Jump to content

denywinarto

Members
  • Posts

    104
  • Joined

  • Last visited

  • Days Won

    1

Posts posted by denywinarto

  1. I'm building a pxe server where IOPS and read performance is crucial, 

    and i was wondering, how is drivepool's performance options compared to raid 0?

    And which one gives the best result for read-intensive purpose?

     

    My plan is to raid 0  3x 4tb black WD,

    I'm also gonna synchronize them with my main server using syncovery for a periodical backup,

    and the files on this pxe server doesnt get updated very often..

    Thats why i choose raid 0.

     

    I'm leaning towards raid 0 for performance,

    but its non-expandability (have to backup and copy again) and DP scanners' ability interests me..t.

  2. On 7/15/2018 at 2:09 AM, Christopher (Drashna) said:

    Well, The file system repair can/will lock the volume until the repair is finished.  So that may be what you're seeing. 

    But more likely is that the errors on the disk are what is causing the issue here. 

    If you stop the StableBit Scanner service, that should be telling.  If the issue persists, then it's not StableBit Scanner.  But if it goes away, then let us know. 

    It's definitely the scanner cause the moment i uninstalled it the problem was gone.

    Is  there anyway to force file system repair during next boot? I don't see it on the options.

    I have to ignore this for now until the next reboot

     

  3. 16 hours ago, Christopher (Drashna) said:

    No, StableBit Scanner shouldn't lock the drive.... unless it's running a file system repair.  

    Otherwise, if the drive in question has SMART errors, and whatever game/app is having issues, then it's likely that the disk is damaged, and that is what is causing the issues. 

    Yeah some drives were having file system damage or something, that's why i ran file system repair, 

    that's what locked it then?

    I''m rma ing the SMART error drive and replace it with new one

    this means i have to spend some maintenance time to fix the file system damage drives...

    Or.. is it safe to just ignore them?

  4. I just bought and installed Scanner this morning on my ISCSI ccboot server which is used to serve diskless clients.

    Strangely when the scanner is running, or maybe the service is active, ISCSI clients have the drive locked and cannot be accessed.

    There's even 1 client that fails to boot,

    ccboot relies on multiple disks on server and each have their own function

    Is there any known conflict with iscsi programs?

    I uninstalled scanner and things go back to normal, so it's definitely has something to do with it,  and it's the only program i add to the server today

     

    NB ; there were some drives that's giving me file system damaged and SMART warning and i tried to fix them

    I think it may have something to do with fixing these drives

    is it safe to ignore file system damage?

     

    I'm getting this log from ccboot :

     

    Quote

    07:36:27 *** Error *** Start game cache failed. drive: G ram: 0 ssd: 63488 reason: 0x1f A device attached to the system is not functioning.
    07:46:59 *** Error *** Start game cache failed. drive: G ram: 0 ssd: 63488 reason: 0x1f A device attached to the system is not functioning.
    07:50:40 *** ERROR *** file_open \\.\G: error 5 reason: 0x5 Access is denied.
    07:50:40 *** ERROR *** Open \\.\G: failed
    07:50:40 *** ERROR *** file_open \\.\G: error 5 reason: 0x5 Access is denied.
    07:50:40 *** ERROR *** Open \\.\G: failed
    07:50:40 *** ERROR *** file_open \\.\G: error 5 reason: 0x5 Access is denied.
    07:50:40 *** ERROR *** Open \\.\G: failed
    07:50:40 *** ERROR *** file_open \\.\G: error 5 reason: 0x5 Access is denied.
    07:50:40 *** ERROR *** Open \\.\G: failed
    07:50:40 *** ERROR *** file_open \\.\G: error 5 reason: 0x5 Access is denied.
    07:50:40 *** ERROR *** Open \\.\G: failed
    07:53:57 *** ERROR *** file_open \\.\F: error 5 reason: 0x5 Access is denied.
    07:53:57 *** ERROR *** Open \\.\F: failed
    07:53:57 *** ERROR *** 192.168.0.59:1024 disconnect: InitDisk (D:\MSI_May_2018.VHD;\\.\F:;\\.\G:;)failed
    07:53:57 *** ERROR *** file_open \\.\F: error 5 reason: 0x5 Access is denied.
    07:53:57 *** ERROR *** Open \\.\F: failed
    07:53:57 *** ERROR *** 192.168.0.59:1025 disconnect: InitDisk (D:\MSI_May_2018.VHD;\\.\F:;\\.\G:;)failed
    07:53:57 *** ERROR *** file_open \\.\F: error 5 reason: 0x5 Access is denied.
    07:53:57 *** ERROR *** Open \\.\F: failed
    07:53:57 *** ERROR *** 192.168.0.59:1026 disconnect: InitDisk (D:\MSI_May_2018.VHD;\\.\F:;\\.\G:;)failed
    07:53:57 *** ERROR *** file_open \\.\F: error 5 reason: 0x5 Access is denied.
    07:53:57 *** ERROR *** Open \\.\F: failed
    07:53:57 *** ERROR *** 192.168.0.59:1027 disconnect: InitDisk (D:\MSI_May_2018.VHD;\\.\F:;\\.\G:;)failed
    07:53:57 *** ERROR *** file_open \\.\F: error 5 reason: 0x5 Access is denied.
    07:53:57 *** ERROR *** Open \\.\F: failed
    07:53:57 *** ERROR *** 192.168.0.59:1028 disconnect: InitDisk (D:\MSI_May_2018.VHD;\\.\F:;\\.\G:;)failed
    07:53:57 *** ERROR *** file_open \\.\F: error 5 reason: 0x5 Access is denied.
    07:53:57 *** ERROR *** Open \\.\F: failed
    07:53:57 *** ERROR *** 192.168.0.59:1029 disconnect: InitDisk (D:\MSI_May_2018.VHD;\\.\F:;\\.\G:;)failed
    .

    .

    .

    .

    .
    08:00:50 *** ERROR *** 192.168.0.76:1034 disconnect: InitDisk (D:\MSI_May_2018.VHD;\\.\F:;\\.\G:;)failed
    08:01:04 *** ERROR *** file_open \\.\F: error 5 reason: 0x5 Access is denied.
    08:01:04 *** ERROR *** Open \\.\F: failed
    08:01:04 *** ERROR *** 192.168.0.61:1024 disconnect: InitDisk (D:\MSI_May_2018.VHD;\\.\F:;\\.\G:;)failed
    08:01:04 *** ERROR *** file_open \\.\F: error 5 reason: 0x5 Access is denied.
    08:01:04 *** ERROR *** Open \\.\F: failed
    08:01:04 *** ERROR *** 192.168.0.61:1025 disconnect: InitDisk (D:\MSI_May_2018.VHD;\\.\F:;\\.\G:;)failed
    08:01:04 *** ERROR *** file_open \\.\F: error 5 reason: 0x5 Access is denied.
    08:01:04 *** ERROR *** Open \\.\F: failed
    08:01:04 *** ERROR *** 192.168.0.61:1026 disconnect: InitDisk (D:\MSI_May_2018.VHD;\\.\F:;\\.\G:;)failed
    08:01:04 *** ERROR *** file_open \\.\F: error 5 reason: 0x5 Access is denied.
    08:01:04 *** ERROR *** Open \\.\F: failed
    08:01:04 *** ERROR *** 192.168.0.61:1027 disconnect: InitDisk (D:\MSI_May_2018.VHD;\\.\F:;\\.\G:;)failed
    08:01:05 *** ERROR *** file_open \\.\F: error 5 reason: 0x5 Access is denied.
    08:01:05 *** ERROR *** Open \\.\F: failed
    08:01:05 *** ERROR *** 192.168.0.61:1028 disconnect: InitDisk (D:\MSI_May_2018.VHD;\\.\F:;\\.\G:;)failed


     

  5. One of my drives is showing smart warning, and i'm gonna RMA it and now i'm buying a new drive,

    just wanna make sure this is correct procedure,

    1. Insert new drive to rack

    2. Add new drive to pool

    3. Remove bad drive pool from DP

    4. Remove bad drive from rack

    Is that it? did i miss anything?

  6. 32 minutes ago, Christopher (Drashna) said:

    Ahhh, yeah, Good ol' Windows.

    This is a known issue, actually.  Windows will do this, when "adding" new drives.  It auto-assigns letters, and may bump them up.  

    So a higher letter is always a good idea (I personally use "S:\" since it VERY far along, and should never be an issue). 

    And mounting the pooled disks to NTFS folders (http://wiki.covecube.com/StableBit_DrivePool_Q4822624) is a good way to help avoid this. Especially, as I manually partition and format disks. 

    So you're saying you use S for drivepool then your drives are in NTFS folders?

    Quote

    For simplicity, you should use a single folder to hold all the folders. Such as "C:\DrivePool\". Then create folders inside of "C:\DrivePool\" for each drive you want to mount.

    This is what i meant, then i guess the best way to do this is to have drivepool mapped to Z drive,

    then the rest of the drives mounted in NTFS folders inside C:\DrivePool\?

  7. On 1/26/2018 at 11:30 PM, Christopher (Drashna) said:

    Ah.  Mount point's don't really matter.  To DrivePool, or Windows really. 

    So unless you're intending to mount the drive on the pool's path, there should be no issue. 

    Hmm something strange just happened, my server reboots (not sure why i think Teamviewer crashed it),

    before reboot Drivepool was using N letter

    after reboot Drivepool changes it drive letter to O, and the new drive is suddenly mounted to N.

    My shares were gone, so i had to reorder the drive letters

    Kinda makes me wonder whats the best way to use NTFS mount points + drivepool and prevent something like this?

    Or maybe at least can i lock drivepool's drive letter somehow?

  8. So i mount my new hdd with mount points, and i put it on C which is reserved for OS.. few questions..

    1. Can i move the mount point file to drivepool partition?

    2. what if my C got corrupted? can i restore the mount point somehow?

  9. 3 hours ago, Christopher (Drashna) said:

    Well, copying from one drive to another (or network location) can cause the permissions to get reset.  So, from C: to Z: would cause that.

    As for the virus thing, a second/closer look, it looks like ESET is losing it's mind over the 'desktop.ini' files.  I would check one, to see what it looks like (eg, calling an EXE or some such).  Especially, as it's mentioned "AutoRun" worms. 

    desktop.ini is indeed a virus, other AV is reporting the same..

    Earlier this morning some of the folders changed its permission again.. and the shortcut virus is back..

    e.g : correct permission should be : Everyone, admin and client

    the infected  folders have extra permission from system, authenticated users , and users

    not sure if its cause i forgot to fix the permssion on those folders, or something else chenged them, but i'm pretty sure i made the parent folder inherited the correct permissions

    It's not as much folders as the last time though.. hopefully it's just me forgetting to change the permssion,  will monitor again..

  10. Now i know whats probably causing it.

    See like i said I'm sharing it with local diskless clients, and some of those clients have personal data that is saved even after reboot.

    Now this personal data is probably infected with virus.

    So i test one of the infected folders from client side.

    And surprisingly, the client can create a text document inside that folder!

    But deleting the text file requires admin account and was halted immediately, 

    and strangely after a while text file is gone..

    I double check the security and it's definitely shared with a standard account, and it doesnt have write permission.

    I tested the other folder that's not infected and it refuses to write anything prompting "destination folder access denied"

    So the writable folder is causing security hole...

    is this related with drivepool somehow?

     

    Edit, attached is the permission difference between writable folder (veep) and read only folder (vinyl)

    It seems like there's extra users for the writable folders..

    I could probably remove those users, but problem is some new folders also have this writable behavior..

    write.jpg

     

    Edit : i changed the whole pool permission inheritance and seems like the shortcut virus arent back.. yet 

  11. Ok i just scanned my drives with nod32 and found 300 viruses.

    But the next day it found them again, through the poolpart instead the normal path to the drives..

    So i was wondering, could one of drivepool features actually bring back the viruses? i'm using balancing feature btw

    virus.PNG

  12. Asking this cause i was hit by a 2010 shortcut virus... My server is shared for public usage to a number of clients so the source might be from infected thumbdrives.. but i cant understand how can it affects read-only smb shares.. does drivepool alter smb share security or is this a known win server 2012 r2 security hole?

    Hopefully its the later,

    I have followed a guude to clean it.  But its still kinda worries me if it happens again.. anyone know a fix or patch?

    VtGX1J2.jpg

  13. Yeah can confirm it's safe.

    And i Finally figured out the cause... my new drive has shorter length than the older ones.. no wonder it didn't detect it cause it's not even plugged in! never new they had different size though.. i moved the caddy screw higher and now it's working :)  

  14. 2 hours ago, Christopher (Drashna) said:

    That's odd.  And makes me wonder if the SAS expander on the enclosure isn't defective in some way. 

    That said, no, there isn't a tool for this.  There are a number of file systems that may be similar though (such as MergerFS, I think). 

     

    As for safe to switch between the two, I'm not really sure, honestly.  You shouldn't, in theory. And worst case, you can reset the permissions in Windows by doing this:
    http://wiki.covecube.com/StableBit_DrivePool_Q5510455

    "You shouldn't" meaning it's not safe?

    yeah tried almost everything, tested yesterday without drives plugged in, turns out there's no zoning

    Card is made in chinese so i'm refunding it and getting the US one (albeit twice more expensive) 

    Wanted to debug it further but im held back by windows' tool limitation..

     

     

  15. Hi, i need to access my jbod under linux to check zoning possibility.. cause strangely my hgst 4u60 is limiting the drives to 11. 

    Ive tried lsi 9300 8e and dell sc8000 and both cant detect 12th drive, a 8tb red wd. But i can move around drives in my jbod as long as its under 11. 

    Unfortunately i cant find any equivalent alternative for smp_utils under windows.

    So im gonna have to boot up ubuntu and access the jbod rack.. question is, is this safe for drivepool once i get back to windows? Cause if i recall i had permission issues when going back and forth with ubuntu and windows.. cant risk my data for this.  Thanks

     

     

  16. Hi, one of my drives is almost full, but there's almost 1TB left in my pool

    How exactly do i make the balancing file work? I tried  volume equalization but it doesn't seem to balance the drives..

    this is after vol equalization :

    balance.thumb.PNG.6d309e052072a4ce513aa17a8d0cd5b0.PNG

    I'm still getting full disk warning and failed torrent download 

     

     

     

     

  17. Sorry for bumping an old thread, but i think i'm experiencing this as well chris.

    Was looking for a movie earlier this morning.

    But I couldn't find it despite kodi listing the movie, and the utorrent has the file downloaded 

    So i redownloaded it again in utorrent

    Strangely when i try to move the file to the exact same path, the movie reappears!

    .. and prompt me if i want to replace the same file or not...

     

    Drivepool is the latest version 2.20 742 beta

  18. You can mount the drives to a folder in windows and remove the drive letter for you data drives

     

    i have 33 drives in one machine but only four drive letters in use

     

    I'm accessing the library from another machine using Kodi,

    I'd thought Kodi can only see driveletters, 

    i read in their forum it could also read UNC paths, guess gotta give it a try first

  19. Could you point me to which part it stucks like you did last time? it's almost finished measuring, like 98-99%

    as of removing hdd, last time i did that the data was scattered or something, is there a removal hdd wiki somewhere?

    And i can't have it removed for too long thats why i prefer "juggling" the data like this

×
×
  • Create New...