I'm not sure if this is an os-x implementation bug or a samba bug, but it worked before and os-x didn't change... I'm running stable on x86 (athlon xp-m 1.66) running samba. I'm running 2 types of clients: an os-x (tiger) ibook and a windows xp desktop. The iBook can connect to the samba shares via "go"->"Map network drive", but can't browse to shares via the network browser (finder -> network -> workgroup name -> server name -> connect). Shortly after connect, it beachballs (hangs indefinitely, until finder is relaunched). per-machine log (at log level 0) shows: [2005/08/08 01:17:49, 0] rpc_parse/parse_prs.c:prs_mem_get(537) prs_mem_get: reading data of size 2 would overrun buffer. [2005/08/08 01:17:49, 0] rpc_server/srv_pipe.c:api_pipe_bind_req(919) api_pipe_bind_req: unable to unmarshall RPC_HDR_RB struct. Windows XP clients can connect fine, every way I can think to connect to it. os-x client can still connect via "go->connect to server" -- as long as we stay away from the "browse" interface (in other words, already know the share names). Reproducible: Always Steps to Reproduce: 1.emerge samba 3.0.14a-r2 2.start it 3.on os-x machine, hit network->workgroup->servername and click "connect". Actual Results: the os-x client hangs until it's killed and relaunched. The server shows the errors in the details section. Expected Results: shouldn't have hung at whereever it's hanging. Portage 2.0.51.22-r2 (default-linux/x86/2005.0, gcc-3.3.5-20050130, glibc-2.3.5-r0, 2.6.11-gentoo-r9 i686) ================================================================= System uname: 2.6.11-gentoo-r9 i686 Unknown CPU Typ Gentoo Base System version 1.6.13 dev-lang/python: 2.3.5 sys-apps/sandbox: 1.2.11 sys-devel/autoconf: 2.13, 2.59-r6 sys-devel/automake: 1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.5 sys-devel/binutils: 2.15.92.0.2-r10 sys-devel/libtool: 1.5.18-r1 virtual/os-headers: 2.6.11-r2 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" CBUILD="i686-pc-linux-gnu" CFLAGS="-O2 -mcpu=athlon-xp -fomit-frame-pointer -funroll-loops -pipe" CHOST="i686-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3/share/config /usr/lib/X11/xkb /usr/share/config /var/qmail/control" CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d" CXXFLAGS="-O2 -mcpu=athlon-xp -fomit-frame-pointer -funroll-loops -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoconfig distlocks sandbox sfperms strict" GENTOO_MIRRORS="ftp://ftp.ussg.iu.edu/pub/linux/gentoo http://gentoo.chem.wisc.edu/gentoo/ http://open-systems.ufl.edu/mirrors/gentoo http://gentoo.llarian.net/ http://gentoo.mirrored.ca/ http://mirror.datapipe.net/gentoo ftp://gentoo.risq.qc.ca/" MAKEOPTS="-j2" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" SYNC="rsync://rsync.gentoo.org/gentoo-portage" USE="x86 apache2 apm async avi berkdb bitmap-fonts bzlib cgi chroot crypt curl eds emboss encode ethereal fam foomaticdb gd gif gnutls gstreamer gtk gtk2 jpeg junit ldap libg++ libwww mad mikmod mmap mp3 mppe-mppc mysql ncurses nptl ntlm offensive ogg pam perl php png posix python quicktime readline samba sdl server slang snmp ssl tcpd truetype-fonts type1-fonts vorbis xmms zlib userland_GNU kernel_linux elibc_glibc" Unset: ASFLAGS, CTARGET, LANG, LC_ALL, LDFLAGS, LINGUAS, PORTDIR_OVERLAY
Created attachment 65388 [details] Debug Level 10 log.smbd from startup to the client hanging
Created attachment 65389 [details] log.nmbd at log level 10 from the same time span
Created attachment 65390 [details] samba log for the ibook itself at log level 10
Created attachment 65391 [details] smb.conf -- in case it helps anyone out.
complich8: instead of the smb.conf, could you please attach the output of
the previous comment would have liked to be something more complete: retry: complich8: instead of the smb.conf, could you please attach the output of 'testparm -v -s > ${file} 2>&1'?
Created attachment 65535 [details] testparm -v -s output you got it.
I'm seeing the same exact behavior connecting to my work fileservers, which are xeon-based gentoo boxes. And I just tested on a solaris box with samba 3.0.14a and found the same exact behavior. Other relevant links include: http://www.stillhq.com/samba/000001.commentform.html and the linked mailing list thread http://www.mail-archive.com/samba@lists.samba.org/msg58641.html
From what I'm seeing, it should be a combination of (missing) 'client NTLMv2 auth = Yes' 'client plaintext auth = No' 'use sendfile= No' (this one shouldn't be needed, but it's here for nice company) These settings are to be applied on all failing unixes' smb.conf. Also look at the man pages of smb.conf for all the recurrencies of 'signing', 'ntlm', 'auth'. The client seems broken, but these settings could stop the failures to show up.
I tried applying the settings you suggested (the client ntlmv2 auth = yes, client plaintext auth = no, and client and server signing both set to auto). I also set ntlm auth to explicitly enabled. Restarted smbd on the server, rebooted the client laptop. That doesn't seem to be it. Clientside still hangs, server still shows the same error logs.
Well, the result I got from samba's bugzilla goes something like this: ------- Additional Comments From jerry@samba.org 2005-08-24 06:26 ------- This should be fixed in 3.0.20. There was one specific problem with our rpc_bind server code. So I guess whenever 3.0.20 rolls down the line, we'll see the problem go away. I'm going to consider it resolved at that.
If you're in a hurry, samba-3.0.20 is being polished on bug #103164
