Hi,
we are using Samba 2.2.5 in a somewhat larger configuration: a big NFS file
server and a separate box acting as a NFS/SMB gateway. Usually there are
about 200-250 connected clients. The box is a Sun Netra T1 AC200 with
Solaris 8.
Sometimes we get the following (loglevel 1):
[2002/07/01 11:48:50, 1] smbd/service.c:make_connection(615)
si17062 (10.2.202.88) connect to service mrd8si as user mrd8si (uid=2674,
gid=155) (pid 23348)
[2002/07/01 11:50:31, 0] lib/util_sock.c:read_socket_with_timeout(298)
read_socket_with_timeout: timeout read. read error = Connection reset by
peer.
[2002/07/01 11:50:31, 1] smbd/password.c:server_cryptkey(1083)
SI21930 rejected the negprot
[2002/07/01 11:50:31, 1] smbd/password.c:server_validate(1117)
password server is not connected
[2002/07/01 11:50:31, 1] smbd/password.c:pass_check_smb(555)
Couldn't find user 'mrd8si' in passdb.
[2002/07/01 11:50:31, 1] smbd/password.c:pass_check_smb(555)
Couldn't find user 'mrd8si' in passdb.
[2002/07/01 11:50:31, 1] smbd/reply.c:reply_sesssetup_and_X(988)
Rejecting user 'mrd8si': authentication failed
[2002/07/01 11:50:31, 1] smbd/password.c:authorise_login(737)
authorise_login: refusing user with no session setup
Last two lines repeated about a dozen times. The Windows client show
something like "Drive N: is already connected to \\server\share".
I'm sorry
I don't have the exact error message. The user has to disconnect and
reconnect his network drive. This is very annoying.
I'm using "security = server". Domain Controller is a W2k server,
clients
are mainly W2k.
I tried a patch to serialize authentication which is a backport from Head,
but it didn't solve my problem.
Next shot was to increase "deadtime" from 60 to 120, but I guess it
won't
help either, because the problem occured only a few minutes after last
successful connection.
Global section in smb.conf looks like this:
[global]
workgroup = DE
netbios name = SI-SAMBA
server string = 10.3.3.49;QI/EES3 Mr;SAMBA %v
interfaces = eri0:1 eri0
security = SERVER
encrypt passwords = Yes
password server = si21931,si8832,si6245,hsm1nt1
username map = /net/samba/fs1/samba/lib/users.map
log file = /net/samba/fs1/samba/var/%m.log
max log size = 300
time server = Yes
change notify timeout = 10
deadtime = 120
kernel oplocks = No
oplocks = No
level2 oplocks = No
max open files = 800
load printers = No
log level = 1
local master = No
wins server = 10.4.4.62
force directory mode = 0700
dos filetimes = Yes
Anybody has a Idea how to solve this?
Mit freundlichen Gr??en / Kind regards / Saludos cordiales
Daniel M?ller
Robert Bosch GmbH
SC Operation, QI/EES3
Tel: 0711/811-34340 FAX: -266689
To me it looks like its time out on connections,
do you think system has enough resources
any user getting semaphore time out while connecting
-----Original Message-----
From: Moeller Daniel (QI/EES3) * [mailto:Daniel.Moeller@de.bosch.com]
Sent: Wednesday, July 03, 2002 9:21 AM
To: samba@lists.samba.org
Subject: [Samba] connection problem to W2k DC
Hi,
we are using Samba 2.2.5 in a somewhat larger configuration: a big NFS file
server and a separate box acting as a NFS/SMB gateway. Usually there are
about 200-250 connected clients. The box is a Sun Netra T1 AC200 with
Solaris 8.
Sometimes we get the following (loglevel 1):
[2002/07/01 11:48:50, 1] smbd/service.c:make_connection(615)
si17062 (10.2.202.88) connect to service mrd8si as user mrd8si (uid=2674,
gid=155) (pid 23348)
[2002/07/01 11:50:31, 0] lib/util_sock.c:read_socket_with_timeout(298)
read_socket_with_timeout: timeout read. read error = Connection reset by
peer.
[2002/07/01 11:50:31, 1] smbd/password.c:server_cryptkey(1083)
SI21930 rejected the negprot
[2002/07/01 11:50:31, 1] smbd/password.c:server_validate(1117)
password server is not connected
[2002/07/01 11:50:31, 1] smbd/password.c:pass_check_smb(555)
Couldn't find user 'mrd8si' in passdb.
[2002/07/01 11:50:31, 1] smbd/password.c:pass_check_smb(555)
Couldn't find user 'mrd8si' in passdb.
[2002/07/01 11:50:31, 1] smbd/reply.c:reply_sesssetup_and_X(988)
Rejecting user 'mrd8si': authentication failed
[2002/07/01 11:50:31, 1] smbd/password.c:authorise_login(737)
authorise_login: refusing user with no session setup
Last two lines repeated about a dozen times. The Windows client show
something like "Drive N: is already connected to \\server\share".
I'm sorry
I don't have the exact error message. The user has to disconnect and
reconnect his network drive. This is very annoying.
I'm using "security = server". Domain Controller is a W2k server,
clients
are mainly W2k.
I tried a patch to serialize authentication which is a backport from Head,
but it didn't solve my problem.
Next shot was to increase "deadtime" from 60 to 120, but I guess it
won't
help either, because the problem occured only a few minutes after last
successful connection.
Global section in smb.conf looks like this:
[global]
workgroup = DE
netbios name = SI-SAMBA
server string = 10.3.3.49;QI/EES3 Mr;SAMBA %v
interfaces = eri0:1 eri0
security = SERVER
encrypt passwords = Yes
password server = si21931,si8832,si6245,hsm1nt1
username map = /net/samba/fs1/samba/lib/users.map
log file = /net/samba/fs1/samba/var/%m.log
max log size = 300
time server = Yes
change notify timeout = 10
deadtime = 120
kernel oplocks = No
oplocks = No
level2 oplocks = No
max open files = 800
load printers = No
log level = 1
local master = No
wins server = 10.4.4.62
force directory mode = 0700
dos filetimes = Yes
Anybody has a Idea how to solve this?
Mit freundlichen Gr??en / Kind regards / Saludos cordiales
Daniel M?ller
Robert Bosch GmbH
SC Operation, QI/EES3
Tel: 0711/811-34340 FAX: -266689
--
To unsubscribe from this list go to the following URL and read the
instructions: http://lists.samba.org/mailman/listinfo/samba
Reasonably Related Threads
- authentification problems
- Allow insecure wide links = yes, wide links =yes; but I still can't "see" files from links to NFS mounts using 3.6.15, after upgrading from 2.2.8a
- AW: authentification problems
- Newbie question - Wrong Way to map a share?
- Samba and OS X 10.4.2 - vfs quota inheritance not correct