[Samba] Address family not supported by protocol
I've compile a Samba 4.0.9 for x86_64 with the following options: CPPFLAGS="-D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -D_GNU_SOURCE -march=atom -O2 -pipe -fomit-frame-pointer" \ samba_cv_HAVE_GETTIMEOFDAY_TZ=yes \ samba_cv_HAVE_IFACE_IFCONF=yes \ samba_cv_HAVE_IFACE_IFREQ=yes \ ac_cv_have_setresuid=yes \ ac_cv_have_setresgid=yes \ ac_cv_file__proc_sys_kernel_core_pattern=yes \ samba_cv_USE_SETRESUID=yes \ samba_cv_HAVE_KERNEL_OPLOCKS_LINUX=yes \ samba_cv_HAVE_WRFILE_KEYTAB=yes \ samba_cv_HAVE_OFF64_T=yes \ samba_cv_have_longlong=yes \ samba_cv_HAVE_MMAP=yes \ samba_cv_HAVE_INO64_T=yes \ samba_cv_CC_NEGATIVE_ENUM_VALUES=yes \ smb_krb5_cv_enctype_to_string_takes_krb5_context_arg=no \ smb_krb5_cv_enctype_to_string_takes_size_t_arg=yes \ ./configure \ --without-pie \ --disable-cups \ --disable-iprint \ --with-configdir=/etc/samba \ --with-logfilebase=/tmp/samba \ --with-lockdir=/tmp/samba \ --with-piddir=/tmp/samba \ --with-privatedir=/etc/samba/private \ --with-sendfile-support \ --with-ldap \ --with-ads \ --with-pam \ --with-pammodulesdir=/lib/x86_64-linux-gnu/security \ --with-pam_smbpass \ --with-winbind \ --with-acl-support \ --with-automount \ --enable-pthreadpool \ --with-dnsupdate \ --with-shared-modules=idmap_ad,idmap_rid \ --localstatedir=/var \ --with-libiconv=/usr \ --with-cachedir=/mnt/system/samba/system \ --prefix=/usr/local/samba \ --without-ad-dc \ --without-swat \ --without-quotas \ --with-aio-support \ --fail-immediately \ --jobs=8 After joining samba on a AD environment, I am not able to list users/groups domains. I changed Log Level =2 and got this weird "Address family not supported by protocol" error on *log.smbd* [2013/10/08 11:49:55, 0] ../source3/smbd/server.c:1201(main) smbd version 4.0.9 started. Copyright Andrew Tridgell and the Samba Team 1992-2012 [2013/10/08 11:49:55.136632, 0] ../source3/smbd/server.c:1281(main) standard input is not a socket, assuming -D option [2013/10/08 11:49:55.763906, 0] ../source3/printing/print_standard.c:68(std_pcap_cache_reload) Unable to open printcap file lpstat for read! [2013/10/08 11:49:55.765678, 0] ../source3/lib/util_sock.c:423(open_socket_in) open_socket_in(): socket() call failed: Address family not supported by protocol [2013/10/08 11:49:55.765901, 0] ../source3/smbd/server.c:685(smbd_open_one_socket) smbd_open_once_socket: open_socket_in: Address family not supported by protocol [2013/10/08 11:49:55.768391, 0] ../source3/lib/util_sock.c:423(open_socket_in) open_socket_in(): socket() call failed: Address family not supported by protocol [2013/10/08 11:49:55.768565, 0] ../source3/smbd/server.c:685(smbd_open_one_socket) smbd_open_once_socket: open_socket_in: Address family not supported by protocol [2013/10/08 11:50:56.374475, 0] ../source3/printing/print_standard.c:68(std_pcap_cache_reload) Unable to open printcap file lpstat for read! [2013/10/08 12:03:57.631802, 0] ../source3/printing/print_standard.c:68(std_pcap_cache_reload) Unable to open printcap file lpstat for read! [2013/10/08 12:16:58.867944, 0] ../source3/printing/print_standard.c:68(std_pcap_cache_reload) Unable to open printcap file lpstat for read! [2013/10/08 12:29:59.868909, 0] ../source3/printing/print_standard.c:68(std_pcap_cache_reload) Unable to open printcap file lpstat for read! [2013/10/08 12:43:01.144145, 0] ../source3/printing/print_standard.c:68(std_pcap_cache_reload) Unable to open printcap file lpstat for read! [2013/10/08 12:56:02.329592, 0] ../source3/printing/print_standard.c:68(std_pcap_cache_reload) Unable to open printcap file lpstat for read! [2013/10/08 13:09:03.247922, 0] ../source3/printing/print_standard.c:68(std_pcap_cache_reload) Unable to open printcap file lpstat for read! At log.winbindd I got: [2013/10/08 13:12:18, 0] ../source3/winbindd/winbindd.c:1409(main) winbindd version 4.0.9 started. Copyright Andrew Tridgell and the Samba Team 1992-2012 [2013/10/08 13:12:18, 2] ../source3/lib/tallocmsg.c:124(register_msg_pool_usage) Registered MSG_REQ_POOL_USAGE [2013/10/08 13:12:18, 2] ../source3/lib/dmallocmsg.c:78(register_dmalloc_msgs) Registered MSG_REQ_DMALLOC_MARK and LOG_CHANGED [2013/10/08 13:12:18.895808, 2] ../source3/param/loadparm.c:544(max_open_files) rlimit_max: increasing rlimit_max (4096) to minimum Windows limit (16384) [2013/10/08 13:12:18.896176, 1] ../source3/param/loadparm.c:3156(lp_do_parameter) WARNING: The "idmap backend" option is deprecated [2013/10/08 13:12:18.896355, 1] ../source3/param/loadparm.c:3156(lp_do_parameter) WARNING: The "idmap uid" option is deprecated [2013/10/08 13:12:18.896521, 1] ../source3/param/loadparm.c:3156(lp_do_parameter) WARNING: The "idmap gid" option is deprecated [2013/10/08 13:12:18.897759,
Re: [Samba] Samba4 consumes more CPU
Googling around copy_user_generic_unrolled() - a kernel space function -
seen in my previous smbd profiling, I found what might be a clue for the
performance drop. It is a comment on line #31 (see below) that says:
31 /*
32 * If CPU has ERMS feature, use copy_user_enhanced_fast_string.
33 * Otherwise, if CPU has rep_good feature, use copy_user_generic_string.
34 * Otherwise, use copy_user_generic_unrolled.
35 */
Which makes me guess that my Atom D2701 (
http://ark.intel.com/products/59683/Intel-Atom-Processor-D2700-1M-Cache-2_13-GHz)
is not compiled with REP_GOOD nor ERMS. It is not clear to me if the
processor does support those features, but apparently it does (looking at
/proc/cpuinfo from another user's NAS -
http://www.foxnetwork.ru/index.php/en/component/content/article/121-thecus-n4800eco.html
)
__
linux/arch/x86/include/asm/uaccess_64.h
Toggle line number - Style:
1 #ifndef _ASM_X86_UACCESS_64_H
2 #define _ASM_X86_UACCESS_64_H
3
4 /*
5 * User space memory access functions
6 */
7 #include
8 #include
9 #include
10 #include
11 #include
12 #include
13
14 /*
15 * Copy To/From Userspace
16 */
17
18 /* Handles exceptions in both to and from, but doesn't do access_ok */
19 __must_check unsigned long
20 copy_user_enhanced_fast_string(void *to, const void *from, unsigned len);
21 __must_check unsigned long
22 copy_user_generic_string(void *to, const void *from, unsigned len);
23 __must_check unsigned long
24 copy_user_generic_unrolled(void *to, const void *from, unsigned len);
25
26 static __always_inline __must_check unsigned long
27 copy_user_generic(void *to, const void *from, unsigned len)
28 {
29 unsigned ret;
30
31 /*
32 * If CPU has ERMS feature, use copy_user_enhanced_fast_string.
33 * Otherwise, if CPU has rep_good feature, use copy_user_generic_string.
34 * Otherwise, use copy_user_generic_unrolled.
35 */
36 alternative_call_2(copy_user_generic_unrolled,
37 copy_user_generic_string,
38 X86_FEATURE_REP_GOOD,
39 copy_user_enhanced_fast_string,
40 X86_FEATURE_ERMS,
41 ASM_OUTPUT2(""=a"" (ret), ""=D"" (to), ""=S"" (from),
42 ""=d"" (len)),
43 ""1"" (to), ""2"" (from), ""3"" (len)
44 : ""memory"", ""rcx"", ""r8"", ""r9"", ""r10"", ""r11"");
45 return ret;
46 }
On Tue, Oct 1, 2013 at 6:04 PM, Thiago Fernandes Crepaldi wrote:
> That is funny. Now that I replaced samba 4 and libc-2.13.so with debug
> symbols, the perf profile seems to be have changed a bit after the same
> tests !
>
> Events: 54K cycles
> - 3.06% smbd [kernel.kallsyms] [k] copy_user_generic_unrolled
>- copy_user_generic_unrolled
> 52.63% __read_nocancel
> 36.20% __write_nocancel
> 2.70% __getdents64
> 2.44% __libc_readv
> + 2.00% do_fcntl
> 0.87% __GI___libc_read
> + 0.77% __fxstat64
> - 2.02% smbd libc-2.13.so [.] _int_malloc
>+ _int_malloc
> - 1.62% smbd [kernel.kallsyms] [k] kmem_cache_alloc
>+ kmem_cache_alloc
> - 1.22% smbd libtalloc.so.2.0.7[.] _talloc_free
>+ _talloc_free
> - 0.99% smbd libtalloc.so.2.0.7[.]
> _talloc_free_children_internal.isra.4
>+ _talloc_free_children_internal.isra.4
> - 0.86% smbd libc-2.13.so [.] __memcpy_ssse3
>+ __memcpy_ssse3
> + 0.81% smbd [kernel.kallsyms] [k] kmem_cache_free
> + 0.81% smbd libc-2.13.so [.] _int_free
> + 0.79% smbd [kernel.kallsyms] [k] __kmalloc
> + 0.66% smbd libtalloc.so.2.0.7[.] _talloc_zero
> + 0.63% smbd [kernel.kallsyms] [k] link_path_walk
> + 0.63% smbd [kernel.kallsyms] [k] ext4_htree_store_dirent
> + 0.55% smbd libtalloc.so.2.0.7[.] talloc_alloc_pool
> + 0.55% smbd libc-2.13.so [.] __memset_sse2
> + 0.53% smbd libc-2.13.so [.] malloc
> + 0.53% smbd [kernel.kallsyms] [k] fcntl_setlk
> + 0.52% smbd [kernel.kallsyms] [k] get_page_from_freelist
> + 0.50% smbd libtalloc.so.2.0.7[.] talloc_get_name
> + 0.50% smbd [kernel.kallsyms] [k] tg3_start_xmit
> + 0.48% smbd [kernel.kallsyms] [k] memset
> + 0.47% smbd libc-2.13.so [.] free
> + 0.47% smbd [kernel.kallsyms] [k] _raw_spin_lock
> + 0.45% smbd [kernel.kallsyms] [k] __d_lookup_rcu
> + 0.45% smbd libc-2.13.so [.] __GI___strcmp_ssse3
> + 0.44% smbd libtalloc.so.2.0.7[.] _talloc_get_type_abort
> + 0.43% smbd [kernel.kallsyms]
Re: [Samba] Samba4 consumes more CPU
That is funny. Now that I replaced samba 4 and libc-2.13.so with debug symbols, the perf profile seems to be have changed a bit after the same tests ! Events: 54K cycles - 3.06% smbd [kernel.kallsyms] [k] copy_user_generic_unrolled - copy_user_generic_unrolled 52.63% __read_nocancel 36.20% __write_nocancel 2.70% __getdents64 2.44% __libc_readv + 2.00% do_fcntl 0.87% __GI___libc_read + 0.77% __fxstat64 - 2.02% smbd libc-2.13.so [.] _int_malloc + _int_malloc - 1.62% smbd [kernel.kallsyms] [k] kmem_cache_alloc + kmem_cache_alloc - 1.22% smbd libtalloc.so.2.0.7[.] _talloc_free + _talloc_free - 0.99% smbd libtalloc.so.2.0.7[.] _talloc_free_children_internal.isra.4 + _talloc_free_children_internal.isra.4 - 0.86% smbd libc-2.13.so [.] __memcpy_ssse3 + __memcpy_ssse3 + 0.81% smbd [kernel.kallsyms] [k] kmem_cache_free + 0.81% smbd libc-2.13.so [.] _int_free + 0.79% smbd [kernel.kallsyms] [k] __kmalloc + 0.66% smbd libtalloc.so.2.0.7[.] _talloc_zero + 0.63% smbd [kernel.kallsyms] [k] link_path_walk + 0.63% smbd [kernel.kallsyms] [k] ext4_htree_store_dirent + 0.55% smbd libtalloc.so.2.0.7[.] talloc_alloc_pool + 0.55% smbd libc-2.13.so [.] __memset_sse2 + 0.53% smbd libc-2.13.so [.] malloc + 0.53% smbd [kernel.kallsyms] [k] fcntl_setlk + 0.52% smbd [kernel.kallsyms] [k] get_page_from_freelist + 0.50% smbd libtalloc.so.2.0.7[.] talloc_get_name + 0.50% smbd [kernel.kallsyms] [k] tg3_start_xmit + 0.48% smbd [kernel.kallsyms] [k] memset + 0.47% smbd libc-2.13.so [.] free + 0.47% smbd [kernel.kallsyms] [k] _raw_spin_lock + 0.45% smbd [kernel.kallsyms] [k] __d_lookup_rcu + 0.45% smbd libc-2.13.so [.] __GI___strcmp_ssse3 + 0.44% smbd libtalloc.so.2.0.7[.] _talloc_get_type_abort + 0.43% smbd [kernel.kallsyms] [k] system_call_after_swapgs + 0.43% smbd [kernel.kallsyms] [k] ext4_mark_iloc_dirty + 0.42% smbd libtalloc.so.2.0.7[.] talloc_is_parent + 0.41% smbd [kernel.kallsyms] [k] __alloc_skb + 0.41% smbd [kernel.kallsyms] [k] __posix_lock_file + 0.40% smbd [kernel.kallsyms] [k] __ext4_get_inode_loc + 0.39% smbd libc-2.13.so [.] __strlen_sse2 + 0.39% smbd [kernel.kallsyms] [k] kfree + 0.39% smbd [kernel.kallsyms] [k] tcp_recvmsg + 0.38% smbd libtalloc.so.2.0.7[.] talloc_named_const + 0.37% smbd libtalloc.so.2.0.7[.] _talloc_array On Mon, Sep 30, 2013 at 6:19 PM, Thiago Fernandes Crepaldi < togn...@gmail.com> wrote: > Agreed. For some strange reason I though perf would "follow" the new smbd > forked and account their data too =) > > Unfortunately, I don't have the libc symbols (at least for today) to see > what is going on there, but here is what I got in the child smbd process on > the server side. The client side is a Windows 7 Virtual machine running > NASPT > > Could this result mean that most of the time the performance drop I am > experiencing is due to libc ? > I've never worked with perf before, but I will still try to resolve those > crazy addresses > > Events: 45K cycles > - 7.37% smbd libc-2.13.so [.] 0x11e465 >- 0x7ffab9f2043c > 41.73% 0 > 5.32% 0x1b3fbe0 > 5.29% 0x2c4dab0 > 3.60% 0x1b0b130 > 3.37% 0x1b0b2a0 > 2.94% 0x1b5af80 > 2.70% 0x1b0d850 > 2.64% 0x2825fb0 > 1.86% 0x28e06d0 > 1.83% 0x2afcc80 > 1.71% 0x1b2ccb0 > 1.64% 0x2a4deb0 > 1.63% 0x1b56e00 > 1.51% 0x1b6bd00 > 1.16% 0x1b49eb0 > 1.15% 0x1b506e0 > 1.13% 0x1b4da00 > 1.07% 0x1b35100 > 0.93% 0x1af9050 > 0.92% 0x2b03680 > 0.91% 0x2ae21f0 > 0.90% 0x1b21210 > 0.89% 0x1b5de80 > 0.89% 0x1b5aa80 > 0.89% 0x1b2e0e0 > 0.88% 0x1b59be0 > 0.87% 0x1b4c600 > 0.86% 0x1b2aa20 > 0.85% 0x1b4a940 > 0.85% 0x1b45f50 > 0.84% 0x1b4a6d0 > 0.84% 0x1b23940 > 0.82% 0x1b37210 > 0.82% 0x1b2cf30 > 0.82% 0x1b33320 > 0.77% 0x2c96d50 > 0.76% 0x202f380 > 0.75% 0x2bd0bd0 > 0.66% 0x1b5e1d0 >- 0x7ffab9f27e10 > 37.72% 0x2f62696c2f3365 > + 23.78% 0 > + 11.24% 0x7fffc9f76d40 > + 6.25% set_unix_security_ctx > 3.13% 0x645f6e656b6f74 > 2.46% 0x10009 > + 2.17% 0x11b9f22aac >
Re: [Samba] Samba4 consumes more CPU
Agreed. For some strange reason I though perf would "follow" the new smbd forked and account their data too =) Unfortunately, I don't have the libc symbols (at least for today) to see what is going on there, but here is what I got in the child smbd process on the server side. The client side is a Windows 7 Virtual machine running NASPT Could this result mean that most of the time the performance drop I am experiencing is due to libc ? I've never worked with perf before, but I will still try to resolve those crazy addresses Events: 45K cycles - 7.37% smbd libc-2.13.so [.] 0x11e465 - 0x7ffab9f2043c 41.73% 0 5.32% 0x1b3fbe0 5.29% 0x2c4dab0 3.60% 0x1b0b130 3.37% 0x1b0b2a0 2.94% 0x1b5af80 2.70% 0x1b0d850 2.64% 0x2825fb0 1.86% 0x28e06d0 1.83% 0x2afcc80 1.71% 0x1b2ccb0 1.64% 0x2a4deb0 1.63% 0x1b56e00 1.51% 0x1b6bd00 1.16% 0x1b49eb0 1.15% 0x1b506e0 1.13% 0x1b4da00 1.07% 0x1b35100 0.93% 0x1af9050 0.92% 0x2b03680 0.91% 0x2ae21f0 0.90% 0x1b21210 0.89% 0x1b5de80 0.89% 0x1b5aa80 0.89% 0x1b2e0e0 0.88% 0x1b59be0 0.87% 0x1b4c600 0.86% 0x1b2aa20 0.85% 0x1b4a940 0.85% 0x1b45f50 0.84% 0x1b4a6d0 0.84% 0x1b23940 0.82% 0x1b37210 0.82% 0x1b2cf30 0.82% 0x1b33320 0.77% 0x2c96d50 0.76% 0x202f380 0.75% 0x2bd0bd0 0.66% 0x1b5e1d0 - 0x7ffab9f27e10 37.72% 0x2f62696c2f3365 + 23.78% 0 + 11.24% 0x7fffc9f76d40 + 6.25% set_unix_security_ctx 3.13% 0x645f6e656b6f74 2.46% 0x10009 + 2.17% 0x11b9f22aac 2.16% 0x1b53000 + 2.12% 0x2a29850 2.08% 0xbe70f04c4c 2.01% 0x1b0af00 1.94% 0x1b07390 1.51% 0x1b49b00 1.41% 0x2010 - 0x7ffab9fc6c10 + 18.08% 0 + 13.63% 0x2c5fc20 + 11.62% 0x2be7b10 + 7.90% 0x2be8560 + 6.61% 0x2a29850 + 6.30% 0x2b3d6c0 5.67% 0x4e6f5479706f43 + 5.64% 0x29d7110 + 5.54% 0x2467130 + 5.53% 0x2b3d5e0 + 5.31% 0x28c81a0 + 4.20% 0x2c5fa30 + 3.98% 0x2a98990 + 0x7ffab9f20438 + 0x7ffab9f2045c 0x7ffab9fc8e03 + 0x7ffab9fc425e + 0x7ffab9f2a715 + 0x7ffab9f2a6d0 0x7ffab9f1f851 0x7ffab9f1f2ac + 0x7ffab9f27e25 + 0x7ffab9f2a648 + 0x7ffab9fc4240 0x7ffab9fc8654 0x7ffab9f206bf + 0x7ffab9f20548 + 0x7ffab9f20bc2 + 0x7ffab9f1f130 + 0x7ffab9f26310 + 0x7ffab9f20422 0x7ffab9f1e0db 0x7ffab9f1f179 + 0x7ffab9f2a6f2 + 0x7ffab9f20572 + 0x7ffab9f2054c + 0x7ffab9fc42c5 - 1.72% smbd [kernel.kallsyms] [k] kmem_cache_alloc + kmem_cache_alloc - 1.30% smbd libtalloc.so.2.0.7[.] _talloc_free + _talloc_free - 1.10% smbd libtalloc.so.2.0.7[.] _talloc_free_children_internal.i + _talloc_free_children_internal.isra.4 - 1.07% smbd [kernel.kallsyms] [k] copy_user_generic_unrolled + copy_user_generic_unrolled - 0.95% smbd [kernel.kallsyms] [k] __kmalloc + __kmalloc - 0.78% smbd [kernel.kallsyms] [k] ext4_htree_store_dirent + ext4_htree_store_dirent + 0x7ffab9f4f2f5 - 0.73% smbd [kernel.kallsyms] [k] kmem_cache_free + kmem_cache_free - 0.73% smbd [kernel.kallsyms] [k] link_path_walk + link_path_walk - 0.69% smbd libc-2.13.so [.] malloc + malloc - 0.69% smbd libtalloc.so.2.0.7[.] _talloc_zero + _talloc_zero - 0.62% smbd [kernel.kallsyms] [k] fcntl_setlk + fcntl_setlk + 0x7ffabcf93238 - 0.59% smbd [kernel.kallsyms] [k] __d_lookup_rcu + __d_lookup_rcu - 0.57% smbd libtalloc.so.2.0.7[.] talloc_alloc_pool + talloc_alloc_pool - 0.55% smbd libtalloc.so.2.0.7[.] talloc_get_name + talloc_get_name - 0.55% smbd [kernel.kallsyms] [k] __posix_lock_file + __posix_lock_file + 0x7ffabcf93238 - 0.50% smbd [kernel.kallsyms] [k] _raw_spin_lock + _raw_spin_lock + 0.49% smbd [kernel.kallsyms] [k] tg3_start_xmit + 0.48% smbd [kernel.kallsyms] [k] system_call_after_swapgs + 0.46% smbd libtalloc.so.2.0.7[.] talloc_named_const + 0.46% smbd [kernel.kallsyms] [k] memset + 0.46% smbd libtalloc.so.2.0.7[.] _talloc_get_type_abort + 0.45% smbd [kernel.kallsyms] [k] str2hashbuf_signed + 0.45% smbd [kernel.kallsyms] [k] kfree + 0.45% smbd libc-2.13.so [.] free + 0.44% smbd [kernel.kallsyms] [k] __alloc_skb + 0.42% smbd libtalloc.so.2.0.7[.] talloc_is_parent + 0.41% smbd libtalloc.so.2.0.7[.] _talloc_array On Mon, Sep 30, 2013 at 5:39 PM, Jeremy Allison wrote: > On Mon, Sep 30, 2013 at 05:21:44PM
Re: [Samba] Samba4 consumes more CPU
Andrew, in my company we are also experiencing a higher CPU usage of Samba 4 (smbd) if compared to Samba 3. In fact, it almost reaches 100% of CPU and uses all the memory during *dir copies* (individual file copy is as good as samba 3's). I strongly believe that this CPU usage is the responsible for a worse samba 4's throughput if compared to Samba 3 tests. Giving that, I would like to contribute with this investigation and share my data regarding perf profiling on smbd (parent process) Events: 7 cycles - 90.01% smbd [kernel.kallsyms] [k] copy_pte_range copy_pte_range __libc_fork smbd_accept_connection - 9.77% smbd [kernel.kallsyms] [k] handle_edge_irq handle_edge_irq smbd_accept_connection - 0.22% smbd [kernel.kallsyms] [k] perf_pmu_rotate_start.isra.57 perf_pmu_rotate_start.isra.57 __poll - 0.00% smbd [kernel.kallsyms] [k] native_write_msr_safe native_write_msr_safe __poll My smb.conf is: [Global] available= yes client signing= auto server signing= auto server string= LenovoEMCâ ¢ px4-400r Workgroup= WORKGROUP security= user domain master= auto preferred master= auto local master= yes os level= 20 invalid users= bin daemon adm sync shutdown halt mail news uucp gopher map to guest= Bad User host msdfs= yes restrict anonymous= 0 strict allocate= yes encrypt passwords= yes passdb backend= smbpasswd printcap name= lpstat printable= no load printers= yes max smbd processes= 500 getwd cache= yes syslog= 0 use sendfile= yes browse directory= /tmp/samba winbind sequence directory= /tmp/samba log level= 0 max log size= 50 unix extensions= no veto files= /.AppleDouble/.AppleDB/.bin/.AppleDesktop/Network Trash Folder/Temporary Items/:2eDS_Store/ [Printers] path= /mnt/system/samba/spool printable= yes only guest= yes use client driver= yes comment= All Printers [USB_UnkVend_USB_DISK_MOD_a_1] path= /mnt/ext/USB_UnkVend_USB_DISK_MOD_a_1/ max connections= 150 directory mode= 0777 create mode= 0777 follow symlinks= yes wide links= no strict allocate= no nt acl support= no dos filemode= no writeable= yes public= yes store dos attributes= yes write list= guest [Backups] path= /mnt/pools/A/A0/Backups/ max connections= 150 directory mode= 0777 create mode= 0777 follow symlinks= yes wide links= no nt acl support= no dos filemode= no writeable= yes public= yes store dos attributes= yes write list= guest [Documents] path= /mnt/pools/A/A0/Documents/ max connections= 150 directory mode= 0777 create mode= 0777 follow symlinks= yes wide links= no nt acl support= no dos filemode= no writeable= yes public= yes store dos attributes= yes write list= guest My samba 4.0.9 was compiled with the following options: CPPFLAGS="-D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -D_GNU_SOURCE -march=atom -O2 -pipe -fomit-frame-pointer" \ samba_cv_HAVE_GETTIMEOFDAY_TZ=yes \ samba_cv_HAVE_IFACE_IFCONF=yes \ samba_cv_HAVE_IFACE_IFREQ=yes \ ac_cv_have_setresuid=yes \ ac_cv_have_setresgid=yes \ ac_cv_file__proc_sys_kernel_core_pattern=yes \ samba_cv_USE_SETRESUID=yes \ samba_cv_HAVE_KERNEL_OPLOCKS_LINUX=yes \ samba_cv_HAVE_WRFILE_KEYTAB=yes \ samba_cv_HAVE_OFF64_T=yes \ samba_cv_have_longlong=yes \ samba_cv_HAVE_MMAP=yes \ samba_cv_HAVE_INO64_T=yes \ samba_cv_CC_NEGATIVE_ENUM_VALUES=yes \ smb_krb5_cv_enctype_to_string_takes_krb5_context_arg=no \ smb_krb5_cv_enctype_to_string_takes_size_t_arg=yes \ ./configure \ --without-pie \ --disable-cups \ --disable-iprint \ --with-configdir=/etc/samba \ --with-logfilebase=/tmp/samba \ --with-lockdir=/tmp/samba \ --with-piddir=/tmp/samba \ --with-privatedir=/etc/samba/private \ --with-sendfile-support \ --with-ldap \ --with-ads \ --with-pam \ --with-pammodulesdir=/lib/x86_64-linux-gnu/security \ --with-pam_smbpass \ --with-winbind \ --with-acl-support \ --with-automount \ --enable-pthreadpool \ --with-dnsupdate \ --with-shared-modules=idmap_ad,idmap_rid \ --localstatedir=/var \ --with-libiconv=/usr \ --with-cachedir=/mnt/system/samba/system \ --prefix=/usr/local/samba \ --without-ad-dc \ --without-swat \ --without-quotas \ --with-aio-support \ --fail-immediately \ --jobs=8 \ --enable-debug \ # i added this to use with perf only # Please, let me know if I can do help more ! Thiago On Mon, Sep 2, 2013 at 5:50 PM, Andrew Bartlett wrote: > On Mon, 2013-08-26 at 22:39 +0530, Prema wrote: > > Dear Andrew, > > > > As per your suggestion , I have attached the gdb log of the samba and > smbd > > process log running in the single server mode. > > Also when I noted in the perf top, libndr.so consumes the maximum cpu. > > I noticed that it happens soon after sometime the samba process is > started > > and the CPU is filled up. > > Since the samba process occupies 100% atleast two or
Re: [Samba] Bug on PAM_Winbind ?
One interesting thing is that if I put pam_winbind in front of pam_smbpass on /etc/pam.d/pwauth, AD domain users will be granted access, workgroup users will NOT and the core dump does not occur. I can't really know if the problem is in pam_smbpass or pam_winbind. pwauth code is correct, though. */etc/pam.d/pwauth which does't not coredump* * * #%PAM-1.0 auth sufficient /lib/security/pam_winbind.so cached_login auth required /lib/security/pam_winbind.so krb5_auth auth sufficient /lib/security/pam_smbpass.so accountrequired /lib/security/pam_nologin.so accountrequired /lib/security/pam_winbind.so accountsufficient /lib/security/pam_smbpass.so password required /lib/security/pam_winbind.so password sufficient /lib/security/pam_smbpass.so sessionrequired /lib/security/pam_unix.so On Wed, Aug 21, 2013 at 12:14 PM, Thiago Fernandes Crepaldi < togn...@gmail.com> wrote: > Hello guys, > > I am using PAM (0.82) to authenticate (samba) workgroup users and it works > fine (pam_smbpass). But after joining a domain and start using pam_winbind > too, pwauth (2.3.10) get a segmentation fault when trying to authenticate > any workgroup or domain user. What happens is that it does authenticate the > user successfully, but when it calls pam_end() to release its resources, it > crashes. Maybe pam_winbind is freeing something that will be freed by > pam_end or something like that - I don't really know about PAM and their > modules (smbpass/winbind), but it seems that pam_winbind might be misusing > libtalloc somehow. Any ideas ? > > I believe this is an important use case because (AFAIK) it is the > recommended way of authenticating users on web servers. Although I am using > Samba 4.0.9 (with symbols), I can also see this issue on Samba 4.0.7 and > 4.0.0 too. The funny thing is that it works fine on samba 3.6.9, though. > > *pwauth backtrace:* > > Program received signal SIGSEGV, Segmentation fault. > 0xb7e32f03 in ?? () > (gdb) bt > #0 0xb7e32f03 in ?? () > *#1 0xb7bb26e1 in _talloc_free_internal (ptr=0x80612d0, > location=0xb7bb54c7 "../lib/talloc/talloc.c:2251") at > ../lib/talloc/talloc.c:831* > *#2 0xb7bb33f0 in _talloc_free_children_internal (tc=0x8060d88, > ptr=0x8060db8, location=0xb7bb54c7 "../lib/talloc/talloc.c:2251") at > ../lib/talloc/talloc.c:1256* > *#3 0xb7bb2830 in _talloc_free_internal (ptr=0x8060db8, > location=0xb7bb54c7 "../lib/talloc/talloc.c:2251") at > ../lib/talloc/talloc.c:851* > *#4 0xb7bb3742 in _talloc_free (ptr=0x8060db8, location=0xb7bb54c7 > "../lib/talloc/talloc.c:2251") at ../lib/talloc/talloc.c:1371* > *#5 0xb7bb4d82 in talloc_autofree () at ../lib/talloc/talloc.c:2251* > *#6 0xb7e865e8 in __cxa_finalize () from /lib/i386-linux-gnu/libc.so.6* > *#7 0xb7bb1a43 in __do_global_dtors_aux () from > /usr/local/samba/lib/private/libtalloc.so.2* > #8 0xb7ff4e52 in ?? () from /lib/ld-linux.so.2 > #9 0xb7ff5947 in ?? () from /lib/ld-linux.so.2 > #10 0xb7e53cc4 in ?? () from /lib/i386-linux-gnu/libdl.so.2 > #11 0xb7fefde6 in ?? () from /lib/ld-linux.so.2 > #12 0xb7e540bc in ?? () from /lib/i386-linux-gnu/libdl.so.2 > #13 0xb7e53cfa in dlclose () from /lib/i386-linux-gnu/libdl.so.2 > #14 0xb7fadf8b in ?? () from /lib/i386-linux-gnu/libpam.so.0 > #15 0xb7fab8ff in ?? () from /lib/i386-linux-gnu/libpam.so.0 > *#16 0xb7fa8da0 in pam_end () from /lib/i386-linux-gnu/libpam.so.0* > #17 0x08048baf in check_auth (login=0xb8bf "admin", passwd=0xb4be > "soho") at auth_pam.c:186 > #18 0x08048952 in main (argc=1, argv=0xbd84) at main.c:92 > > *cat /etc/nsswitch.conf * > > passwd: files winbind > shadow: files winbind > group: files winbind > hosts: files wins dns > bootparams: nisplus [NOTFOUND=return] files > ethers: files > netmasks: files > networks: files > protocols: files > rpc:files > services: files > netgroup: nisplus > publickey: nisplus > automount: files nisplus > aliases:files nisplus > > *cat /etc/pam.d/pwauth * > #%PAM-1.0 > auth sufficient /lib/security/pam_smbpass.so > auth sufficient /lib/security/pam_winbind.so cached_login > auth required /lib/security/pam_winbind.so krb5_auth > accountrequired /lib/security/pam_nologin.so > accountsufficient /lib/security/pam_smbpass.so > accountrequired /lib/security/pam_winbind.so > password sufficient /lib/security/pam_smbpass.so > password required /lib/security/pam_winbind.so > sessionrequired /lib/security/pam_unix.so > > *cat /etc/samba/smb.conf* > [Global] > available= yes > client signing= auto > se
Re: [Samba] Bug on PAM_Winbind ?
As expected, if I comment out talloc_free() call, pwauth's pam_end() will
not crash and everything will work (user credentials will be verified and
access will be granted - memory will leak, though)
talloc.c:2249
static void talloc_autofree(void)
{
talloc_free(autofree_context);
}
On Wed, Aug 21, 2013 at 12:14 PM, Thiago Fernandes Crepaldi <
togn...@gmail.com> wrote:
> Hello guys,
>
> I am using PAM (0.82) to authenticate (samba) workgroup users and it works
> fine (pam_smbpass). But after joining a domain and start using pam_winbind
> too, pwauth (2.3.10) get a segmentation fault when trying to authenticate
> any workgroup or domain user. What happens is that it does authenticate the
> user successfully, but when it calls pam_end() to release its resources, it
> crashes. Maybe pam_winbind is freeing something that will be freed by
> pam_end or something like that - I don't really know about PAM and their
> modules (smbpass/winbind), but it seems that pam_winbind might be misusing
> libtalloc somehow. Any ideas ?
>
> I believe this is an important use case because (AFAIK) it is the
> recommended way of authenticating users on web servers. Although I am using
> Samba 4.0.9 (with symbols), I can also see this issue on Samba 4.0.7 and
> 4.0.0 too. The funny thing is that it works fine on samba 3.6.9, though.
>
> *pwauth backtrace:*
>
> Program received signal SIGSEGV, Segmentation fault.
> 0xb7e32f03 in ?? ()
> (gdb) bt
> #0 0xb7e32f03 in ?? ()
> *#1 0xb7bb26e1 in _talloc_free_internal (ptr=0x80612d0,
> location=0xb7bb54c7 "../lib/talloc/talloc.c:2251") at
> ../lib/talloc/talloc.c:831*
> *#2 0xb7bb33f0 in _talloc_free_children_internal (tc=0x8060d88,
> ptr=0x8060db8, location=0xb7bb54c7 "../lib/talloc/talloc.c:2251") at
> ../lib/talloc/talloc.c:1256*
> *#3 0xb7bb2830 in _talloc_free_internal (ptr=0x8060db8,
> location=0xb7bb54c7 "../lib/talloc/talloc.c:2251") at
> ../lib/talloc/talloc.c:851*
> *#4 0xb7bb3742 in _talloc_free (ptr=0x8060db8, location=0xb7bb54c7
> "../lib/talloc/talloc.c:2251") at ../lib/talloc/talloc.c:1371*
> *#5 0xb7bb4d82 in talloc_autofree () at ../lib/talloc/talloc.c:2251*
> *#6 0xb7e865e8 in __cxa_finalize () from /lib/i386-linux-gnu/libc.so.6*
> *#7 0xb7bb1a43 in __do_global_dtors_aux () from
> /usr/local/samba/lib/private/libtalloc.so.2*
> #8 0xb7ff4e52 in ?? () from /lib/ld-linux.so.2
> #9 0xb7ff5947 in ?? () from /lib/ld-linux.so.2
> #10 0xb7e53cc4 in ?? () from /lib/i386-linux-gnu/libdl.so.2
> #11 0xb7fefde6 in ?? () from /lib/ld-linux.so.2
> #12 0xb7e540bc in ?? () from /lib/i386-linux-gnu/libdl.so.2
> #13 0xb7e53cfa in dlclose () from /lib/i386-linux-gnu/libdl.so.2
> #14 0xb7fadf8b in ?? () from /lib/i386-linux-gnu/libpam.so.0
> #15 0xb7fab8ff in ?? () from /lib/i386-linux-gnu/libpam.so.0
> *#16 0xb7fa8da0 in pam_end () from /lib/i386-linux-gnu/libpam.so.0*
> #17 0x08048baf in check_auth (login=0xb8bf "admin", passwd=0xb4be
> "soho") at auth_pam.c:186
> #18 0x08048952 in main (argc=1, argv=0xbd84) at main.c:92
>
> *cat /etc/nsswitch.conf *
>
> passwd: files winbind
> shadow: files winbind
> group: files winbind
> hosts: files wins dns
> bootparams: nisplus [NOTFOUND=return] files
> ethers: files
> netmasks: files
> networks: files
> protocols: files
> rpc:files
> services: files
> netgroup: nisplus
> publickey: nisplus
> automount: files nisplus
> aliases:files nisplus
>
> *cat /etc/pam.d/pwauth *
> #%PAM-1.0
> auth sufficient /lib/security/pam_smbpass.so
> auth sufficient /lib/security/pam_winbind.so cached_login
> auth required /lib/security/pam_winbind.so krb5_auth
> accountrequired /lib/security/pam_nologin.so
> accountsufficient /lib/security/pam_smbpass.so
> accountrequired /lib/security/pam_winbind.so
> password sufficient /lib/security/pam_smbpass.so
> password required /lib/security/pam_winbind.so
> sessionrequired /lib/security/pam_unix.so
>
> *cat /etc/samba/smb.conf*
> [Global]
> available= yes
> client signing= auto
> server signing= auto
> server string= Bla
> Workgroup= DISNEY
> netbios name= vmstore-4
> realm= DISNEY.XXTEST.ASD-ABC.LOCALDOMAIN
> password server= *
> idmap backend= tdb
> idmap uid= 5000-999
> idmap gid= 5000-999
> idmap config DISNEY : backend= rid
> idmap config DISNEY : range= 1000-1999
> security= ADS
> name resolve order= wins host bcast lmhosts
> client use spnego= yes
> dns proxy= no
> winbind use default domain= no
> winbind nested groups= yes
> inherit acls= yes
> winbind enum users= yes
> winbind enum groups= yes
> winbind separator= \\
> winbind cache time= 300
> winbind offline logon= true
> encrypt passwords= yes
> passdb backend= smbpasswd
>
--
Thiago Fernandes Crepaldi (aka Crepaldi)
--
To unsubscribe from this list go to the following URL and read the
instructions: https://lists.samba.org/mailman/options/samba
[Samba] Bug on PAM_Winbind ?
Hello guys, I am using PAM (0.82) to authenticate (samba) workgroup users and it works fine (pam_smbpass). But after joining a domain and start using pam_winbind too, pwauth (2.3.10) get a segmentation fault when trying to authenticate any workgroup or domain user. What happens is that it does authenticate the user successfully, but when it calls pam_end() to release its resources, it crashes. Maybe pam_winbind is freeing something that will be freed by pam_end or something like that - I don't really know about PAM and their modules (smbpass/winbind), but it seems that pam_winbind might be misusing libtalloc somehow. Any ideas ? I believe this is an important use case because (AFAIK) it is the recommended way of authenticating users on web servers. Although I am using Samba 4.0.9 (with symbols), I can also see this issue on Samba 4.0.7 and 4.0.0 too. The funny thing is that it works fine on samba 3.6.9, though. *pwauth backtrace:* Program received signal SIGSEGV, Segmentation fault. 0xb7e32f03 in ?? () (gdb) bt #0 0xb7e32f03 in ?? () *#1 0xb7bb26e1 in _talloc_free_internal (ptr=0x80612d0, location=0xb7bb54c7 "../lib/talloc/talloc.c:2251") at ../lib/talloc/talloc.c:831* *#2 0xb7bb33f0 in _talloc_free_children_internal (tc=0x8060d88, ptr=0x8060db8, location=0xb7bb54c7 "../lib/talloc/talloc.c:2251") at ../lib/talloc/talloc.c:1256* *#3 0xb7bb2830 in _talloc_free_internal (ptr=0x8060db8, location=0xb7bb54c7 "../lib/talloc/talloc.c:2251") at ../lib/talloc/talloc.c:851* *#4 0xb7bb3742 in _talloc_free (ptr=0x8060db8, location=0xb7bb54c7 "../lib/talloc/talloc.c:2251") at ../lib/talloc/talloc.c:1371* *#5 0xb7bb4d82 in talloc_autofree () at ../lib/talloc/talloc.c:2251* *#6 0xb7e865e8 in __cxa_finalize () from /lib/i386-linux-gnu/libc.so.6* *#7 0xb7bb1a43 in __do_global_dtors_aux () from /usr/local/samba/lib/private/libtalloc.so.2* #8 0xb7ff4e52 in ?? () from /lib/ld-linux.so.2 #9 0xb7ff5947 in ?? () from /lib/ld-linux.so.2 #10 0xb7e53cc4 in ?? () from /lib/i386-linux-gnu/libdl.so.2 #11 0xb7fefde6 in ?? () from /lib/ld-linux.so.2 #12 0xb7e540bc in ?? () from /lib/i386-linux-gnu/libdl.so.2 #13 0xb7e53cfa in dlclose () from /lib/i386-linux-gnu/libdl.so.2 #14 0xb7fadf8b in ?? () from /lib/i386-linux-gnu/libpam.so.0 #15 0xb7fab8ff in ?? () from /lib/i386-linux-gnu/libpam.so.0 *#16 0xb7fa8da0 in pam_end () from /lib/i386-linux-gnu/libpam.so.0* #17 0x08048baf in check_auth (login=0xb8bf "admin", passwd=0xb4be "soho") at auth_pam.c:186 #18 0x08048952 in main (argc=1, argv=0xbd84) at main.c:92 *cat /etc/nsswitch.conf * passwd: files winbind shadow: files winbind group: files winbind hosts: files wins dns bootparams: nisplus [NOTFOUND=return] files ethers: files netmasks: files networks: files protocols: files rpc:files services: files netgroup: nisplus publickey: nisplus automount: files nisplus aliases:files nisplus *cat /etc/pam.d/pwauth * #%PAM-1.0 auth sufficient /lib/security/pam_smbpass.so auth sufficient /lib/security/pam_winbind.so cached_login auth required /lib/security/pam_winbind.so krb5_auth accountrequired /lib/security/pam_nologin.so accountsufficient /lib/security/pam_smbpass.so accountrequired /lib/security/pam_winbind.so password sufficient /lib/security/pam_smbpass.so password required /lib/security/pam_winbind.so sessionrequired /lib/security/pam_unix.so *cat /etc/samba/smb.conf* [Global] available= yes client signing= auto server signing= auto server string= Bla Workgroup= DISNEY netbios name= vmstore-4 realm= DISNEY.XXTEST.ASD-ABC.LOCALDOMAIN password server= * idmap backend= tdb idmap uid= 5000-999 idmap gid= 5000-999 idmap config DISNEY : backend= rid idmap config DISNEY : range= 1000-1999 security= ADS name resolve order= wins host bcast lmhosts client use spnego= yes dns proxy= no winbind use default domain= no winbind nested groups= yes inherit acls= yes winbind enum users= yes winbind enum groups= yes winbind separator= \\ winbind cache time= 300 winbind offline logon= true encrypt passwords= yes passdb backend= smbpasswd -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] Samba4 + Winbind + PAM Installation/Configuration
I am justs tarting o samba too, but I think it is normal to have lots of "not found". You have to pay attention for the features you want, though. You might need libpam0g-dev package for PAM support (At least that is the one for Debian) and you can also consider libacl-dev for ACL support (changing permissions for a share on windows, for example) On Sun, Aug 18, 2013 at 11:38 AM, Andreas Krupp wrote: > > Hi, > > I have not set any home var yet in my smb.conf. > If you're asking for that, I am probably missing a lot of important > parameters. > Below my smb.conf for the moment: > > # Global parameters > [global] > workgroup = MYDOMAIN > realm = MYDOMAIN.HOME > netbios name = DC > server role = active directory domain controller > dns forwarder = 10.33.66.99 > template shell = /bin/bash > wins support = yes > > [netlogon] > path = /usr/local/samba/var/locks/sysvol/mydomain.home/scripts > read only = No > > [sysvol] > path = /usr/local/samba/var/locks/sysvol > read only = No > > > Otherwise I checked for all the lines during "./configure" that mention > "not found"... I have more than 100 of these. Is that normal? > Among the things missing are e.g. ldap, pam_start, NFS QUOTAS, and lots of > other stuff... I tried to follow the list of packages to install on the > Samba4 Wiki for CentOS but it seems, that is not really enough, is it? > > Cheers & best, > Andreas > > *On 16 August 2013 08:37, Daniel Müller has written: * > > -- > To unsubscribe from this list go to the following URL and read the > instructions: https://lists.samba.org/mailman/options/samba > -- Thiago Fernandes Crepaldi (aka Crepaldi) -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
