Re: [Samba] Difference between real and reported disk usage

2012-12-17 Thread Berni CED
Pieter De Wit insync.za.net> writes: > What is the underlying file system, ext4 ? I wonder if it will be worth > trying this on XFS, perhaps it behaves better with newer Windows clients > ? Is this an option for you ? Hi Pieter! Yes i'm using ext4 but since the file transferring was done on a t

Re: [Samba] Difference between real and reported disk usage

2012-12-17 Thread Pieter De Wit
On 17/12/2012 21:07, Berni CED wrote: Berni Ced bernispa.com> writes: For example, i've analized the data copied with strict allocation activated and i've found that Windows Server 2008 make files space allocation in 1024 kiB blocks. Err, i meant Windows Server 2003, not 2008. In the meantime

Re: [Samba] Difference between real and reported disk usage

2012-12-17 Thread Berni CED
Berni CED bernispa.com> writes: > Also, do you know a command that can restore back the 11 MB wasted in the > latter two cases? Obviously, having enough space, i can make a copy of the entire folder hosting the share: cp -a origin-folder dest-folder But my question is if there is a command i can

Re: [Samba] Difference between real and reported disk usage

2012-12-17 Thread Berni CED
Berni Ced bernispa.com> writes: > For example, i've analized the data copied with strict allocation > activated and i've found that Windows Server 2008 make files space > allocation in 1024 kiB blocks. Err, i meant Windows Server 2003, not 2008. In the meantime i've tested that Windows 7 behave

Re: [Samba] Difference between real and reported disk usage

2012-12-16 Thread Berni Ced
Hi Bjoern and thank you for your reply. Bjoern Baumbach sernet.de> writes: > This is by design. > With the strict allocate = yes Samba stops creating sparse files. If you > copy a sparse file with a real size of e.g. 100MB and a > reserved/allocated size of 1GB to a server with "strict allocate =

Re: [Samba] Difference between real and reported disk usage

2012-12-14 Thread Bjoern Baumbach
Dear Cesare, On 12/14/2012 03:51 PM, Berni CED wrote: >> > I've retried commenting out this line and repeating the entire copy with >> > robocopy: now the disk usage corresponds to the one reported by Windows >> > and by >> > "du -s -h --apparent-size". > To summarize, in my case "strict allocate

Re: [Samba] Difference between real and reported disk usage

2012-12-14 Thread Berni CED
Berni CED bernispa.com> writes: > Berni CED bernispa.com> writes: > > strict allocate = yes > > I've retried commenting out this line and repeating the entire copy with > robocopy: now the disk usage corresponds to the one reported by Windows and by > "du -s -h --apparent-size". To summari

Re: [Samba] Difference between real and reported disk usage

2012-12-14 Thread Berni CED
Berni CED bernispa.com> writes: > strict allocate = yes I've retried commenting out this line and repeating the entire copy with robocopy: now the disk usage corresponds to the one reported by Windows and by "du -s -h --apparent-size". Are there any technical reason for that? Should i file a

[Samba] Difference between real and reported disk usage

2012-12-13 Thread Berni CED
(I'm not subscribed to the list: please keep me in CC) I'm copying files between two samba server using robocopy from a third Windows PC and i'm experiencing disk usage that is much higher on the destination PC. The origin samba server is Debian Lenny with samba 3.2.5. The destination samba s