search for: local_password_chang

...ay (util.c:754) ==28624== by 0x5731BE1: smbldap_set_mod_internal (smbldap.c:325) ==28624== by 0x5732166: smbldap_set_mod (smbldap.c:392) ==28624== by 0x4AA410F: ldapsam_create_user (pdb_ldap.c:5305) ==28624== by 0x4AC046E: pdb_create_user (pdb_interface.c:542) ==28624== by 0x4AB1401: local_password_change (passdb.c:745) ==28624== by 0x10B4AB: password_change (smbpasswd.c:264) ==28624== by 0x10BC87: process_root (smbpasswd.c:466) ==28624== by 0x10C1AC: main (smbpasswd.c:627) ==28624== ==28624== Invalid read of size 1 ==28624== at 0x56B4578: ldap_mods_free (in /usr/lib/i386-linux-gnu/lib...

On Fri, Mar 04, 2016 at 10:53:17AM +0100, Volker Lendecke wrote: > On Wed, Mar 02, 2016 at 09:23:34AM +0000, Rowland penny wrote: > > OK, here is the output: > > Can you try the attached (UNTESTED!) patch? > > Thanks, Gna, should have at least compiled it.... New patch :-) Volker -- SerNet GmbH, Bahnhofsallee 1b, 37081 Göttingen phone: +49-551-370000-0, fax:

...sswd file by the users that exist in the system's /etc/passwd file. I was relativly pleased to learn that it does, but only does so by virtue of core dumping when you try to do it. Here's why: (Source code is from samba 2.0.3 downloaded yesterday) Note that at swat.c:635, we are calling local_password_change with trust_account = False. Then, inside local_password_change (which in the case that is core dumping is being called by clicking 'add user' with a username not in /etc/passwd) you check to see if the user exists; but only if trust_account is true. I guess this is fine; I don't know...

...sswd file by the users that exist in the system's /etc/passwd file. I was relativly pleased to learn that it does, but only does so by virtue of core dumping when you try to do it. Here's why: (Source code is from samba 2.0.3 downloaded yesterday) Note that at swat.c:635, we are calling local_password_change with trust_account = False. Then, inside local_password_change (which in the case that is core dumping is being called by clicking 'add user' with a username not in /etc/passwd) you check to see if the user exists; but only if trust_account is true. I guess this is fine; I don't know...

This patch allows net rpc vampire to create accounts in the same way that smbpasswd does, i.e. it will attempt to use the appropriate account creation function for the backend in use. From reading the comments on the top of the local_password_change function, either I shouldn't be going this route or local_password_change is due for some sort of change in status. The only major caveat I've found is that because net rpc vampire does a getpwnam_alloc() immediately after account creation, which fails in my case because I'm using ldap...

...c_array (util.c:754) ==2405== by 0x5728B31: smbldap_set_mod_internal (smbldap.c:325) ==2405== by 0x57290B6: smbldap_set_mod (smbldap.c:392) ==2405== by 0x4A9FC2B: ldapsam_create_user (pdb_ldap.c:5304) ==2405== by 0x4ABB376: pdb_create_user (pdb_interface.c:542) ==2405== by 0x4AACBF9: local_password_change (passdb.c:745) ==2405== by 0x10B29B: password_change (smbpasswd.c:264) ==2405== by 0x10BA77: process_root (smbpasswd.c:466) ==2405== by 0x10BF9C: main (smbpasswd.c:627) ==2405== ==2405== Invalid read of size 4 ==2405== at 0x56AB548: ldap_mods_free (in /usr/lib/i386-linux-gnu/libldap_r...

On 01/03/16 21:35, Garming Sam wrote: > Hi Rowland, > > This new segfault seems unrelated to the previous one. It's probably > something like a double free, which typically shouldn't be that hard to > fix. If you try running the tool under valgrind, it should provide > enough information to fix the issue. > > > Cheers, > > Garming > Who or what is

...tially have to add 8000 users from a constantly changing database. Is there any workaround for this? I've noticed in the source that the check for this is done in passdb/passdb.c approx line 947 if (!NT_STATUS_IS_OK(pdb_init_sam_new(&sam_pass, user_name, 0))) But this is in the function local_password_change() -- If this is modifying the smbpasswd database, why would it need to check /etc/passwd? Is this just a sanity check, or do I have my samba configs incorrect? Call me naive, but could I just comment out this section of code and see what happens? Thanks for any help in advance, --Cal Heldenbra...

...! Failed to modify entry for user ZTEST. root at uuu:/# uname -a Linux uuu 6.5.0-14-generic #14~22.04.1-Ubuntu SMP... I checked samba code then I found the problem is order of gerpwnam call in Get_Pwnam_internals(). Lowercase first order will get domain user (UUQ/ztest) but local account(ZTEST). local_password_change() pdb_getsampwnam() smbpasswd_getsampwnam() build_sam_account() Get_Pwnam_alloc() Get_Pwnam_internals() <-- turn "ZTEST" to "ztest" getpwnam_alloc_cached() getpwnam() (nss_winbinadd) winbindd_getpwnam_send() pars...

...root at uuu:/# uname -a > Linux uuu 6.5.0-14-generic #14~22.04.1-Ubuntu SMP... > > > I checked samba code then I found the problem is order of gerpwnam > call in Get_Pwnam_internals(). Lowercase first order will get domain > user (UUQ/ztest) but local account(ZTEST). > > local_password_change() > pdb_getsampwnam() > smbpasswd_getsampwnam() > build_sam_account() > Get_Pwnam_alloc() > Get_Pwnam_internals() <-- turn "ZTEST" to "ztest" > getpwnam_alloc_cached() > getpwnam() > (nss_winbinadd) >...

...nux uuu 6.5.0-14-generic #14~22.04.1-Ubuntu SMP... > > > > > > I checked samba code then I found the problem is order of gerpwnam > > call in Get_Pwnam_internals(). Lowercase first order will get domain > > user (UUQ/ztest) but local account(ZTEST). > > > > local_password_change() > > pdb_getsampwnam() > > smbpasswd_getsampwnam() > > build_sam_account() > > Get_Pwnam_alloc() > > Get_Pwnam_internals() <-- turn "ZTEST" to "ztest" > > getpwnam_alloc_cached() > > getpwnam() >...

...EXC_BAD_ACCESS (0x0001) >Codes: KERN_PROTECTION_FAILURE (0x0002) at 0x000000df > >Thread 0 Crashed: > #0 0x0007c3c0 in dos_struni2 > #1 0x000182a4 in E_md4hash > #2 0x0001832c in nt_lm_owf_gen > #3 0x00031ef0 in pdb_set_plaintext_passwd > #4 0x000311b0 in local_password_change > #5 0x0000596c in change_password > #6 0x00005c40 in chg_passwd > #7 0x00005ea8 in passwd_page > #8 0x00006704 in main > #9 0x00002124 in _start > #10 0x00001f54 in start The browser returns a "document contained no data." Trying to work with the use...

...not be changed into a machine account but stays as it was before, a user account. This is a little bit confusing because trying to add a machine to the domain failes with "Specified user already exists". I appended a patch which could solve the problem. It changes passdb.c to assure that local_password_change always honors the LOCAL_TRUST_ACCOUNT flag. The current cvs only uses it when adding an account. It's not a bug but an annoying behaviour. But why do you want such stupid things? Changing a user account to a machine account doesn't make sense to me? Well we use it to create a dummy acco...

...1-Ubuntu SMP... > > > > > > > > > I checked samba code then I found the problem is order of gerpwnam > > > call in Get_Pwnam_internals(). Lowercase first order will get > > > domain user (UUQ/ztest) but local account(ZTEST). > > > > > > local_password_change() > > > pdb_getsampwnam() > > > smbpasswd_getsampwnam() > > > build_sam_account() > > > Get_Pwnam_alloc() > > > Get_Pwnam_internals() <-- turn "ZTEST" to "ztest" > > > getpwnam_alloc_cached() >...

...s_object () #4 0xef69b2d8 in xdr_array () #5 0xef69f528 in xdr_nis_result () #6 0xef69bb5c in xdr_free () #7 0xef69f854 in nis_freeresult () #8 0x38a48 in add_nisp21pwd_entry () #9 0x390ec in add_nisppwd_entry () #10 0x35900 in add_smbpwd_entry () #11 0x39258 in add_new_user () #12 0x393d8 in local_password_change () #13 0x2322c in password_change () #14 0x236cc in process_root () #15 0x23b70 in main ()

...rc = 0 __FUNCTION__ = "ldapsam_create_user" #4 0xb7d6346f in pdb_create_user (mem_ctx=0x8001e888, name=0x800081c0 <user_name> "ldap09", flags=16, rid=0xbffffb58) at ../source3/passdb/pdb_interface.c:542 pdb = 0x80027e28 #5 0xb7d54402 in local_password_change ( user_name=0x800081c0 <user_name> "ldap09", local_flags=577, new_passwd=0x800118b8 "p4ssword", pp_err_str=0xbffffbb8, pp_msg_str=0xbffffbbc) at ../source3/passdb/passdb.c:745 tosctx = 0x8001e888 sam_pass = 0x80028100 acb = 16...

On Fri, Mar 04, 2016 at 04:16:44PM +0000, Rowland penny wrote: > On 04/03/16 09:58, Volker Lendecke wrote: > >On Fri, Mar 04, 2016 at 10:53:17AM +0100, Volker Lendecke wrote: > >>On Wed, Mar 02, 2016 at 09:23:34AM +0000, Rowland penny wrote: > >>>OK, here is the output: > >>Can you try the attached (UNTESTED!) patch? > >> > >>Thanks, >

...> > > > > > > > I checked samba code then I found the problem is order of gerpwnam > > > > call in Get_Pwnam_internals(). Lowercase first order will get > > > > domain user (UUQ/ztest) but local account(ZTEST). > > > > > > > > local_password_change() > > > > pdb_getsampwnam() > > > > smbpasswd_getsampwnam() > > > > build_sam_account() > > > > Get_Pwnam_alloc() > > > > Get_Pwnam_internals() <-- turn "ZTEST" to "ztest" > > > >...

...net ads testjoin". * BUG 6253: Use correct value for password expiry calculation in pam_winbind. * BUG 6305: Correctly prompt for a password when a username was given. * BUG 6451: net/libnetapi user rename using wrong access bits. * BUG 6458: Fix uninitialized variable in local_password_change(). * BUG 6465: Fix enumeration of empty aliases. o Volker Lendecke <vl@samba.org> * BUG 4699: Remove pidfile on clean shutdown. * BUG 6449: ''net rap user add'' crashes without -C option. o David Markey <admin@dmarkey.com> * BUG 6328: Add support...

...net ads testjoin". * BUG 6253: Use correct value for password expiry calculation in pam_winbind. * BUG 6305: Correctly prompt for a password when a username was given. * BUG 6451: net/libnetapi user rename using wrong access bits. * BUG 6458: Fix uninitialized variable in local_password_change(). * BUG 6465: Fix enumeration of empty aliases. o Volker Lendecke <vl@samba.org> * BUG 4699: Remove pidfile on clean shutdown. * BUG 6449: ''net rap user add'' crashes without -C option. o David Markey <admin@dmarkey.com> * BUG 6328: Add support...