[Samba] Horrible write performance from XP to Samba

How about Windows client to Windows client speeds?

I'm getting the feeling the problem might be the speed of the hard
drives on your clients.

Sebastian Held wrote:
> Hi,
>
> Am Donnerstag, 5. Oktober 2006 14:56 schrieb Peter Daum:
>
>> maybe some people would be adventurous enough to just try it out:
>>
>
> My results (only 1 measurement taken)
> WinXP64 -> Samba-3.0.23b/SuSE-Linux-9.2: 2:08 min => 8192 kiB/s (smbd
> @18% CPU)
> other direction: 2:21 min => 7437 kiB/s (smbd @10% CPU)
>
> Samba -> Win2k: 4:52 min => 3591 kiB/s (smbd @5% CPU)
> Win2k -> Samba: 2:22 min => 7384 kiB/s (smbd @15% CPU)
>
> The windows machines are not the same.
>
> compare to nfs performance:
> Samba machine -> other nfs machine: 1:58 min => 8886 kiB/s
> other direction: 1:55 min => 9118 kiB/s
>
>
> br,
> Sebastian
>
>

--
To unsubscribe from this list go to the following URL and read the
instructions: https://lists.samba.org/mailman/listinfo/samba

Have you shut down all other processes including anti-virus to make sure
there isn't something else causing a wide variance?

Sebastian Held wrote:
> Am Donnerstag, 5. Oktober 2006 16:41 schrieben Sie:
>
>> How about Windows client to Windows client speeds?
>>
>> I'm getting the feeling the problem might be the speed of the hard
>> drives on your clients.
>>
>
> For me and my budget, 8 MB/s must be enough.
>
> But here are the requested values:
>
> Copying initiated from WinXP64:
> WinXP64 -> Win2k: 3:10 min
> Win2k -> WinXP64: 2:08 min
>
> Copying initiated from Win2k:
> WinXP64 -> Win2k: 5:28 min
> Win2k -> WinXP64: 1:53 min
>
> I don't know, why these values differ that much, if the copying machine is
> changed...
>
> br, Sebastian
>

--
To unsubscribe from this list go to the following URL and read the
instructions: https://lists.samba.org/mailman/listinfo/samba

On Thu, Oct 05, 2006 at 03:55:00PM +0200, Sebastian Held wrote:
> Hi,
>
> Am Donnerstag, 5. Oktober 2006 14:56 schrieb Peter Daum:
> > maybe some people would be adventurous enough to just try it out:
>
> My results (only 1 measurement taken)
> WinXP64 -> Samba-3.0.23b/SuSE-Linux-9.2: 2:08 min => 8192 kiB/s (smbd
> @18% CPU)
> other direction: 2:21 min => 7437 kiB/s (smbd @10% CPU)
>
> Samba -> Win2k: 4:52 min => 3591 kiB/s (smbd @5% CPU)
> Win2k -> Samba: 2:22 min => 7384 kiB/s (smbd @15% CPU)
>
> The windows machines are not the same.
>
> compare to nfs performance:
> Samba machine -> other nfs machine: 1:58 min => 8886 kiB/s
> other direction: 1:55 min => 9118 kiB/s

Unless the "other nfs machine" is also windows this is
an apples to oranges comparison. If you want an apples
to apples comparison compare using cifsfs on the linux
client as your cifs client, not windows.

Jeremy.
--
To unsubscribe from this list go to the following URL and read the
instructions: https://lists.samba.org/mailman/listinfo/samba

Sebastian Held wrote:

> But the only thing I wanted to show is, that my WinXP has no problems writing
> to a Samba share. Performance is good (at least in my opinion).

.... so obviously the problems that I encounter do not occur in all
cases - of course it would be good to know what makes the
difference. I could reproduce the issue on 4 different Samba
servers (with totally different configuration) and with 5
differently configured Windows-XP clients (all XP professional+
SP2) and with different network switches in between ?!

Any clues?

--
To unsubscribe from this list go to the following URL and read the
instructions: https://lists.samba.org/mailman/listinfo/samba

I am still desperately trying to figure out why I get such a horrible
performance trying to copy files from XP to Samba (below is my
original post to illustrate the context. Meanwhile, I found a
pretty remarkable pattern in the network traces.

Generally, the packets send from the XP machine look like this:

01: WriteAndXRequest 60 kBytes (1392 bytes payload)
02: 1460 bytes
... (usual time delta between 2 packets < 0.0001 secs)
42: 1460 bytes
< < < < delta: 0.05 - 0.3 secs > > > >
43: last 188 bytes
... WriteAndXResponse from Samba machine, da capo

It seems like what is slowing the transfer down so dramatically, is
the long gap before the last 188 bytes are sent. I can't see anything
reasonable the XP machine could be waiting for - it already got the
ack for the last preceding packet.

To emphasize again, this is not a general networking problem between
the 2 machines; it only (at leas AFAIK) occurs when copying large
files from Windows XP to Samba shares (and at that occasion I could
reproduce it with all the Samba servers and XP clients that I checked
so far)

Any help would be greatly appreciated.

Regards,
Peter Daum

> I noted an extremely poor performance when copying big files from
> a windows xp client to a samba share. The exact version of samba
> does not seem to matter: I tried several different samba servers
> with versions between 3.014 and 3.0.23b running on Linux 2.4.32
> and 2.6.17 (machines and network otherwise idle, clients connected
> via fast ethernet, servers via Gbit; network performance in both
> directions around 95 Mbit/s). I made several tests copying a 1GB
> file with Windows 98 and Windows XP clients. Reading the file from
> the server takes predictably around 105 seconds (~9.75 MB/s).
>
> Writing to the server takes only slightly longer on Win98 (130
> seconds, ~8 MB/s) while the same takes approximately 45 minutes
> from a XP client (I don't know whether this matters, I noted that
> on the XP write test, the directory listing on the server
> immediately shows a file with the final size - obviously a sparse
> file, repeatedly invoking du shows the gradually increasing actual
> size).
>
> I wrote a little test program that just writes data to a file and
> shows the throughput; the transfer rates I get that way are pretty
> reasonable, so it is not a general problem but something that only
> occurs on specific operations like copying.
>
> Tracing the network traffic also didn't tell me what the problem
> might be: XP uses for copying as well as for other write
> operations WriteAndXRequest, the only peculiarity I noticed is the
> slightly exotic block size of 61440 bytes per request when copying
> (which also doesn't seem to be the problem - Win98 uses the same
> block size with WriteRaw)
>
> Has anybody else made similar experiences? (Since I could see this
> issue with differently configured servers/clients, it should not
> be just my personal problem. Of course in most settings where the
> data usually goes mostly from the server to the client it is not
> obvious)
>
> Any ideas what's going on and what to do about it?
>

--
To unsubscribe from this list go to the following URL and read the
instructions: https://lists.samba.org/mailman/listinfo/samba

I am also having performance problems writing to a Samba share (see previous
post about tcpdump) and I have seen a strange phenomenon: if I start tcpdump
on the Samba machine, performance increases 20 times. It would be
interesting to know if you also experience the same, just start tcpdump and
redirect output to /dev/null while/before copying files.

Regards,
Daniel Lindgren

2006/10/9, Peter Daum <gator_ml.DeleteThis@yahoo.de>:
>
> I am still desperately trying to figure out why I get such a horrible
> performance trying to copy files from XP to Samba (below is my
> original post to illustrate the context. Meanwhile, I found a
> pretty remarkable pattern in the network traces.
>
> Generally, the packets send from the XP machine look like this:
>
> 01: WriteAndXRequest 60 kBytes (1392 bytes payload)
> 02: 1460 bytes
> ... (usual time delta between 2 packets < 0.0001 secs)
> 42: 1460 bytes
> < < < < delta: 0.05 - 0.3 secs > > > >
> 43: last 188 bytes
> .. WriteAndXResponse from Samba machine, da capo
>
> It seems like what is slowing the transfer down so dramatically, is
> the long gap before the last 188 bytes are sent. I can't see anything
> reasonable the XP machine could be waiting for - it already got the
> ack for the last preceding packet.
>
> To emphasize again, this is not a general networking problem between
> the 2 machines; it only (at leas AFAIK) occurs when copying large
> files from Windows XP to Samba shares (and at that occasion I could
> reproduce it with all the Samba servers and XP clients that I checked
> so far)
>
> Any help would be greatly appreciated.
>
> Regards,
> Peter Daum
>
> > I noted an extremely poor performance when copying big files from
> > a windows xp client to a samba share. The exact version of samba
> > does not seem to matter: I tried several different samba servers
> > with versions between 3.014 and 3.0.23b running on Linux 2.4.32
> > and 2.6.17 (machines and network otherwise idle, clients connected
> > via fast ethernet, servers via Gbit; network performance in both
> > directions around 95 Mbit/s). I made several tests copying a 1GB
> > file with Windows 98 and Windows XP clients. Reading the file from
> > the server takes predictably around 105 seconds (~9.75 MB/s).
> >
> > Writing to the server takes only slightly longer on Win98 (130
> > seconds, ~8 MB/s) while the same takes approximately 45 minutes
> > from a XP client (I don't know whether this matters, I noted that
> > on the XP write test, the directory listing on the server
> > immediately shows a file with the final size - obviously a sparse
> > file, repeatedly invoking du shows the gradually increasing actual
> > size).
> >
> > I wrote a little test program that just writes data to a file and
> > shows the throughput; the transfer rates I get that way are pretty
> > reasonable, so it is not a general problem but something that only
> > occurs on specific operations like copying.
> >
> > Tracing the network traffic also didn't tell me what the problem
> > might be: XP uses for copying as well as for other write
> > operations WriteAndXRequest, the only peculiarity I noticed is the
> > slightly exotic block size of 61440 bytes per request when copying
> > (which also doesn't seem to be the problem - Win98 uses the same
> > block size with WriteRaw)
> >
> > Has anybody else made similar experiences? (Since I could see this
> > issue with differently configured servers/clients, it should not
> > be just my personal problem. Of course in most settings where the
> > data usually goes mostly from the server to the client it is not
> > obvious)
> >
> > Any ideas what's going on and what to do about it?
> >
>
> --
> To unsubscribe from this list go to the following URL and read the
> instructions: https://lists.samba.org/mailman/listinfo/samba
>
--
To unsubscribe from this list go to the following URL and read the
instructions: https://lists.samba.org/mailman/listinfo/samba

Peter Daum schrieb:
> Any clues?
>
How about posting one of your smb.conf?

Greetings

Martin Sundmacher

--
To unsubscribe from this list go to the following URL and read the
instructions: https://lists.samba.org/mailman/listinfo/samba

Hi @all,

I don't know why, but my PDC Stopps serving the netlogon service over the
weekend.

\\HUNTER\netlogon is avaliable and readeable!

Here's output of nbtstat -ac HUNTER

Name Typ Status
---------------------------------------------
HUNTER <00> EINDEUTIG Registriert
HUNTER <03> EINDEUTIG Registriert
HUNTER <20> EINDEUTIG Registriert
..__MSBROWSE__.<01> GRUPPE Registriert
SNEAKER <1D> EINDEUTIG Registriert
SNEAKER <1B> EINDEUTIG Registriert
SNEAKER <1C> GRUPPE Registriert
SNEAKER <1E> GRUPPE Registriert
SNEAKER <00> GRUPPE Registriert

Smb.conf

[global]
interfaces = lo eth0 eth1 eth2
bind interfaces only = Yes
name resolve order = wins bcast lmhosts host
unix charset = ISO8859-1
display charset = ISO8859-1
workgroup = SNEAKER
netbios name = HUNTER
admin users = @"Domain Admins"
guest account = nobody
server string = SoundServer %v
security = user
encrypt passwords = Yes
log level = 2 vfs:2
log file = /var/log/samba/%U.%m.log
syslog = 0
max log size = 100000
domain logons = Yes
os level = 255
preferred master = Yes
domain master = Yes
local master = Yes
wins support = Yes
wins proxy = Yes
dns proxy = Yes
time server = Yes
#ldap##
passdb backend = ldapsam:"ldap://127.0.0.1/"
ldap admin dn = cn=Manager,dc=radiogong,dc=intern
ldap suffix = dc=radiogong,dc=intern
ldap group suffix = ou=Groups
ldap user suffix = ou=Users
ldap machine suffix = ou=Computers
ldap idmap suffix = ou=Users
ldap ssl = no
ldap delete dn = Yes
add user script = /usr/local/sbin/smbldap-useradd -a -P "%u"
delete user script = /usr/local/sbin/smbldap-userdel -r "%u"; rm -r
/home/"%u"; rm -r /opt/profiles/"%u"
add machine script = /usr/local/sbin/smbldap-useradd -w "%u"
add group script = /usr/local/sbin/smbldap-groupadd "%g"
delete group script = /usr/local/sbin/smbldap-groupdel "%g"
add user to group script = /usr/local/sbin/smbldap-groupmod -m "%u" "%g"
delete user from group script = /usr/local/sbin/smbldap-groupmod -x "%u"
"%g"
set primary group script = /usr/local/sbin/smbldap-usermod -g "%g" "%u"
ldap passwd sync = Yes
utmp = Yes
idmap uid = 1000-20000
idmap gid = 1000-20000
idmap backend = ldap:ldap://127.0.0.1/
shutdown script = /sbin/shutdown
abort shutdown script = /sbin/shutdown -c
winbind enum users = yes
winbind enum groups = yes
winbind use default domain = yes
winbind uid = 1000-20000
winbind gid = 1000-20000
winbind cache time = 10
nt acl support = yes
kernel oplocks = yes
enable privileges = Yes
template shell = /bin/false
logon script = logon.bat
logon path =
logon home =


DNS is working fine, also WinS!

Do you have any clues?

Thanx in advance

Sascha

--
To unsubscribe from this list go to the following URL and read the
instructions: https://lists.samba.org/mailman/listinfo/samba

Martin Sundmacher wrote:
> Peter Daum schrieb:
>> Any clues?
>>
> How about posting one of your smb.conf?

.... not particularly helpful, because the phenomenon is obviously
not dependent on any particularly configuration and also occurs
with a smb.conf that only defines the path for one share and
keeps all the Samba default settings.

Regards,
Peter Daum

--
To unsubscribe from this list go to the following URL and read the
instructions: https://lists.samba.org/mailman/listinfo/samba

Unfortunately, running a packet sniffer on the server does not make
things any better in my case (otherwise it would be at least an
easy workaround

Regards,
Peter Daum

Daniel Lindgren wrote:
> I am also having performance problems writing to a Samba share (see
> previous
> post about tcpdump) and I have seen a strange phenomenon: if I start
> tcpdump
> on the Samba machine, performance increases 20 times. It would be
> interesting to know if you also experience the same, just start tcpdump and
> redirect output to /dev/null while/before copying files.
>
> Regards,
> Daniel Lindgren
>
> 2006/10/9, Peter Daum <gator_ml.TakeThisOut@yahoo.de>:
>>
>> I am still desperately trying to figure out why I get such a horrible
>> performance trying to copy files from XP to Samba (below is my
>> original post to illustrate the context. Meanwhile, I found a
>> pretty remarkable pattern in the network traces.
>>
>> Generally, the packets send from the XP machine look like this:
>>
>> 01: WriteAndXRequest 60 kBytes (1392 bytes payload)
>> 02: 1460 bytes
>> ... (usual time delta between 2 packets < 0.0001 secs)
>> 42: 1460 bytes
>> < < < < delta: 0.05 - 0.3 secs > > > >
>> 43: last 188 bytes
>> .. WriteAndXResponse from Samba machine, da capo
>>
>> It seems like what is slowing the transfer down so dramatically, is
>> the long gap before the last 188 bytes are sent. I can't see anything
>> reasonable the XP machine could be waiting for - it already got the
>> ack for the last preceding packet.
>>
>> To emphasize again, this is not a general networking problem between
>> the 2 machines; it only (at leas AFAIK) occurs when copying large
>> files from Windows XP to Samba shares (and at that occasion I could
>> reproduce it with all the Samba servers and XP clients that I checked
>> so far)
>>
>> Any help would be greatly appreciated.
>>
>> Regards,
>> Peter Daum
>>
>> > I noted an extremely poor performance when copying big files from
>> > a windows xp client to a samba share. The exact version of samba
>> > does not seem to matter: I tried several different samba servers
>> > with versions between 3.014 and 3.0.23b running on Linux 2.4.32
>> > and 2.6.17 (machines and network otherwise idle, clients connected
>> > via fast ethernet, servers via Gbit; network performance in both
>> > directions around 95 Mbit/s). I made several tests copying a 1GB
>> > file with Windows 98 and Windows XP clients. Reading the file from
>> > the server takes predictably around 105 seconds (~9.75 MB/s).
>> >
>> > Writing to the server takes only slightly longer on Win98 (130
>> > seconds, ~8 MB/s) while the same takes approximately 45 minutes
>> > from a XP client (I don't know whether this matters, I noted that
>> > on the XP write test, the directory listing on the server
>> > immediately shows a file with the final size - obviously a sparse
>> > file, repeatedly invoking du shows the gradually increasing actual
>> > size).
>> >
>> > I wrote a little test program that just writes data to a file and
>> > shows the throughput; the transfer rates I get that way are pretty
>> > reasonable, so it is not a general problem but something that only
>> > occurs on specific operations like copying.
>> >
>> > Tracing the network traffic also didn't tell me what the problem
>> > might be: XP uses for copying as well as for other write
>> > operations WriteAndXRequest, the only peculiarity I noticed is the
>> > slightly exotic block size of 61440 bytes per request when copying
>> > (which also doesn't seem to be the problem - Win98 uses the same
>> > block size with WriteRaw)
>> >
>> > Has anybody else made similar experiences? (Since I could see this
>> > issue with differently configured servers/clients, it should not
>> > be just my personal problem. Of course in most settings where the
>> > data usually goes mostly from the server to the client it is not
>> > obvious)
>> >
>> > Any ideas what's going on and what to do about it?
>> >
>>
>> --
>> To unsubscribe from this list go to the following URL and read the
>> instructions: https://lists.samba.org/mailman/listinfo/samba
>>

--
To unsubscribe from this list go to the following URL and read the
instructions: https://lists.samba.org/mailman/listinfo/samba