variety of HW configurations. To date I have not experienced any issues obtaining a VSS snapshot from individual drives in a DB pool on Win 8.1 x64. I have other pools on Win 7 machines but have yet to image the individual drives. Link -> http://www.macrium.com/reflectfree.aspx
  4. I can't comment on Win 7 but drives in my DB pool successfully sleep on Win 8.1. I also can't vouch for the timing but I have no reason to believe it doesn't follow the schedule outlined in the active power profile which is configured to sleep hard drives after 60 min. I have also noticed the drives spin up independently depending on how files in the pool are distributed.
  5. The latest version of Everything (1.3.4.686 as of this posting) doesn't error/hang on DB volumes listed in the NTFS section...but it still doesn't include them in the DB database as there is no USN journal. As a workaround you can designate the DB volume (or selected subfolders) as an Everything 'folder' but that requires a manual indexing run (plus periodic refreshes) and is usually not suitable for drives with a large number of files. It works for me as my DB mount point only contains a few thousand files which can be indexed in seconds. Note: the DB index only contains file metadata - it does not consider file content as other indexers do. Windows 8.1 x64 running on older hardware.
  6. Same. Service start, health check and balancing all working fine. Just installed over 2.0.3.8 and rebooted. Happy camper!
  7. Had the same problem locating a copy of v2.0.3.8 after v2.0.4.0 proved faulty. I have both now; PM me if you need a copy. v2.0.4.0 installer SHA-1 which others can confirm: 9FEA8511BB3D0344131F20B3CF65AF01CED855FD
  8. Spoke too soon. Just noticed queued tasks for Health Check and Drive Balancing dating back two days. Manual start doesn't work either. Rolling back to v2.0.3.8 ...
  9. My numbers start and remain small (under 3 MB) for the DB service. However, my implementation is relatively simple with only small amounts of data written each day (10-20 GB). Just another data point. Win 8.1 x64.
  10. Interesting - I had the same problem with v2.0.2.9 beta which was never resolved. No one else reported the same issue. I wonder of some 'services' don't get installed correctly vs a problem with the underlying code.
  11. @DIE_KUH - v2.0.4.0 runs fine on my Win 8.1 (Update 1) x64 box with a couple TrueCrypt drives. I had troubles with earlier betas and ultimately had to revert to v1 but all has been well since as did v2.0.3.8. Just adding a data point; obviously your difficulties are real and may be shared by other perspective users.
  12. So far v2.0.2.9 beta has remained stable although a few UI inconsistencies persist which I assume will be sorted in time. However, I can not seem to trigger a manual balancing run and automatic balancing has yet to kick in (I tweaked the schedule to 'aggressive'...normally set to once/day). When I look in the real-time monitor I see the balancing job is scheduled but it never runs. Same thing with heath checks. However, file validations run perfectly. Go figure. There is a 100GB difference in free space across the two drives in my primary pool with lots of small files and directories on both sides. Balancing on v1.9.5 seems to work just fine. I verified the v5 drivers are installed and did a full reinstall/rebuild/repair just to make sure. No sign of the v4 drivers. Any ideas?
  13. I found system performance much improved during heavy pool activity (writes or balancing) by excluding the processes DriveBenderManager.exe and DriveBenderService.exe from Microsoft Security Essentials (Defender on Windows).
  14. The release notes suggested some significant performance improvements in this release and I can confirm they are working - at least for me. Large file IO rates now approach that of writing directly to the drive while CPU utilization is down dramatically. I haven't done any formal timings with small files yet but subjectively they appear to be copied faster than with previous versions. Well done.
×
×
  • Create New...