Jump to content

Spider99

Members
  • Posts

    544
  • Joined

  • Last visited

  • Days Won

    39

Spider99 last won the day on May 19 2022

Spider99 had the most liked content!

About Spider99

  • Birthday 09/10/1964

Profile Information

  • Gender
    Male
  • Location
    UK

Recent Profile Visitors

1568 profile views

Spider99's Achievements

  1. is your ssd cache filling up as 3600 torrents is a lot?
  2. Should be fine IF you have duplication turn off - but if i remember correctly turning duplication off may make the pool delete the copies automatically hence why it says to stop the service so that does not happen i would do a test on a small folder with only a few files and see if turning duplication off does what i remember
  3. Updated to .1285 beta via Dp interface - completed fine for me
  4. Partition the larger ssd(s) so dp sees more than three but dont add all the ssd's to the ssd optimizer plugin - so one or more are still archive disks so the placement rules will work or use the third ssd as the archive disk and not part of the ssd optimizer etc etc
  5. i have seen this with files that get ms differences in timestamps which get detected by backup software and other software this can happen on a pool rebuild as well as moving files around due to balancers
  6. wow 56 usb drives looking at the wdf_violation its a "driver" related issue if the sas controller works ok without usb drives and vice versa then its probably a conflict between the drivers which is at fault - difficult to say - could be the usb controllers at fault - as i guess you are using more than one type off the motherboard as they usually come with a couple of different types dont think its that you are over doing it - just getting bitten by a bug - as they work when used separately however - the sas card is the way to go if you can add another with a sas expander card or two - rather than using usb there a lots of HP sas expanders on ebay for circa $20 that support 24 drives (and dont need to be plugged into the pci bus - just provide power via a molex connector and the performance will be much better than usb especially as you must have a daisy chain going on.
  7. When a NVME drive is being scanned it does not appear on DP dashboard two drives being scanned only one on DP dashboard - i waited a few mins for it to appear
  8. @gtaus remeasure time now less than 30 mins for 45TB of data
  9. the usb controller will be the bottleneck for 7200 rpm drives especially if they are newer models i tend to shuck my usb drives and a couple of WD Blacks i got recently can do 240MB/s plus once they have room to breath
  10. For me a remeasure is about 80-90mins for 45TB with a similar number of disks - the make up of the files on the disk will affect this greatly - lots of small files = slow i recently got rid of 500k of images - which accounted for a lot of the time above - but have not remeasured since with the HDD DP disks connected to a 9211 the speed is mainly affected by the files sizes/number, controller and HDD speed - cpu is only a very minor player in this
  11. Should be fine - i am in the process of moving/removing/replacing disks at the moment and DP does not care which sata port the disk is connected to only difference is that i dont use drive letters but mount points - so windows getting confused about drive letters is not something i see each disk has a unique poolpart - but as windows had a ghost drive which would have appeared with the same poolpart thats prob what caused DP to choke until you rebooted and it did its checks etc
  12. ok - thats a bit clearer will have a play and report back after some testing
  13. I have a new server that now boots considerably quicker than my old server - in just 5 seconds to the windows login screen as a consequence not all of the 20+ HDD i have - get a chance to mount in that time and DP then does a remeasure etc on every boot which i am trying to stop as its a long unnecessary process. So i would like to change the time that DP waits for the volumes to mount but its a bit confusing as there are three different options with "time" values CoveFs_WaitForVolumesOnMount (Default: "True") - Some volumes arrive late in the boot process so DrivePool will wait for all the volumes to arrive before allowing the pool drive to mount. The delay is adaptive, so it will be different for each system. Turning this off might shorten boot times, but is not recommended because it could expose a partially populated pool drive to system services that start right after booting. CoveFs_WaitForVolumesOnMountMs (Default: "10000") - How long to wait for all the volumes to arrive in MS. CoveFs_WaitForKnownPoolPartsOnMountMs (Default: "10000") - (Newer pools may be aware of it's parts) If a pool is aware of how many pool parts it's made of, at mount time, it will wait for all of those pool parts to arrive before allowing the OS to finish mounting the pool. CoveFs_WaitForPoolsAndPoolPartsAfterMountMs (Default: "10000") - How long the service will wait for all of the detected pools to finish mounting, and then for all of the known pool parts to get recognized on those pools. so 10 seconds is too quick do i need to change all three to a larger value or is the first option the one i need to change - so DP waits x seconds to "see" all the disks before it mounts the poolparts and the service runs up etc as i want the pool to mount as quick as it can without triggering a remeasure on each boot Just after some clarity
×
×
  • Create New...