[Samba] samba as PDC, win 2003 as bdc
Hello, is it possible to add a wind 2003 machine as BDC to a samba domain. Is documentation documentation available ? Thanks Andreas -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] samba as PDC, win 2003 as bdc
is it possible to add a wind 2003 machine as BDC to a samba domain. The Samba3-HOWTO, Chapter 5, contains the following: « Samba-3 cannot participate in true SAM replication and is therefore not able to employ precisely the same protocols used by MS Windows NT4. A Samba-3 BDC will not create SAM update delta files. It will not interoperate with a PDC (NT4 or Samba) to synchronize the SAM from delta files that are held by BDCs. Samba-3 cannot function as a BDC to an MS Windows NT4 PDC, and Samba-3 cannot function correctly as a PDC to an MS Windows NT4 BDC. Both Samba-3 and MS Windows NT4 can function as a BDC to its own type of PDC. » Is documentation documentation available ? Yes. You can start with: The Official Samba 3.5.x HOWTO and Reference Guide Samba3 By Example Both come with the Samba packages, in HTML and PDF formats, or can be viewed online at www.samba.org. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] StorNext CVFS
Morning All! Anyone ever tried exporting a StorNext CVFS filesystem from a Linux box??? I¹ve got this Samba server (3.5.6) running on CentOS 5.4 and it¹s working fine, exporting ext3, nfs and an IBM GPFS filesystem just fine. So I know Samba is good an my configuration is working. I tried to add the exportation of a StorNext CVFS volume and that doesn¹t work. All the other volumes still work just fine, but when I try to connect to this new volume I get ... Is not accessible, you might not have permission, etc ... The parameter is incorrect. Hoping someone else might have gone down this road before... For those who haven¹t, Quantum StorNext is a clustered/HA filesystem that really sits somewhere between the category of vaporware and smokeware! It¹s very poorly documented and their support sucks standard answer is ³you can¹t do that² for just about everything. Now my StorNext server is cheerfully exporting his filesystem natively to Windows, Linux and Mac clients, so I know he¹s working fine. I also know this is most likely not a Samba problem, but with Quantum¹s ³you can¹t do that² attitude, I¹m hoping someone here has some knowledge that the documentation missed! Thanks, -brian -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] StorNext CVFS
On Sat, Nov 13, 2010 at 10:39:30AM -0800, Brian McGrew wrote: Morning All! Anyone ever tried exporting a StorNext CVFS filesystem from a Linux box??? I¹ve got this Samba server (3.5.6) running on CentOS 5.4 and it¹s working fine, exporting ext3, nfs and an IBM GPFS filesystem just fine. So I know Samba is good an my configuration is working. I tried to add the exportation of a StorNext CVFS volume and that doesn¹t work. All the other volumes still work just fine, but when I try to connect to this new volume I get ... Is not accessible, you might not have permission, etc ... The parameter is incorrect. That should work fine, if that file system is anything like posix. Do you have a debug level 10 log of smbd? You might also want to strace smbd on that file system. Volker -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] getting error with setfacl
On November 3, 2010, James D. Parra jam...@musicreports.com wrote: Hello, I have joined the Linux server (Suse 11.2) to the Windows domain (win2003) and users can login to the server using their window's domain credentials. Also can view all of the domain groups using 'wbinfo -g', however when I try to set the acls on a local dir' I get the follow error; #setfacl -m g:DOMAIN\\groupname:r /tmp/testacl setfacl: Option -m: Invalid argument near character 3 The same error occurs if I use a domain user, although I can add a local Linux user or group. What am I overlooking? setfacl has no idea about smb group or user names. You probably need to setup winbindd/nsswitch to map domain groups and users to local groups and users before you can assign acls. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] NMB is flooding logfile
The hermes bags historical http://www.earlshop.com mens watches past news charcoal absorbing and arresting till date, and the designs with the http://www.sonbags.com designer handbags accoutrements from Hermes, Paris, seem to become an afflatus for abounding clean and accessible http://www.earlshop.com rolex watches look designers, and in addition for those alive accurately in the backpack market. The http://www.earlshop.com/Watches-Breitling.html breitling watches Hermes accoutrements may be calmly start at any from the official Hermes outlets or with authorised sellers of Hermes bags. You will find furthermore http://www.sonbags.com/gucci.html gucci bags abounding dealers of Hermes accoutrements who settle for web sites of their own, and advertise these accoutrements on the internet with http://www.sonbags.com/gucci-handbags.html gucci handbags considerable delivers of chargeless aircraft beyond the world, and some abundant discounts as well. You will discover abounding altered colour alternatives obtainable, and in excess of the a long time, the affectionate of abstracts referred http://www.earlshop.com/Watches-Franck-Muller.html Franck Muller to as to architecture the Kelly handbag settle for in addition changed, giving the archetypal structure abounding clean versions. It is possible to acquisition out about all these accoutrements at the Hermes web site, or at any of their retailers. So in the event you as well are absorbed with the news and http://www.sonbags.com/balenciaga.html balenciaga historical past abaft the authoritative of your Kelly Travelling bag, once again go and get your self a person of these accoutrements right now! 138. Search Louis Vuitton http://www.earlshop.com/Watches-Gucci.html gucci watches Bags on the internet -- View this message in context: http://samba.2283325.n4.nabble.com/NMB-is-flooding-logfile-tp2456892p3028087.html Sent from the Samba - General mailing list archive at Nabble.com. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] Mount error 13
The hermes bags historical http://www.earlshop.com mens watches past news charcoal absorbing and arresting till date, and the designs with the http://www.sonbags.com designer handbags accoutrements from Hermes, Paris, seem to become an afflatus for abounding clean and accessible http://www.earlshop.com rolex watches look designers, and in addition for those alive accurately in the backpack market. The http://www.earlshop.com/Watches-Breitling.html breitling watches Hermes accoutrements may be calmly start at any from the official Hermes outlets or with authorised sellers of Hermes bags. You will find furthermore http://www.sonbags.com/gucci.html gucci bags abounding dealers of Hermes accoutrements who settle for web sites of their own, and advertise these accoutrements on the internet with http://www.sonbags.com/gucci-handbags.html gucci handbags considerable delivers of chargeless aircraft beyond the world, and some abundant discounts as well. You will discover abounding altered colour alternatives obtainable, and in excess of the a long time, the affectionate of abstracts referred http://www.earlshop.com/Watches-Franck-Muller.html Franck Muller to as to architecture the Kelly handbag settle for in addition changed, giving the archetypal structure abounding clean versions. It is possible to acquisition out about all these accoutrements at the Hermes web site, or at any of their retailers. So in the event you as well are absorbed with the news and http://www.sonbags.com/balenciaga.html balenciaga historical past abaft the authoritative of your Kelly Travelling bag, once again go and get your self a person of these accoutrements right now! 138. Search Louis Vuitton http://www.earlshop.com/Watches-Gucci.html gucci watches Bags on the internet -- View this message in context: http://samba.2283325.n4.nabble.com/Mount-error-13-tp2441308p3028089.html Sent from the Samba - General mailing list archive at Nabble.com. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] No data on inotify fd error
The hermes bags historical http://www.earlshop.com mens watches past news charcoal absorbing and arresting till date, and the designs with the http://www.sonbags.com designer handbags accoutrements from Hermes, Paris, seem to become an afflatus for abounding clean and accessible http://www.earlshop.com rolex watches look designers, and in addition for those alive accurately in the backpack market. The http://www.earlshop.com/Watches-Breitling.html breitling watches Hermes accoutrements may be calmly start at any from the official Hermes outlets or with authorised sellers of Hermes bags. You will find furthermore http://www.sonbags.com/gucci.html gucci bags abounding dealers of Hermes accoutrements who settle for web sites of their own, and advertise these accoutrements on the internet with http://www.sonbags.com/gucci-handbags.html gucci handbags considerable delivers of chargeless aircraft beyond the world, and some abundant discounts as well. You will discover abounding altered colour alternatives obtainable, and in excess of the a long time, the affectionate of abstracts referred http://www.earlshop.com/Watches-Franck-Muller.html Franck Muller to as to architecture the Kelly handbag settle for in addition changed, giving the archetypal structure abounding clean versions. It is possible to acquisition out about all these accoutrements at the Hermes web site, or at any of their retailers. So in the event you as well are absorbed with the news and http://www.sonbags.com/balenciaga.html balenciaga historical past abaft the authoritative of your Kelly Travelling bag, once again go and get your self a person of these accoutrements right now! 138. Search Louis Vuitton http://www.earlshop.com/Watches-Gucci.html gucci watches Bags on the internet -- View this message in context: http://samba.2283325.n4.nabble.com/No-data-on-inotify-fd-error-tp2452847p3028091.html Sent from the Samba - General mailing list archive at Nabble.com. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] idmap uid range missing or invalid
The hermes bags historical http://www.earlshop.com mens watches past news charcoal absorbing and arresting till date, and the designs with the http://www.sonbags.com designer handbags accoutrements from Hermes, Paris, seem to become an afflatus for abounding clean and accessible http://www.earlshop.com rolex watches look designers, and in addition for those alive accurately in the backpack market. The http://www.earlshop.com/Watches-Breitling.html breitling watches Hermes accoutrements may be calmly start at any from the official Hermes outlets or with authorised sellers of Hermes bags. You will find furthermore http://www.sonbags.com/gucci.html gucci bags abounding dealers of Hermes accoutrements who settle for web sites of their own, and advertise these accoutrements on the internet with http://www.sonbags.com/gucci-handbags.html gucci handbags considerable delivers of chargeless aircraft beyond the world, and some abundant discounts as well. You will discover abounding altered colour alternatives obtainable, and in excess of the a long time, the affectionate of abstracts referred http://www.earlshop.com/Watches-Franck-Muller.html Franck Muller to as to architecture the Kelly handbag settle for in addition changed, giving the archetypal structure abounding clean versions. It is possible to acquisition out about all these accoutrements at the Hermes web site, or at any of their retailers. So in the event you as well are absorbed with the news and http://www.sonbags.com/balenciaga.html balenciaga historical past abaft the authoritative of your Kelly Travelling bag, once again go and get your self a person of these accoutrements right now! 138. Search Louis Vuitton http://www.earlshop.com/Watches-Gucci.html gucci watches Bags on the internet -- View this message in context: http://samba.2283325.n4.nabble.com/idmap-uid-range-missing-or-invalid-tp2451200p3028092.html Sent from the Samba - General mailing list archive at Nabble.com. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] Windows XP can't join to Domain by Samba
Hi everbody, Two monts ago, i had got problem with my Domain build by Samba 3.0.24 by Debian Etch. I had configured DNS, WINS and SAMBA domain and when i set up IP addres gateway and dns on Windows XP i could connect to DOMAIN But now when i setup this and push join computer to domain i got info that domain is unavailable. I Check log in Windows XP Systemdir\windows\debug\NetSetup.log And I found error message: NetpDsGetDcName: failed to find a DC in the specified domain So i tri reconfigure DNS - its sounds like good issues - but nothing Stations what i hade join to domain work normaly - they logons to domiain the use share disk provide by samba, but when i want change password i had got erro that domain is unavailable. One options is setup on each computer WINS server in network setting, or fix this problem. [global] display charset = UTF-8 name resolve order = wins hosts bcast lmhosts domain master = Yes passwd chat = *Enter\snew\s*\spassword:* %n\n *Retype\snew\s*\spassword:* %n\n *password\supdated\ssuccessfully* . admin users = root administrator @it posix locking = no time server = Yes passwd program = /usr/bin/smbpasswd %u dns proxy = no netbios name = server string locking = no printing = bsd logon script = logon.vbs dos charset = CP852 local master = Yes workgroup = DOMAIN server signing = Auto printcap name = /dev/null security = user add machine script = /usr/sbin/useradd -s /bin/false -d /dev/null '%u' delete user script = /usr/sbin/userdel -r '%u' max log size = 1000 server schannel = Auto log level = 2 log file = /var/log/samba/%m load printers = No add group script = /usr/sbin/groupadd '%g' socket options = TCP_NODELAY delete group script = /usr/sbin/groupdel '%g' add user to group script = /usr/sbin/usermod -G '%g' '%u' logon drive = H: deadtime = 15 hide dot files = yes interfaces = eth0 eth2 lo encrypt passwords = Yes logon home = \\%L\profiles\%u keepalive = 300 wins proxy = no wins support = Yes server string = ATENTO unix password sync = No logon path = \\%L\profiles\%u acl compatibility = Auto add user script = /usr/sbin/useradd -d /dev/null -g 100 -s /bin/false -m %u preferred master = Yes unix charset = LOCALE bind interfaces only = Yes domain logons = Yes pam password change = Yes -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] krb5.conf Second domain controller
Hello I did net/vampire with success. 1) I noticed that second domain controller not listening on port 3268. Is this normal ? ( /usr/local/samba/private/dns_update_list file suggest to have record for gc service for second domain controller ) 2) Should I add record for dc2 in /etc/krb5.conf on both servers ? I suggest [realms] TEST.LOCAL = { kdc = dc1.test.local:88 *kdc = dc2.test.local:88* admin_server = dc1.test.local:749 *admin_server = dc2.test.local:749* default_domain = test.local -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] Win98 client samba 3.3.8-52.el5_5.2 on ext4
Hello, I get a General failure reading drive i Abort, Retry, Fail? When I try to access some files, not all files do this. On the CentOS5.5 (ext4) Side: I can copy the file and if it has the same name, same error. If I copy the file to another name I can read it. From the DOS client, if I type the file, type file's contents are displayed followed by the error message. copy fails. This is a new installation where files are rsynched from another server running CentOs4.8. The other servers that receive these files are running CentOS4.8 ext3 for the file system and smb version 3.0.28.-0.el4.9 and have no problems. Also, my win7 system has no problems either. This is a guest acount with no password. The guest account does have an account and the file has appropriate guest permissions the the same group ownership. file permissions are rw-r--r-- Samba configuration is like (I know its not secure, I'll worry about that later): workgroup = SOMETHING server string = SOMETHING netbios name = SOMETHING Else netbios aliases = another name for something else security = share passdb backend = smbpasswd os level = 65 preferred master = yes guest account = tech guest ok = yes null passwords = yes wide links = yes wins support = yes [xx_files] comment = Test Files path = /mnt/xx_files read only = no create mask = 0774 directory mask = 0755 locking = no browseable = yes I noticed in the log.smbd file when the dos client initially connects: pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2010/11/08 19:01:35, 3] lib/privileges.c:get_privileges(63) get_privileges: No privileges assigned to SID [S-1-22-1-500] [2010/11/08 19:01:35, 3] lib/privileges.c:get_privileges(63) get_privileges: No privileges assigned to SID [S-1-22-2-500] [2010/11/08 19:01:35, 3] lib/privileges.c:get_privileges(63) get_privileges: No privileges assigned to SID [S-1-5-2] [2010/11/08 19:01:35, 3] lib/privileges.c:get_privileges(63) get_privileges: No privileges assigned to SID [S-1-5-32-546] [2010/11/08 19:01:35, 3] smbd/sec_ctx.c:push_sec_ctx(224) push_sec_ctx(0, 0) : sec_ctx_stack_ndx = 1 [2010/11/08 19:01:35, 3] smbd/uid.c:push_conn_ctx(440) push_conn_ctx(0) : conn_ctx_stack_ndx = 0 [2010/11/08 19:01:35, 3] smbd/sec_ctx.c:set_sec_ctx(324) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 1 [2010/11/08 19:01:35, 3] smbd/sec_ctx.c:pop_sec_ctx(432) pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2010/11/08 19:01:35, 3] smbd/sec_ctx.c:push_sec_ctx(224) push_sec_ctx(0, 0) : sec_ctx_stack_ndx = 1 [2010/11/08 19:01:35, 3] smbd/uid.c:push_conn_ctx(440) push_conn_ctx(0) : conn_ctx_stack_ndx = 0 [2010/11/08 19:01:35, 3] smbd/sec_ctx.c:set_sec_ctx(324) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 1 [2010/11/08 19:01:35, 3] smbd/sec_ctx.c:pop_sec_ctx(432) pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2010/11/08 19:01:35, 3] smbd/sec_ctx.c:push_sec_ctx(224) push_sec_ctx(0, 0) : sec_ctx_stack_ndx = 1 [2010/11/08 19:01:35, 3] smbd/uid.c:push_conn_ctx(440) push_conn_ctx(0) : conn_ctx_stack_ndx = 0 [2010/11/08 19:01:35, 3] smbd/sec_ctx.c:set_sec_ctx(324) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 1 [2010/11/08 19:01:35, 3] smbd/sec_ctx.c:pop_sec_ctx(432) pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2010/11/08 19:01:35, 3] smbd/service.c:make_connection_snum(879) Connect path is '/mnt/ae_files' for service [ae_files] [2010/11/08 19:01:35, 3] smbd/vfs.c:vfs_init_default(96) Initialising default vfs hooks [2010/11/08 19:01:35, 3] smbd/vfs.c:vfs_init_custom(130) Initialising custom vfs hooks from [/[Default VFS]/] [2010/11/08 19:01:35, 3] smbd/sec_ctx.c:set_sec_ctx(324) setting sec ctx (500, 500) - sec_ctx_stack_ndx = 0 [2010/11/08 19:01:35, 1] smbd/service.c:make_connection_snum(1119) ___10.20.4.248 (:::10.20.4.248) connect to service ae_files initially as user tech (uid=500, gid=500) (pid 18556) And when I type the file I get: [2010/11/08 19:02:57, 3] smbd/process.c:process_smb(1576) Transaction 6 of length 98 (0 toread) [2010/11/08 19:02:57, 3] smbd/process.c:switch_message(1393) switch message SMBopen (pid 18559) conn 0x2aef08498fa0 [2010/11/08 19:02:57, 3] smbd/sec_ctx.c:set_sec_ctx(324) setting sec ctx (500, 500) - sec_ctx_stack_ndx = 0 [2010/11/08 19:02:57, 3] smbd/dosmode.c:unix_mode(124) unix_mode(utils/fits32/SCRIPTS/GEN_TEST/RELEASED/FITS/FITS.INI) returning 0764 [2010/11/08 19:02:57, 2] smbd/open.c:open_file(551) tech opened file utils/fits32/SCRIPTS/GEN_TEST/RELEASED/FITS/FITS.INI read=Yes write=No (numopen=1) [2010/11/08 19:02:57, 3] smbd/process.c:process_smb(1576) Transaction 7 of length 52 (0 toread) [2010/11/08 19:02:57, 3] smbd/process.c:switch_message(1393) switch message SMBread (pid 18559) conn 0x2aef08498fa0 [2010/11/08 19:02:57, 3] smbd/reply.c:reply_read(3193) read fnum=12835 num=524 nread=524 [2010/11/08 19:02:57, 3] smbd/process.c:process_smb(1576) Transaction 8 of
Re: [Samba] Problems with ACL jumbo patch
I have attached the config.log file, and a capture of the messages from making that I called make.log. I have been able to build working binaries from the git, though I run into problems when I tried to patch git branch and compile. -Original Message- From: Michael Wood [mailto:esiot...@gmail.com] Sent: Monday, November 08, 2010 3:35 AM To: Derek Lewis Cc: Miguel Medalha; samba@lists.samba.org Subject: Re: [Samba] Problems with ACL jumbo patch On 8 November 2010 10:31, Derek Lewis dle...@mtu.edu wrote: I have been able to get the unpatched versions to compile from git successfully, though not with the patch implemented. I followed these steps to compile Samba 3.5.6 with the patch: 1. sudo git clone git://git.samba.org/samba.git samba102510 2. sudo wget http://samba.org/~jra/samba-3-5-x-acl-jumbo-patch.tgz 3. sudo tar -xvf samba-3-5-x-acl-jumbo-patch.tgz 4. cd samba102510 5. sudo git checkout -b my_branch release-3-5-6 6. sudo git am -3 ../samba_patches/samba-3-5-x-acl-jumbo-patch/*.patch 7. cd source3 8. sudo ./autogen.sh 9. sudo ./configure.developer --prefix=/usr/local/samba35 10. sudo make [The build failed on the error 'cli_krb5_get_ticket', in the function 'spnego_gen_negTokenTarg'. Make: *** [libsmb/clispnego.o] Error 1] You should generally not compile things as root. You do not need the sudos above. Only when you make install should you need it, because then it will need to write to /usr/local/samba35 where a normal user would not have write access. This is, however, not the cause of your problems. I have just tried the above and it compiled successfully for me. What is the full output of the errors you get? It should provide more information than you have quoted above. I am still learning how to use git, I wondered if the error messages are the result of missing a step. I did not explicitly commit the changes, do I need to update the index or pull in remote files? What you have above should work. git am automatically commits the patches. I am searching for references to this make error though I have not found much, except the library: libldap2-dev. I installed it and the problem persists. Could you recommend other debug options? Post the full error and maybe someone will be able to figure out what the problem is. Are you able to build 3.5.6 without the patches applied? -- Michael Wood esiot...@gmail.com -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] Samba for z/OS 1.10
Hello everybody, is there any samba version available for download, which runs on IBM MVS, i.e. IBM z/OS 1.10 ...? if not, is there a possibility to compile the source code myself for the target system? with kind regards, M. The information transmitted may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] aio and roaming profiles
Dear list, has anyone expierienced syncronisation issues with roaming profiles and aio enabled? Is there a way to disable aio for profiles? Or has somebody a good knowledge playing with samba and aio? Thank you and best regards Ralf -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] StorNext CVFS
On 11/13/10 11:30 AM, Volker Lendecke volker.lende...@sernet.de wrote: On Sat, Nov 13, 2010 at 10:39:30AM -0800, Brian McGrew wrote: Morning All! Anyone ever tried exporting a StorNext CVFS filesystem from a Linux box??? I¹ve got this Samba server (3.5.6) running on CentOS 5.4 and it¹s working fine, exporting ext3, nfs and an IBM GPFS filesystem just fine. So I know Samba is good an my configuration is working. I tried to add the exportation of a StorNext CVFS volume and that doesn¹t work. All the other volumes still work just fine, but when I try to connect to this new volume I get ... Is not accessible, you might not have permission, etc ... The parameter is incorrect. That should work fine, if that file system is anything like posix. Do you have a debug level 10 log of smbd? You might also want to strace smbd on that file system. There is a level 10 log attached. I stopped samba, cleared the logs, restarted samba, tried to connect, stopped samba and grabbed the logs. I'm not having very good luck getting strace to attach to the smbd process??? StorNext CVFS is supposed to be a posix filesystem, and in Unix it appears to be posix - but I can't say for sure! Maybe I just don't know what I'm looking for in the logs! -brian -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] Problems with ACL jumbo patch
Hi On 10 November 2010 10:22, Derek Lewis dle...@mtu.edu wrote: Okay, I search the apt repositories and found three Kerberos libraries: libpam-krb5, krb5-auth-dialog and libkrb5-dev. libkrb5-dev is the one you want. After installing, I configured my build with autogen and configure.developer as before. Attempting to make the binaries resulted in the same cli_krb5_get_ticket' error as before. Try making sure you're starting from a clean checkout. i.e. get rid of any previous build attempts. The following should work assuming you're in the checked out directory and on the right branch: $ git reset --hard HEAD $ git clean -dxf Then do the autogen and configure again. After running configure.developer, do this (from the source3 directory): $ grep 'HAVE_KRB5\' include/config.h It should print out: #define HAVE_KRB5 1 since you now have the libkrb5-dev package installed. I looked through the options for configure.developer, can I disable Kerberos in the configure step and bypass the problem? I don't know. Jeremy, any idea why Derek is getting link errors with your ACL jumbo patch for 3.5.x despite the HAVE_KRB5 check in source3/libsmb/clikrb5.c? -Original Message- From: Michael Wood [mailto:esiot...@gmail.com] Sent: Tuesday, November 09, 2010 3:52 AM To: Derek Lewis Cc: Miguel Medalha; samba@lists.samba.org; Jeremy Allison Subject: Re: [Samba] Problems with ACL jumbo patch On 9 November 2010 11:20, Derek Lewis dle...@mtu.edu wrote: I have attached the config.log file, and a capture of the messages from making that I called make.log. OK, then it does seem to have something to do with Kerberos. The cli_krb5_get_ticket function is defined in source3/libsmb/clikrb5.c and if you don't have HAVE_KRB5 defined, then it's supposed to do this: /* this saves a few linking headaches */ int cli_krb5_get_ticket(const char *principal, time_t time_offset, DATA_BLOB *ticket, DATA_BLOB *session_key_krb5, uint32 extra_ap_opts, const char *ccname, time_t *tgs_expire, const char *impersonate_princ_s) { DEBUG(0,(NO KERBEROS SUPPORT\n)); return 1; } but for some reason that's not happening for you, so you get link errors. The solution to your problem is, of course, to install the Kerberos libs (either MIT or Heimdal). Try apt-get install libkrb5-dev. -- Michael Wood esiot...@gmail.com -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] Samba for z/OS 1.10
On 10/11/2010 16:54, martin.h...@helvetia.ch wrote: Hello everybody, is there any samba version available for download, which runs on IBM MVS, i.e. IBM z/OS 1.10 ...? if not, is there a possibility to compile the source code myself for the target system? Which samba ? I guess we do not have a prepackaged version for Z-OS Matthieu -- Matthieu Patou Samba Teamhttp://samba.org Private repo http://git.samba.org/?p=mat/samba.git;a=summary -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] samba as PDC, win 2003 as bdc
Hi On 13 November 2010 18:32, Andreas Moroder andreas.moro...@gmx.net wrote: Hello, is it possible to add a wind 2003 machine as BDC to a samba domain. Is documentation documentation available ? Samba 3 can work as an NT-style PDC, but does not support the native PDC-BDC replication protocol, so as Miguel said you can't have a Samba 3 PDC and a Windows 2003 BDC. (You also can't have Samba 3 as a BDC.) Samba 4 works as an Active Directory domain controller and supports replication to other DCs using the native Active Directory replication protocols. So it works with other Samba 4 DCs and also Windows DCs. It is, however, still alpha software, so if you find something doesn't work, post to the samba-technical mailing list. See the Samba 4 HOWTO for details: http://wiki.samba.org/index.php/Samba4/HOWTO See also: http://wiki.samba.org/index.php/Samba4/HOWTO/Join_a_domain_as_a_DC http://wiki.samba.org/index.php/Samba4/FAQ http://wiki.samba.org/index.php/Samba4/videos#Demo4_:_dcpromo -- Michael Wood esiot...@gmail.com -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] winbind sometimes does not resolve sid to a name
Hi Shirish, Shirish Pargaonkar wrote: On Mon, Nov 8, 2010 at 1:47 PM, Jeremy Allison j...@samba.org wrote: On Mon, Nov 08, 2010 at 01:21:30PM -0600, Shirish Pargaonkar wrote: Sometimes a group sid does not get resolved to its name. Is this a settings problem? Looks like winbind deamon went dormant for a while and then woke up? I am using interface wbcLookupSid provided by the library libwbclient.so for resolving sids to names. These are the winbind related parameters in /etc/samba/smb.conf Not enough information for useful debugging. What do the winbindd logs say ? ps -eaf | grep winbind root 20085 1 0 14:03 ?00:00:00 /usr/sbin/winbindd -D root 20086 20085 0 14:03 ?00:00:00 /usr/sbin/winbindd -D root 20089 20085 0 14:03 ?00:00:00 /usr/sbin/winbindd -D Cleared /var/log/samba/winbindd.log just before issueing command getcifsacl which could not resolve the group SID winbindd.log attached. not really. :-) Cheers - Michael pgpmiRUIOzSAA.pgp Description: PGP signature -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] Samba for z/OS 1.10
On Wed, Nov 10, 2010 at 02:54:48PM +0100, martin.h...@helvetia.ch wrote: is there any samba version available for download, which runs on IBM MVS, i.e. IBM z/OS 1.10 ...? There used to be somthing that claimed to run on MVS ages (and I mean AGES, my rough guess would be 10 years) ago. if not, is there a possibility to compile the source code myself for the target system? Sure, if you have something vaguely looking like Posix with a C compiler, then the chances are not too bad. I'd love to see Samba running on z/OS, but I fear that without access to such a box by someone knowledgeable with the Samba build process you will not get very far. I would be happy to assist you to get it up and running there. Volker -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] winbind sometimes does not resolve sid to a name
On Sat, Nov 13, 2010 at 4:52 PM, Michael Adam ob...@samba.org wrote: Hi Shirish, Shirish Pargaonkar wrote: On Mon, Nov 8, 2010 at 1:47 PM, Jeremy Allison j...@samba.org wrote: On Mon, Nov 08, 2010 at 01:21:30PM -0600, Shirish Pargaonkar wrote: Sometimes a group sid does not get resolved to its name. Is this a settings problem? Looks like winbind deamon went dormant for a while and then woke up? I am using interface wbcLookupSid provided by the library libwbclient.so for resolving sids to names. These are the winbind related parameters in /etc/samba/smb.conf Not enough information for useful debugging. What do the winbindd logs say ? ps -eaf | grep winbind root 20085 1 0 14:03 ? 00:00:00 /usr/sbin/winbindd -D root 20086 20085 0 14:03 ? 00:00:00 /usr/sbin/winbindd -D root 20089 20085 0 14:03 ? 00:00:00 /usr/sbin/winbindd -D Cleared /var/log/samba/winbindd.log just before issueing command getcifsacl which could not resolve the group SID winbindd.log attached. not really. :-) Cheers - Michael Michael, not sure what is implied. The log is not sufficient? I see two error messages in the log. [2010/11/08 14:32:56, 5] winbindd/winbindd_async.c:lookupsid_recv2(138) lookupsid (forest root) returned an error [2010/11/08 14:32:56, 5] winbindd/winbindd_sid.c:lookupsid_recv(61) lookupsid returned an error -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] winbind sometimes does not resolve sid to a name
On 14 November 2010 01:16, Shirish Pargaonkar shirishpargaon...@gmail.com wrote: On Sat, Nov 13, 2010 at 4:52 PM, Michael Adam ob...@samba.org wrote: Hi Shirish, Shirish Pargaonkar wrote: On Mon, Nov 8, 2010 at 1:47 PM, Jeremy Allison j...@samba.org wrote: On Mon, Nov 08, 2010 at 01:21:30PM -0600, Shirish Pargaonkar wrote: Sometimes a group sid does not get resolved to its name. Is this a settings problem? Looks like winbind deamon went dormant for a while and then woke up? I am using interface wbcLookupSid provided by the library libwbclient.so for resolving sids to names. These are the winbind related parameters in /etc/samba/smb.conf Not enough information for useful debugging. What do the winbindd logs say ? ps -eaf | grep winbind root 20085 1 0 14:03 ?00:00:00 /usr/sbin/winbindd -D root 20086 20085 0 14:03 ?00:00:00 /usr/sbin/winbindd -D root 20089 20085 0 14:03 ?00:00:00 /usr/sbin/winbindd -D Cleared /var/log/samba/winbindd.log just before issueing command getcifsacl which could not resolve the group SID winbindd.log attached. not really. :-) Cheers - Michael Michael, not sure what is implied. The log is not sufficient? No, the mailing list (sometimes) strips attachments. There was no log file attached to your e-mail when I received it. I see two error messages in the log. [2010/11/08 14:32:56, 5] winbindd/winbindd_async.c:lookupsid_recv2(138) lookupsid (forest root) returned an error [2010/11/08 14:32:56, 5] winbindd/winbindd_sid.c:lookupsid_recv(61) lookupsid returned an error -- Michael Wood esiot...@gmail.com -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] Problems with ACL jumbo patch
So far so good...I wanted to thank you, Miguel and the Samba team for all of the help. The krb5 test was successful. The hard reset and make of Samba 3.5.6 was successful though I have not completed the install yet. I have a question, does the git reset/clean commands remove the acl patch? Can I confirm that the patch files were included into the build? From the Samba side, how would I test the patch viability? Derek -Original Message- From: Michael Wood [mailto:esiot...@gmail.com] Sent: Saturday, November 13, 2010 12:43 PM To: Derek Lewis Cc: samba@lists.samba.org; Jeremy Allison; samba-techni...@lists.samba.org Subject: Re: [Samba] Problems with ACL jumbo patch Hi On 10 November 2010 10:22, Derek Lewis dle...@mtu.edu wrote: Okay, I search the apt repositories and found three Kerberos libraries: libpam-krb5, krb5-auth-dialog and libkrb5-dev. libkrb5-dev is the one you want. After installing, I configured my build with autogen and configure.developer as before. Attempting to make the binaries resulted in the same cli_krb5_get_ticket' error as before. Try making sure you're starting from a clean checkout. i.e. get rid of any previous build attempts. The following should work assuming you're in the checked out directory and on the right branch: $ git reset --hard HEAD $ git clean -dxf Then do the autogen and configure again. After running configure.developer, do this (from the source3 directory): $ grep 'HAVE_KRB5\' include/config.h It should print out: #define HAVE_KRB5 1 since you now have the libkrb5-dev package installed. I looked through the options for configure.developer, can I disable Kerberos in the configure step and bypass the problem? I don't know. Jeremy, any idea why Derek is getting link errors with your ACL jumbo patch for 3.5.x despite the HAVE_KRB5 check in source3/libsmb/clikrb5.c? -Original Message- From: Michael Wood [mailto:esiot...@gmail.com] Sent: Tuesday, November 09, 2010 3:52 AM To: Derek Lewis Cc: Miguel Medalha; samba@lists.samba.org; Jeremy Allison Subject: Re: [Samba] Problems with ACL jumbo patch On 9 November 2010 11:20, Derek Lewis dle...@mtu.edu wrote: I have attached the config.log file, and a capture of the messages from making that I called make.log. OK, then it does seem to have something to do with Kerberos. The cli_krb5_get_ticket function is defined in source3/libsmb/clikrb5.c and if you don't have HAVE_KRB5 defined, then it's supposed to do this: /* this saves a few linking headaches */ int cli_krb5_get_ticket(const char *principal, time_t time_offset, DATA_BLOB *ticket, DATA_BLOB *session_key_krb5, uint32 extra_ap_opts, const char *ccname, time_t *tgs_expire, const char *impersonate_princ_s) { DEBUG(0,(NO KERBEROS SUPPORT\n)); return 1; } but for some reason that's not happening for you, so you get link errors. The solution to your problem is, of course, to install the Kerberos libs (either MIT or Heimdal). Try apt-get install libkrb5-dev. -- Michael Wood esiot...@gmail.com -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] winbind - wbinfo problem
Hi Guys, I have configured SAMBA with Windows 2003 AD. But #wbinfo -u and #wbinfo -g does not list the users 1. Domain joined successfully. # net rpc testjoin -U Administrator Join to 'DOMAIN' is OK 2. wbinfo -a works ( User authentication ) # wbinfo -a 'DOMAIN\user' Enter DOMAIN\user's password: plaintext password authentication succeeded Enter DOMAIN\user's password: challenge/response password authentication succeeded 3. wbinfo -u and wbinfo -g does list nothing # wbinfo -u # wbinfo -g # wbinfo -r 'DOMAIN\user' Could not get groups for user DOMAIN\user SAMBA config : - [global] workgroup = DOMAIN realm = DOMAIN.BIZ security = ADS password server = EIS printcap name = cups idmap uid = 1000-2000 idmap gid = 1000-2000 winbind enum users = Yes winbind enum groups = Yes winbind use default domain = Yes winbind nss info = rfc2307 cups options = raw Versions :- # smbd -V Version 3.4.2-1.1.3.1-2229-SUSE-SL11.2 # winbindd -V Version 3.4.2-1.1.3.1-2229-SUSE-SL11.2 Share your ideas... Regards, Vivek -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated via b9cfe10 s4-drs: fixed a crash in writspn from 8e34df4 s4-test: we need to import testtools before subunit/python http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit b9cfe109453937f5c2ae5008bee4b12226d68914 Author: Andrew Tridgell tri...@samba.org Date: Sat Nov 13 19:08:45 2010 +1100 s4-drs: fixed a crash in writspn sam_ctx_system may be NULL for non-privileged users Autobuild-User: Andrew Tridgell tri...@samba.org Autobuild-Date: Sat Nov 13 08:52:53 UTC 2010 on sn-devel-104 --- Summary of changes: source4/rpc_server/drsuapi/writespn.c | 10 -- 1 files changed, 8 insertions(+), 2 deletions(-) Changeset truncated at 500 lines: diff --git a/source4/rpc_server/drsuapi/writespn.c b/source4/rpc_server/drsuapi/writespn.c index 23b0626..320232d 100644 --- a/source4/rpc_server/drsuapi/writespn.c +++ b/source4/rpc_server/drsuapi/writespn.c @@ -160,6 +160,7 @@ WERROR dcesrv_drsuapi_DsWriteAccountSpn(struct dcesrv_call_state *dce_call, TALL int ret; unsigned spn_count=0; bool passed_checks = true; + struct ldb_context *sam_ctx; req = r-in.req-req1; count = req-count; @@ -215,9 +216,14 @@ WERROR dcesrv_drsuapi_DsWriteAccountSpn(struct dcesrv_call_state *dce_call, TALL } } + if (passed_checks b_state-sam_ctx_system) { + sam_ctx = b_state-sam_ctx_system; + } else { + sam_ctx = b_state-sam_ctx; + } + /* Apply to database */ - ret = dsdb_modify(passed_checks?b_state-sam_ctx_system:b_state-sam_ctx, - msg, DSDB_MODIFY_PERMISSIVE); + ret = dsdb_modify(sam_ctx, msg, DSDB_MODIFY_PERMISSIVE); if (ret != LDB_SUCCESS) { DEBUG(0,(Failed to modify SPNs on %s: %s\n, ldb_dn_get_linearized(msg-dn), -- Samba Shared Repository
Re: [SCM] Samba Shared Repository - branch master updated
Well basically yes, ekacnet. But I don't see any problem to use modify_ldif/add_ldif when sending very special requests as password changes and others. Sometimes it's really needed due to encoding constraints. Greets, Matthias Matthieu Patou wrote: Hi Mathias, We try to get rid of the modify_ldif way of doing ldb modification. It's better to create a ldb.Message() object and then do a ldb.modify() with it like this: --- a/source4/scripting/python/samba/upgradehelpers.py +++ b/source4/scripting/python/samba/upgradehelpers.py @@ -835,14 +835,13 @@ def update_machine_account_password(samdb, secrets_ldb, names): res = samdb.search(expression=expression, attrs=[]) assert(len(res) == 1) +msg = ldb.Message(res[0].dn) machinepass = samba.generate_random_password(128, 255) - -samdb.modify_ldif( -dn: + str(res[0].dn) + -changetype: modify -replace: clearTextPassword -clearTextPassword:: + base64.b64encode(machinepass.encode('utf-16-le')) + -) +mpu = machinepass.encode('utf-16-le') +msg[clearTextPassword] = ldb.MessageElement(mpu, +ldb.FLAG_MOD_REPLACE, +clearTextPassword) +samdb.modify(msg) res = samdb.search(expression=(samAccountName=%s$ % names.netbiosname), attrs=[msDs-keyVersionNumber]) Matthieu. On 11/11/2010 12:48, Matthias Dieter WallnXXfer wrote: The branch, master has been updated via bb241f5 s4:pytevent.c - fix a discard const warning via f036790 ldb:ldb_ldap.c rename operation - check for the RDN name and value via feb00fe s4:dsdb - proof against empty RDN values where expected via 4fe63d9 Cannot create OU using custom Schema class via e96c9df s4:objectclass LDB module - allow RDNs also to come from superclasses via 4f86f29 s4:passwords.py - add a test for the normal userPassword behaviour via 7f171a9 s4:password_hash and acl LDB modules - handle the userPassword attribute according to the dSHeuristics via d6c78fb s4:password_hash LDB module - move samdb_msg_find_old_and_new_ldb_val into the password_hash LDB module via eff1e8c s4:libnet/libnet_samsync_ldb.c - remove userPassword remove code via 39f8661 s4:local_password LDB module - remove schema checking code and fix some typos via ed704c2 s4:ldb_modules/util.c - dsHeuristics - dSHeuristics via 79548f0 s4:selftest/tests.py - skip the passwords.py suite on Windows 2000 domain function level via 5ded90e s4:acl.py - two password change tests are expected to fails on Windows 2000 function level via 2403aaa s4:upgradehelpers.py - use clearTextPassword rather than userPassword via 7c59ece s4:speedtest.py - use unicodePwd for setting user's password via 0e94569 s4:speedtest.py - remove duplicated code via 0a29e55 s4:speedtest.py - fix script name in the help text via ed1ca1c s4:speedtest.py - make it executable via cc7f390 s4:python tests - fix script names in the help text from ee50bdd s4-loadparm: fix the FLAG_DEFAULT settings on specially handled parameters http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit bb241f5cf8424c2576d5bc33ac149e5720b82068 Author: Matthias Dieter Wallnöferm...@samba.org Date: Wed Nov 10 16:21:41 2010 +0100 s4:pytevent.c - fix a discard const warning Autobuild-User: Matthias Dieter Wallnöferm...@samba.org Autobuild-Date: Thu Nov 11 09:47:55 UTC 2010 on sn-devel-104 commit f0367905d9a6db76712f1dcf9734f64fe5c5e1b3 Author: Matthias Dieter Wallnöferm...@samba.org Date: Wed Nov 10 16:20:38 2010 +0100 ldb:ldb_ldap.c rename operation - check for the RDN name and value Make it more similar to ldb_ildap.c and also more save commit feb00fe7cc238a78b3832c116cb4634936597735 Author: Matthias Dieter Wallnöferm...@samba.org Date: Wed Nov 10 16:05:16 2010 +0100 s4:dsdb - proof against empty RDN values where expected This should prevent crashes as pointed out on the mailing list. commit 4fe63d927e5e684d3e2bec9642a1e77b056ab2ed Author: Zahari Zaharievzahari.zahar...@postpath.com Date: Tue Nov 9 14:55:32 2010 +0200 Cannot create OU using custom Schema class If we define our own child class 'subClassOf' system Schema class e.g. organizationalUnit then we cannot create OU in the Dafualt Naming Context that has this custom Schama class in the objectClass attribute. commit e96c9df817326197a0866a18ad53621405b8bee8 Author: Matthias Dieter Wallnöferm...@samba.org Date: Wed Nov 10 15:12:02 2010 +0100 s4:objectclass LDB module - allow RDNs also to come from superclasses Detected by a testcase written by Zahari Zahariev. commit
[SCM] build.samba.org - branch master updated
The branch, master has been updated via e4024b8 Various fixes found during ad-hoc testing. via e771174 Add convenience function for avoiding database. Move database code to separate module. via d4df7e4 Fix some formatting. via ff7cecb More fixes for import-and-analyse. via 1ca3392 Use named tuples, some more fixes for hostdb. via ae0821a Warn early when creating host that already exists, move some storm code around. via 99f99ba Move some functionality to Host. via b26943a More storm usage. via 549ef86 Use storm in host database. via af937d5 Use storm to find builds. via 75f1b64 Remove unecessary argument from Build(). via 75e2697 use storm. from 2a37ec3 Be verbose when there's no previous build. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit e4024b82466d4b4966a4ac21ec0884cc296bf984 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 13:08:07 2010 +0100 Various fixes found during ad-hoc testing. commit e77117456e12043754d1da19e19880d1c5899722 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 06:40:18 2010 +0100 Add convenience function for avoiding database. Move database code to separate module. commit d4df7e442cb6acd8a4aacd8135a1b38c159925bc Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 06:05:10 2010 +0100 Fix some formatting. commit ff7cecb4b0af868781a1a84563bc78921a250588 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 05:49:39 2010 +0100 More fixes for import-and-analyse. commit 1ca339299de89b486979c73222e9faa9ead01b7e Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 05:24:47 2010 +0100 Use named tuples, some more fixes for hostdb. commit ae0821a89cfc6810c4410f9f1db26d629c7ee41d Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 05:11:05 2010 +0100 Warn early when creating host that already exists, move some storm code around. commit 99f99bab13945b6d6cdd499ecfdae845e08d8ad1 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 04:43:47 2010 +0100 Move some functionality to Host. commit b26943a223f8215cf132c0176a0f1385d66d1f7c Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 04:16:47 2010 +0100 More storm usage. commit 549ef86335847197cc58ddca5dc4d407f925814f Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 04:09:04 2010 +0100 Use storm in host database. commit af937d5dc9d2c68aa0b69f7247e59ad0320ec570 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 03:50:04 2010 +0100 Use storm to find builds. commit 75f1b64035ab24de2dcd38e5d0d914cb992cb2b7 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 03:09:46 2010 +0100 Remove unecessary argument from Build(). commit 75e2697b4116421545c64780fc4610c636461891 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 02:55:39 2010 +0100 use storm. --- Summary of changes: admin.py | 46 +--- buildfarm/__init__.py | 36 +- buildfarm/data.py | 128 -- buildfarm/hostdb.py | 117 +++- buildfarm/sqldb.py| 217 + buildfarm/tests/__init__.py | 19 +++- buildfarm/tests/test_buildfarm.py |9 +-- buildfarm/tests/test_data.py | 33 +-- buildfarm/tests/test_hostdb.py| 67 +--- buildfarm/tests/test_sqldb.py | 78 + import-and-analyse.py |9 +- mail-dead-hosts.py| 11 +- web/build.py | 41 13 files changed, 517 insertions(+), 294 deletions(-) create mode 100644 buildfarm/sqldb.py create mode 100644 buildfarm/tests/test_sqldb.py Changeset truncated at 500 lines: diff --git a/admin.py b/admin.py index f17b26e..79c6ce7 100755 --- a/admin.py +++ b/admin.py @@ -70,16 +70,21 @@ else: if op == remove: hostname = raw_input(Please enter hostname to delete: ) try: -db.deletehost(hostname) +db.deletehost(hostname.decode(utf-8)) except hostdb.NoSuchHost, e: print No such host '%s' % e.name sys.exit(1) else: +db.commit() update_rsyncd_secrets() update_hosts_list() elif op == modify: hostname = raw_input(Please enter hostname to modify: ) -host = db.host(hostname) +try: +host = db.host(hostname.decode(utf-8)) +except hostdb.NoSuchHost, e: +print No such host '%s' % e.name +sys.exit(1) print Owner: %s %s % host.owner print Platform: %s % host.platform print @@ -88,19 +93,13 @@ elif op == modify: mod_op = platform if mod_op == platform: platform = raw_input(Enter new platform: ) -try: -
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated via 7d2260c s4:password_hash LDB module - return ERR_CONSTRAINT_VIOLATION on password conversion errors via ac0dcd1 s4:upgradeprovision - why not directly use provision:0? via 113a9c1 s4:objectclass LDB module - multiple objectClass change elements are unfortunately still allowed from b9cfe10 s4-drs: fixed a crash in writspn http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit 7d2260cdd18b7354c372d6e8833d1554ab7894d0 Author: Matthias Dieter Wallnöfer m...@samba.org Date: Sat Nov 13 12:47:53 2010 +0100 s4:password_hash LDB module - return ERR_CONSTRAINT_VIOLATION on password conversion errors This errors can happen also on a regular basis - then we shouldn't return ERR_OPERATIONS_ERROR (this error code is reserved for very serious failures). Autobuild-User: Matthias Dieter Wallnöfer m...@samba.org Autobuild-Date: Sat Nov 13 12:37:36 UTC 2010 on sn-devel-104 commit ac0dcd1e07e8f2642c01912d9dfbc457a18fdfae Author: Matthias Dieter Wallnöfer m...@samba.org Date: Sat Nov 13 12:33:26 2010 +0100 s4:upgradeprovision - why not directly use provision:0? commit 113a9c1806863f9794c3a611ed4c4d910c3bf11b Author: Matthias Dieter Wallnöfer m...@samba.org Date: Sat Nov 13 12:25:40 2010 +0100 s4:objectclass LDB module - multiple objectClass change elements are unfortunately still allowed The test message has been compressed - therefore I've now used modify_ldif. --- Summary of changes: source4/dsdb/samdb/ldb_modules/objectclass.c | 330 source4/dsdb/samdb/ldb_modules/password_hash.c | 15 +- source4/dsdb/tests/python/ldap.py | 21 +- source4/scripting/bin/upgradeprovision |8 +- 4 files changed, 191 insertions(+), 183 deletions(-) Changeset truncated at 500 lines: diff --git a/source4/dsdb/samdb/ldb_modules/objectclass.c b/source4/dsdb/samdb/ldb_modules/objectclass.c index d2b4f10..e863d48 100644 --- a/source4/dsdb/samdb/ldb_modules/objectclass.c +++ b/source4/dsdb/samdb/ldb_modules/objectclass.c @@ -921,7 +921,7 @@ static int objectclass_do_mod(struct oc_context *ac) TALLOC_CTX *mem_ctx; struct class_list *sorted, *current; const struct dsdb_class *objectclass; - unsigned int i, j; + unsigned int i, j, k; bool found, replace = false; int ret; @@ -939,13 +939,6 @@ static int objectclass_do_mod(struct oc_context *ac) return ldb_operr(ldb); } - oc_el_change = ldb_msg_find_element(ac-req-op.mod.message, - objectClass); - if (oc_el_change == NULL) { - /* we should have an objectclass change operation */ - return ldb_operr(ldb); - } - /* use a new message structure */ msg = ldb_msg_new(ac); if (msg == NULL) { @@ -959,195 +952,210 @@ static int objectclass_do_mod(struct oc_context *ac) return ldb_oom(ldb); } - switch (oc_el_change-flags LDB_FLAG_MOD_MASK) { - case LDB_FLAG_MOD_ADD: - /* Merge the two message elements */ - for (i = 0; i oc_el_change-num_values; i++) { - for (j = 0; j oc_el_entry-num_values; j++) { - if (strcasecmp((char *)oc_el_change-values[i].data, - (char *)oc_el_entry-values[j].data) == 0) { - /* we cannot add an already existing object class */ + /* We've to walk over all objectClass message elements */ + for (k = 0; k ac-req-op.mod.message-num_elements; k++) { + if (ldb_attr_cmp(ac-req-op.mod.message-elements[k].name, +objectClass) != 0) { + continue; + } + + oc_el_change = ac-req-op.mod.message-elements[k]; + + switch (oc_el_change-flags LDB_FLAG_MOD_MASK) { + case LDB_FLAG_MOD_ADD: + /* Merge the two message elements */ + for (i = 0; i oc_el_change-num_values; i++) { + for (j = 0; j oc_el_entry-num_values; j++) { + if (ldb_attr_cmp((char *)oc_el_change-values[i].data, +(char *)oc_el_entry-values[j].data) == 0) { + ldb_asprintf_errstring(ldb, + objectclass: cannot re-add an existing objectclass: '%.*s'!, + (int)oc_el_change-values[i].length, +
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 1a6118b Support overriding cache page on command line for web/build.py Put pages into a separate class. Several other fixes. from e4024b8 Various fixes found during ad-hoc testing. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 1a6118b447040ec837cba9e7c832f779a3a548e6 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 13:55:45 2010 +0100 Support overriding cache page on command line for web/build.py Put pages into a separate class. Several other fixes. --- Summary of changes: buildfarm/history.py |2 +- web/build.py | 1204 ++ 2 files changed, 620 insertions(+), 586 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/history.py b/buildfarm/history.py index 537ce48..d52bec1 100644 --- a/buildfarm/history.py +++ b/buildfarm/history.py @@ -106,5 +106,5 @@ class GitBranch(object): commit = self.repo[revision] f = StringIO() for (oldpath, newpath), (oldmode, newmode), (oldsha, newsha) in self._changes_for(commit): -write_blob_diff((oldpath, oldmode, self.store[oldsha]), (newpath, newmode, self.store[newsha])) +write_blob_diff(f, (oldpath, oldmode, self.store[oldsha]), (newpath, newmode, self.store[newsha])) return (self._revision_from_commit(commit), f.getvalue()) diff --git a/web/build.py b/web/build.py index dc1ddb1..f8d74d5 100755 --- a/web/build.py +++ b/web/build.py @@ -1,12 +1,16 @@ #!/usr/bin/python # This CGI script presents the results of the build_farm build + +# Copyright (C) Jelmer Vernooij jel...@samba.org 2010 +# Copyright (C) Matthieu Patou m...@matws.net 2010 +# +# Based on the original web/build.pl: # # Copyright (C) Andrew Tridgell tri...@samba.org 2001-2005 # Copyright (C) Andrew Bartlett abart...@samba.org 2001 # Copyright (C) Vance Lankhaar va...@samba.org 2002-2005 # Copyright (C) Martin Pool m...@samba.org2001 -# Copyright (C) Jelmer Vernooij jel...@samba.org 2007-2010 -# Copyright (C) Matthieu Patou m...@matws.net 2010-2010 +# Copyright (C) Jelmer Vernooij jel...@samba.org 2007 # # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by @@ -32,6 +36,7 @@ sys.path.insert(0, os.path.join(os.path.dirname(__file__), ..)) from buildfarm import ( CachingBuildFarm, data, +hostdb, util, ) @@ -44,16 +49,6 @@ standalone = 0 webdir = os.path.dirname(__file__) basedir = os.path.abspath(os.path.join(webdir, ..)) -buildfarm = CachingBuildFarm() - -db = data.BuildResultStore(basedir) -hostsdb = buildfarm.hostdb - -compilers = buildfarm.compilers -# host.properties are unicode object and the framework expect string object -hosts = dict([(host.name.encode(utf-8), host) for host in hostsdb.hosts()]) -trees = buildfarm.trees -OLDAGE = buildfarm.OLDAGE UNPACKED_BASE = http://svn.samba.org/ftp/unpacked; GITWEB_BASE = http://gitweb.samba.org; @@ -123,13 +118,6 @@ def build_status_html(myself, build): return build_link(myself, build.tree, build.host, build.compiler, build.revision, status) -def red_age(age): -show an age as a string -if age OLDAGE: -return span clsas='old'%s/span % util.dhm_time(age) -return util.dhm_time(age) - - def build_status_vals(status): translate a status into a set of int representing status status = util.strip_html(status) @@ -142,89 +130,6 @@ def build_status_vals(status): return status.split(/) -def view_summary(myself, output_type): -view build summary -i = 0 -cols = 2 -broken = 0 -broken_count = {} -panic_count = {} -host_count = {} - -# zero broken and panic counters -for tree in trees: -broken_count[tree] = 0 -panic_count[tree] = 0 -host_count[tree] = 0 - -# set up a variable to store the broken builds table's code, so we can -# output when we want -broken_table = -last_host = - -# for the text report, include the current time -if output_type == 'text': -t = time.gmtime() -yield Build status as of %s\n\n % t - -for host in hosts.values(): -for compiler in compilers: -for tree in trees: -try: -build = buildfarm.get_build(tree, host.name.encode(utf-8), compiler) -status = build_status_html(myself, build) -except data.NoSuchBuildError: -continue -age_mtime = build.age_mtime() -host_count[tree]+=1 - -if status failed in status: -broken_count[tree]+=1 -if PANIC in
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 1d36022 Kill a global. from 1a6118b Support overriding cache page on command line for web/build.py Put pages into a separate class. Several other fixes. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 1d3602271d3b5898b46b2c3890a98ac8d92559b6 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 14:08:19 2010 +0100 Kill a global. --- Summary of changes: web/build.py | 97 + 1 files changed, 49 insertions(+), 48 deletions(-) Changeset truncated at 500 lines: diff --git a/web/build.py b/web/build.py index f8d74d5..d933554 100755 --- a/web/build.py +++ b/web/build.py @@ -163,14 +163,12 @@ def format_subunit_reason(reason): return div class=\reason\%s/div % reason -def print_log_pretty(log): -# prints the log in a visually appealing manner -global indice -indice = 0 +class LogPrettyPrinter(object): + +def __init__(self): +self.indice = 0 -# do some pretty printing for the actions -def pretty_print(m): -global indice +def _pretty_print(self, m): output = m.group(1) actionName = m.group(2) status = m.group(3) @@ -178,37 +176,19 @@ def print_log_pretty(log): if actionName == 'cc_checker': output = print_log_cc_checker(output) -indice += 1 -return make_collapsible_html('action', actionName, output, indice, status) - -pattern = re.compile((Running action\s+([\w\-]+)$(?:\s^.*$)*?\sACTION\ (PASSED|FAILED):\ ([\w\-]+)$), re.M) -log = pattern.sub(pretty_print, log) +self.indice += 1 +return make_collapsible_html('action', actionName, output, self.indice, status) # log is already CGI-escaped, so handle '' in test name by handling gt -def format_stage(m): -indice += 1 -return make_collapsible_html('test', m.group(1), m.group(2), indice, m.group(3)) - -log = re.sub( - --==--==--==--==--==--==--==--==--==--==--.*? - Running\ test\ ([\w\-=,_:\ /.;]+).*? - --==--==--==--==--==--==--==--==--==--==-- - (.*?) - ==.*? - TEST\ (FAILED|PASSED|SKIPPED):.*? - ==\s+ -, format_stage, log) +def _format_stage(self, m): +self.indice += 1 +return make_collapsible_html('test', m.group(1), m.group(2), self.indice, m.group(3)) -def format_skip_testsuite(m): -global indice -indice += 1 -return make_collapsible_html('test', m.group(1), '', indice, 'skipped') +def _format_skip_testsuite(self, m): +self.indice += 1 +return make_collapsible_html('test', m.group(1), '', self.indice, 'skipped') -log = re.sub(skip-testsuite: ([\w\-=,_:\ /.; \(\)]+).*?, -format_skip_testsuite, log) - -def format_testsuite(m): -global indice +def _format_testsuite(self, m): testName = m.group(1) content = m.group(2) status = subunit_to_buildfarm_result(m.group(3)) @@ -216,23 +196,44 @@ def print_log_pretty(log): errorReason = format_subunit_reason(m.group(4)) else: errorReason = -indice += 1 -return make_collapsible_html('test', testName, content+errorReason, indice, status) - -pattern = re.compile(^testsuite: (.+)$\s((?:^.*$\s)*?)testsuite-(\w+): .*?(?:(\[$\s(?:^.*$\s)*?^\]$)|$), re.M) -log = pattern.sub(format_testsuite, log) +self.indice += 1 +return make_collapsible_html('test', testName, content+errorReason, self.indice, status) + +def _format_test(self, m): +self.indice += 1 +return make_collapsible_html('test', m.group(1), m.group(2)+format_subunit_reason(m.group(4)), self.indice, subunit_to_buildfarm_result(m.group(3))) + +def pretty_print(self, log): +# do some pretty printing for the actions +pattern = re.compile((Running action\s+([\w\-]+)$(?:\s^.*$)*?\sACTION\ (PASSED|FAILED):\ ([\w\-]+)$), re.M) +log = pattern.sub(self._pretty_print, log) + +log = re.sub( + --==--==--==--==--==--==--==--==--==--==--.*? + Running\ test\ ([\w\-=,_:\ /.;]+).*? + --==--==--==--==--==--==--==--==--==--==-- + (.*?) + ==.*? + TEST\ (FAILED|PASSED|SKIPPED):.*? + ==\s+ +, self._format_stage, log) + +log = re.sub(skip-testsuite: ([\w\-=,_:\ /.; \(\)]+).*?, +self._format_skip_testsuite, log) + +pattern = re.compile(^testsuite: (.+)$\s((?:^.*$\s)*?)testsuite-(\w+):
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 7fd815d fix git root. from 1d36022 Kill a global. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 7fd815d0a196ed363342b2aa77707b6863c825d7 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 14:21:40 2010 +0100 fix git root. --- Summary of changes: buildfarm/__init__.py |9 +++-- buildfarm/history.py | 16 2 files changed, 19 insertions(+), 6 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/__init__.py b/buildfarm/__init__.py index ba87fb1..76a4146 100644 --- a/buildfarm/__init__.py +++ b/buildfarm/__init__.py @@ -21,6 +21,8 @@ import ConfigParser import os import re +GIT_ROOT = /data/git + class Tree(object): A tree to build. @@ -35,8 +37,11 @@ class Tree(object): self.scm = scm def get_branch(self): -from buildfarm.history import GitBranch -return GitBranch(self.repo, self.branch) +if self.scm == git: +from buildfarm.history import GitBranch +return GitBranch(os.path.join(GIT_ROOT, self.repo), self.branch) +else: +raise NotImplementedError(self.scm) def __repr__(self): return %s %r % (self.__class__.__name__, self.name) diff --git a/buildfarm/history.py b/buildfarm/history.py index d52bec1..ffe392f 100644 --- a/buildfarm/history.py +++ b/buildfarm/history.py @@ -27,8 +27,11 @@ from dulwich.repo import Repo class Branch(object): +A version control branch. def authors(self): +Determine all authors that have contributed to this project. + ret = set() for rev in self.log(): ret.add(rev.author) @@ -43,7 +46,8 @@ class Branch(object): class Revision(object): -def __init__(self, revision, date, author, message, modified=[], added=[], removed=[]): +def __init__(self, revision, date, author, message, modified=[], added=[], +removed=[]): self.revision = revision self.date = date self.author = author @@ -78,7 +82,9 @@ class GitBranch(object): removed.add(oldpath) else: modified.add(newpath) -return Revision(commit.id, commit.commit_time, commit.author, commit.message, modified=modified, removed=removed, added=added) +return Revision(commit.id, commit.commit_time, commit.author, +commit.message, modified=modified, removed=removed, +added=added) def log(self, from_rev=None, exclude_revs=None): if from_rev is None: @@ -105,6 +111,8 @@ class GitBranch(object): def diff(self, revision): commit = self.repo[revision] f = StringIO() -for (oldpath, newpath), (oldmode, newmode), (oldsha, newsha) in self._changes_for(commit): -write_blob_diff(f, (oldpath, oldmode, self.store[oldsha]), (newpath, newmode, self.store[newsha])) +changes = self._changes_for(commit) +for (oldpath, newpath), (oldmode, newmode), (oldsha, newsha) in changes: +write_blob_diff(f, (oldpath, oldmode, self.store[oldsha]), +(newpath, newmode, self.store[newsha])) return (self._revision_from_commit(commit), f.getvalue()) -- build.samba.org
[SCM] build.samba.org - branch master updated
The branch, master has been updated via c868b29 Allow limiting the number of revisions that we look back. from de84a30 Provide GitBranch.authors. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit c868b2951d467200f728c15049f5099cec5bc98d Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 14:36:29 2010 +0100 Allow limiting the number of revisions that we look back. --- Summary of changes: buildfarm/history.py |6 -- web/build.py |3 ++- 2 files changed, 6 insertions(+), 3 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/history.py b/buildfarm/history.py index 1c96120..383782b 100644 --- a/buildfarm/history.py +++ b/buildfarm/history.py @@ -29,11 +29,13 @@ from dulwich.repo import Repo class Branch(object): A version control branch. -def authors(self): +def authors(self, limit=None): Determine all authors that have contributed to this project. ret = set() -for rev in self.log(): +for i, rev in enumerate(self.log()): +if i == limit: +break ret.add(rev.author) return ret diff --git a/web/build.py b/web/build.py index d933554..88e99aa 100755 --- a/web/build.py +++ b/web/build.py @@ -52,6 +52,7 @@ basedir = os.path.abspath(os.path.join(webdir, ..)) UNPACKED_BASE = http://svn.samba.org/ftp/unpacked; GITWEB_BASE = http://gitweb.samba.org; +HISTORY_HORIZON = 1000 # this is automatically filled in deadhosts = [] @@ -875,7 +876,7 @@ class RecentCheckinsPage(BuildFarmPage): def render(self, myself, tree, author=None): t = self.buildfarm.trees[tree] authors = set([ALL]) -authors.update(t.get_branch().authors(tree)) +authors.update(t.get_branch().authors(limit=HISTORY_HORIZON)) yield h2Recent checkins for %s (%s branch %s)/h2\n % ( tree, t.scm, t.branch) -- build.samba.org
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 0ab262b attempt to speed up log parsing by compiling regexes beforehand. from c868b29 Allow limiting the number of revisions that we look back. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 0ab262b677046d51191a2598d040e09f0d59ff7b Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 14:40:18 2010 +0100 attempt to speed up log parsing by compiling regexes beforehand. --- Summary of changes: buildfarm/data.py | 32 1 files changed, 16 insertions(+), 16 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/data.py b/buildfarm/data.py index 8629842..0bd1498 100644 --- a/buildfarm/data.py +++ b/buildfarm/data.py @@ -109,30 +109,17 @@ def build_status_from_logs(log, err): ret = BuildStatus() stages = [] +re_status = re.compile(^([A-Z_]+) STATUS:(\s*\d+)$) +re_action = re.compile(^ACTION (PASSED|FAILED):\s+test$) for l in log: -m = re.match(^([A-Z_]+) STATUS:(\s*\d+)$, l) -if m: -stages.append(BuildStageResult(m.group(1), int(m.group(2).strip( -if m.group(1) == TEST: -test_seen = 1 -continue -m = re.match(^ACTION (PASSED|FAILED):\s+test$, l) -if m and not test_seen: -if m.group(1) == PASSED: -stages.append(BuildStageResult(TEST, 0)) -else: -stages.append(BuildStageResult(TEST, 1)) -continue - if l.startswith(No space left on device): ret.other_failures.add(disk full) continue if l.startswith(maximum runtime exceeded): ret.other_failures.add(timeout) continue -m = re.match(^(PANIC|INTERNAL ERROR):.*$, l) -if m: +if l.startswith(PANIC:) or l.startswith(INTERNAL ERROR:): ret.other_failures.add(panic) continue if l.startswith(testsuite-failure: ) or l.startswith(testsuite-error: ): @@ -141,6 +128,19 @@ def build_status_from_logs(log, err): if l.startswith(testsuite-success: ): test_successes += 1 continue +m = re_status.match(l) +if m: +stages.append(BuildStageResult(m.group(1), int(m.group(2).strip( +if m.group(1) == TEST: +test_seen = 1 +continue +m = re_action.match(l) +if m and not test_seen: +if m.group(1) == PASSED: +stages.append(BuildStageResult(TEST, 0)) +else: +stages.append(BuildStageResult(TEST, 1)) +continue # Scan err file for specific errors for l in err: -- build.samba.org
Re: [SCM] Samba Shared Repository - branch master updated
On Sat, 2010-11-13 at 11:50 +0100, Matthias Dieter Wallnöfer wrote: Well basically yes, ekacnet. But I don't see any problem to use modify_ldif/add_ldif when sending very special requests as password changes and others. Sometimes it's really needed due to encoding constraints. Can you give an example of this? E.g. in this situation below it would be simpler without the ldif. Since LDIFs are always converted to ldb Message anyway before they are processed I don't see how they could be used in a larger number of situations. Cheers, Jelmer u Patou wrote: Hi Mathias, We try to get rid of the modify_ldif way of doing ldb modification. It's better to create a ldb.Message() object and then do a ldb.modify() with it like this: --- a/source4/scripting/python/samba/upgradehelpers.py +++ b/source4/scripting/python/samba/upgradehelpers.py @@ -835,14 +835,13 @@ def update_machine_account_password(samdb, secrets_ldb, names): res = samdb.search(expression=expression, attrs=[]) assert(len(res) == 1) +msg = ldb.Message(res[0].dn) machinepass = samba.generate_random_password(128, 255) - -samdb.modify_ldif( -dn: + str(res[0].dn) + -changetype: modify -replace: clearTextPassword -clearTextPassword:: + base64.b64encode(machinepass.encode('utf-16-le')) + -) +mpu = machinepass.encode('utf-16-le') +msg[clearTextPassword] = ldb.MessageElement(mpu, +ldb.FLAG_MOD_REPLACE, +clearTextPassword) +samdb.modify(msg) res = samdb.search(expression=(samAccountName=%s$ % names.netbiosname), attrs=[msDs-keyVersionNumber]) Matthieu. On 11/11/2010 12:48, Matthias Dieter WallnXXfer wrote: The branch, master has been updated via bb241f5 s4:pytevent.c - fix a discard const warning via f036790 ldb:ldb_ldap.c rename operation - check for the RDN name and value via feb00fe s4:dsdb - proof against empty RDN values where expected via 4fe63d9 Cannot create OU using custom Schema class via e96c9df s4:objectclass LDB module - allow RDNs also to come from superclasses via 4f86f29 s4:passwords.py - add a test for the normal userPassword behaviour via 7f171a9 s4:password_hash and acl LDB modules - handle the userPassword attribute according to the dSHeuristics via d6c78fb s4:password_hash LDB module - move samdb_msg_find_old_and_new_ldb_val into the password_hash LDB module via eff1e8c s4:libnet/libnet_samsync_ldb.c - remove userPassword remove code via 39f8661 s4:local_password LDB module - remove schema checking code and fix some typos via ed704c2 s4:ldb_modules/util.c - dsHeuristics - dSHeuristics via 79548f0 s4:selftest/tests.py - skip the passwords.py suite on Windows 2000 domain function level via 5ded90e s4:acl.py - two password change tests are expected to fails on Windows 2000 function level via 2403aaa s4:upgradehelpers.py - use clearTextPassword rather than userPassword via 7c59ece s4:speedtest.py - use unicodePwd for setting user's password via 0e94569 s4:speedtest.py - remove duplicated code via 0a29e55 s4:speedtest.py - fix script name in the help text via ed1ca1c s4:speedtest.py - make it executable via cc7f390 s4:python tests - fix script names in the help text from ee50bdd s4-loadparm: fix the FLAG_DEFAULT settings on specially handled parameters http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit bb241f5cf8424c2576d5bc33ac149e5720b82068 Author: Matthias Dieter Wallnöferm...@samba.org Date: Wed Nov 10 16:21:41 2010 +0100 s4:pytevent.c - fix a discard const warning Autobuild-User: Matthias Dieter Wallnöferm...@samba.org Autobuild-Date: Thu Nov 11 09:47:55 UTC 2010 on sn-devel-104 commit f0367905d9a6db76712f1dcf9734f64fe5c5e1b3 Author: Matthias Dieter Wallnöferm...@samba.org Date: Wed Nov 10 16:20:38 2010 +0100 ldb:ldb_ldap.c rename operation - check for the RDN name and value Make it more similar to ldb_ildap.c and also more save commit feb00fe7cc238a78b3832c116cb4634936597735 Author: Matthias Dieter Wallnöferm...@samba.org Date: Wed Nov 10 16:05:16 2010 +0100 s4:dsdb - proof against empty RDN values where expected This should prevent crashes as pointed out on the mailing list. commit 4fe63d927e5e684d3e2bec9642a1e77b056ab2ed Author: Zahari Zaharievzahari.zahar...@postpath.com Date: Tue Nov 9 14:55:32 2010 +0200 Cannot create OU using custom Schema class If we define our own child class
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 758927a Fix cache. from be38b55 Limit history. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 758927a28262520944ed6f16029ebe993c8f1498 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 15:11:10 2010 +0100 Fix cache. --- Summary of changes: buildfarm/filecache.py |2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/filecache.py b/buildfarm/filecache.py index 52e0183..d9c4a16 100644 --- a/buildfarm/filecache.py +++ b/buildfarm/filecache.py @@ -24,6 +24,8 @@ from buildfarm import ( from buildfarm.data import ( Build, BuildResultStore, +BuildStageResult, +BuildStatus, NoSuchBuildError, UploadBuildResultStore, check_dir_exists, -- build.samba.org
Re: [SCM] Samba Shared Repository - branch master updated
On 13/11/2010 17:09, Matthias Dieter Wallnöfer wrote: Sure, Jelmer! Consider this modify request message - doesn't make much sense but was needed for a test. There's obviously a bug that the previous message element is overwritten by the second one. Only with modify_ldif it works correctly. Greets, Matthias -m = Message() -m.dn = Dn(ldb, cn=ldaptestuser,cn=users, + self.base_dn) -m[objectClass] = MessageElement(bootableDevice, FLAG_MOD_DELETE, - objectClass) -m[objectClass] = MessageElement(bootableDevice, FLAG_MOD_ADD, - objectClass) -try: -ldb.modify(m) -self.fail() -except LdbError, (num, _): -self.assertEquals(num, ERR_ATTRIBUTE_OR_VALUE_EXISTS) +# More than one change operation is allowed +ldb.modify_ldif( +dn: cn=ldaptestuser,cn=users, + self.base_dn + +changetype: modify +delete: objectClass +objectClass: bootableDevice +add: objectClass +objectClass: bootableDevice +) Well that's pretty obvious that the m[objecClass] way is not correct as you expect to it to have both the add and remove in your message but in reality you'll have just the add. The first question: did you try the FLAG_MOD_MODIFY ? The second question: it seems that you are trying to trigger some behavior by changing the objectclass what is the goal of this ? So for the example you gave modify_ldif might be correct but clearly for the change you made in upgradeprovisionhelper it wasn't as we can do it with Message() + modify(). Matthieu. -- Matthieu Patou Samba Teamhttp://samba.org Private repo http://git.samba.org/?p=mat/samba.git;a=summary
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 4295e5e Fix CSS in standalone mode, move static file hack out of app, get rid of global. from c24daf3 Remove obsolete tree, more history horizon. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 4295e5e2b719ac3eb73402f696ce842cefb2a253 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 16:06:49 2010 +0100 Fix CSS in standalone mode, move static file hack out of app, get rid of global. --- Summary of changes: web/build.py | 49 ++--- 1 files changed, 26 insertions(+), 23 deletions(-) Changeset truncated at 500 lines: diff --git a/web/build.py b/web/build.py index 38949af..a8e96a7 100755 --- a/web/build.py +++ b/web/build.py @@ -47,7 +47,6 @@ import re import time import wsgiref.util -standalone = 0 webdir = os.path.dirname(__file__) basedir = os.path.abspath(os.path.join(webdir, ..)) @@ -944,22 +943,6 @@ class BuildFarmApp(object): fn_name = get_param(form, 'function') or '' myself = wsgiref.util.application_uri(environ) -if standalone and environ['PATH_INFO']: -dir = os.path.join(os.path.dirname(__file__)) -if re.match(^/[a-zA-Z0-9_-]+(\.[a-zA-Z0-9_-]+)?, environ['PATH_INFO']): -static_file = %s/%s % (dir, environ['PATH_INFO']) -if os.path.exists(static_file): -tab = environ['PATH_INFO'].split('.') -if len(tab) 1: -extension = tab[-1] -import mimetypes -mimetypes.init() -type = mimetypes.types_map[.%s % extension] -start_response('200 OK', [('Content-type', type)]) -data = open(static_file, 'rb').read() -yield data -return - if fn_name == 'text_diff': start_response('200 OK', [('Content-type', 'application/x-diff')]) tree = get_param(form, 'tree') @@ -982,10 +965,7 @@ class BuildFarmApp(object): yield meta name='description' contents='Home of the Samba Build Farm, the automated testing facility.'/ yield meta name='robots' contents='noindex'/ yield link rel='stylesheet' href='/build_farm.css' type='text/css' media='all'/ -if standalone: -yield link rel='stylesheet' href='common.css' type='text/css' media='all'/ -else: -yield link rel='stylesheet' href='http://master.samba.org/samba/style/common.css' type='text/css' media='all'/ +yield link rel='stylesheet' href='http://master.samba.org/samba/style/common.css' type='text/css' media='all'/ yield link rel='shortcut icon' href='http://www.samba.org/samba/images/favicon.ico'/ yield /head yield body @@ -1038,13 +1018,36 @@ if __name__ == '__main__': parser = optparse.OptionParser([options]) parser.add_option(--standalone, help=Run as standalone server (useful for debugging), action=store_true) parser.add_option(--cachedirname, help=Cache directory name, type=str) +parser.add_option(--port, help=Port to listen on (in standalone mode) [localhost:8000], default=localhost:8000, type=str) opts, args = parser.parse_args() buildfarm = CachingBuildFarm(cachedirname=opts.cachedirname) buildApp = BuildFarmApp(buildfarm) if opts.standalone: -standalone = 1 from wsgiref.simple_server import make_server -httpd = make_server('localhost', 8000, buildApp) + +def standaloneApp(environ, start_response): +if environ['PATH_INFO']: +dir = os.path.join(os.path.dirname(__file__)) +if re.match(^/[a-zA-Z0-9_-]+(\.[a-zA-Z0-9_-]+)?, environ['PATH_INFO']): +static_file = %s/%s % (dir, environ['PATH_INFO']) +if os.path.exists(static_file): +tab = environ['PATH_INFO'].split('.') +if len(tab) 1: +extension = tab[-1] +import mimetypes +mimetypes.init() +type = mimetypes.types_map[.%s % extension] +start_response('200 OK', [('Content-type', type)]) +data = open(static_file, 'rb').read() +yield data +return +yield .join(buildApp(environ, start_response)) +try: +(address, port) = opts.port.rsplit(:, 1) +except ValueError: +address = localhost +port = opts.port +httpd =
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 32b2b08 Mention correct port in message. from 4295e5e Fix CSS in standalone mode, move static file hack out of app, get rid of global. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 32b2b08294b9c2e98828d98df30ae77eab25ac92 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 16:08:34 2010 +0100 Mention correct port in message. --- Summary of changes: buildfarm/data.py |3 --- web/build.py |2 +- 2 files changed, 1 insertions(+), 4 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/data.py b/buildfarm/data.py index 4120730..d6d5732 100644 --- a/buildfarm/data.py +++ b/buildfarm/data.py @@ -376,6 +376,3 @@ class BuildResultStore(object): def get_latest_revision(self, tree, host, compiler): raise NoSuchBuildError(tree, host, compiler) - - - diff --git a/web/build.py b/web/build.py index a8e96a7..227512a 100755 --- a/web/build.py +++ b/web/build.py @@ -1048,7 +1048,7 @@ if __name__ == '__main__': address = localhost port = opts.port httpd = make_server(address, int(port), standaloneApp) -print Serving on port 8000... +print Serving on %s:%d... % (address, int(port)) httpd.serve_forever() else: import wsgiref.handlers -- build.samba.org
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 398247a Improve regressed_since. from 7176a9c Fix old revisions table. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 398247a96c40d927d0f0b81ee8a816f2ff16d5a7 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 16:48:01 2010 +0100 Improve regressed_since. --- Summary of changes: buildfarm/data.py |6 ++ 1 files changed, 6 insertions(+), 0 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/data.py b/buildfarm/data.py index d6d5732..47a76b4 100644 --- a/buildfarm/data.py +++ b/buildfarm/data.py @@ -71,6 +71,12 @@ class BuildStatus(object): Check if this build has regressed since another build. if disk full in self.other_failures: return False +if timeout in self.other_failures and timeout in other.other_failures: +# When the timeout happens exactly can differ slightly, so it's okay +# if the numbers are a bit different.. +return False +if panic in self.other_failures and not panic in other.other_failures: +return True return cmp(self._status_tuple(), other._status_tuple()) def __cmp__(self, other): -- build.samba.org
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 73d45eb Cope with system version of sqlite. from 731e40e Fix commit handling from specific author. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 73d45eb068a9cbed4712e3e7742f09c9707746aa Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 17:26:31 2010 +0100 Cope with system version of sqlite. --- Summary of changes: buildfarm/sqldb.py |7 +-- 1 files changed, 5 insertions(+), 2 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/sqldb.py b/buildfarm/sqldb.py index d1ca87d..ca7d235 100644 --- a/buildfarm/sqldb.py +++ b/buildfarm/sqldb.py @@ -33,7 +33,10 @@ from buildfarm.hostdb import ( ) import os -import pysqlite2 +try: +import sqlite3 +except ImportError: +from pysqlite2 import dbapi2 as sqlite3 from storm.database import create_database from storm.locals import Bool, Desc, Int, Unicode, RawStr from storm.store import Store @@ -100,7 +103,7 @@ class StormHostDatabase(HostDatabase): try: self.store.add(newhost) self.store.flush() -except pysqlite2.dbapi2.IntegrityError: +except sqlite3.IntegrityError: raise HostAlreadyExists(name) return newhost -- build.samba.org
[SCM] build.samba.org - branch master updated
The branch, master has been updated via d5fc97f Don't kill upload builds just yet, improve verboseness. from 9ff770d Cope with previous builds disappearing. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit d5fc97f2d69f206b0db768b9d0f094d92af76806 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 18:08:33 2010 +0100 Don't kill upload builds just yet, improve verboseness. --- Summary of changes: buildfarm/data.py |1 + buildfarm/sqldb.py|8 import-and-analyse.py |9 + 3 files changed, 10 insertions(+), 8 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/data.py b/buildfarm/data.py index 47a76b4..bc0eac9 100644 --- a/buildfarm/data.py +++ b/buildfarm/data.py @@ -376,6 +376,7 @@ class BuildResultStore(object): os.link(build.basename+.log, new_basename+.log) if os.path.exists(build.basename+.err): os.link(build.basename+.err, new_basename+.err) +return Build(new_basename, build.tree, build.host, build.compiler, rev) def get_previous_revision(self, tree, host, compiler, revision): raise NoSuchBuildError(tree, host, compiler, revision) diff --git a/buildfarm/sqldb.py b/buildfarm/sqldb.py index a244bd9..5ab71df 100644 --- a/buildfarm/sqldb.py +++ b/buildfarm/sqldb.py @@ -170,8 +170,6 @@ class StormCachingBuildResultStore(BuildResultStore): return build.revision def upload_build(self, build): -super(StormCachingBuildResultStore, self).upload_build(build) -rev, timestamp = build.revision_details() result = self.store.find(StormBuild, StormBuild.checksum == build.log_checksum()) existing_build = result.one() if existing_build is not None: @@ -179,14 +177,16 @@ class StormCachingBuildResultStore(BuildResultStore): assert build.tree == existing_build.tree assert build.host == existing_build.host assert build.compiler == existing_build.compiler -assert rev == existing_build.revision -return +return existing_build +rev, timestamp = build.revision_details() +super(StormCachingBuildResultStore, self).upload_build(build) new_basename = self.build_fname(build.tree, build.host, build.compiler, rev) new_build = StormBuild(new_basename, unicode(build.tree), unicode(build.host), unicode(build.compiler), rev) new_build.checksum = build.log_checksum() new_build.age = build.age_mtime() new_build.status = unicode(str(build.status())) self.store.add(new_build) +return new_build class StormCachingBuildFarm(BuildFarm): diff --git a/import-and-analyse.py b/import-and-analyse.py index 4f99b5c..a1cb47f 100755 --- a/import-and-analyse.py +++ b/import-and-analyse.py @@ -38,7 +38,7 @@ def check_and_send_mails(tree, host, compiler, cur, old): old_status = old.status() if not cur_status.regressed_since(old_status): -if opts.verbose = 1: +if opts.verbose = 3: print ... hasn't regressed since %s: %s % (old_rev, old_status) return @@ -83,7 +83,7 @@ The build may have been broken by one of the following commits: for build in buildfarm.get_new_builds(): -if opts.verbose = 1: +if opts.verbose = 2: print Processing %s... % build, if not opts.dry_run: @@ -91,7 +91,7 @@ for build in buildfarm.get_new_builds(): (rev, rev_timestamp) = build.revision_details() -if opts.verbose = 1: +if opts.verbose = 2: print str(build.status()) try: @@ -114,7 +114,8 @@ for build in buildfarm.get_new_builds(): check_and_send_mails(build.tree, build.host, build.compiler, build, prev_build) if not opts.dry_run: -build.remove() +# When the new web script is introduced, kill the build here: +# build.remove() buildfarm.commit() smtp.quit() -- build.samba.org
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 2ec5988 Cope with missing revision info in logs. from 926046e Fix override of status method. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 2ec59886e606ae73e0c9305450a9e1c148b7b41e Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 18:30:56 2010 +0100 Cope with missing revision info in logs. --- Summary of changes: buildfarm/data.py | 13 ++--- buildfarm/sqldb.py|2 +- import-and-analyse.py | 12 ++-- 3 files changed, 21 insertions(+), 6 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/data.py b/buildfarm/data.py index 10966f4..5c25aab 100644 --- a/buildfarm/data.py +++ b/buildfarm/data.py @@ -42,6 +42,13 @@ class BuildSummary(object): BuildStageResult = collections.namedtuple(BuildStageResult, name result) +class MissingRevisionInfo(Exception): +Revision info could not be found in the build log. + +def __init__(self, build): +self.build = build + + class BuildStatus(object): def __init__(self, stages=None, other_failures=None): @@ -259,6 +266,9 @@ class Build(object): finally: f.close() +if revid is None: +raise MissingRevisionInfo(self) + return (revid, timestamp) def status(self): @@ -375,9 +385,6 @@ class BuildResultStore(object): def upload_build(self, build): (rev, rev_timestamp) = build.revision_details() -if not rev: -raise Exception(Unable to find revision in %r log % build) - new_basename = self.build_fname(build.tree, build.host, build.compiler, rev) try: existing_build = self.get_build(build.tree, build.host, build.compiler, rev) diff --git a/buildfarm/sqldb.py b/buildfarm/sqldb.py index 2f65931..709784f 100644 --- a/buildfarm/sqldb.py +++ b/buildfarm/sqldb.py @@ -192,7 +192,7 @@ class StormCachingBuildResultStore(BuildResultStore): new_build = StormBuild(new_basename, unicode(build.tree), unicode(build.host), unicode(build.compiler), rev) new_build.checksum = build.log_checksum() new_build.age = build.age_mtime() -new_build.status = unicode(str(build.status())) +new_build.status_str = unicode(str(build.status())) self.store.add(new_build) return new_build diff --git a/import-and-analyse.py b/import-and-analyse.py index dcfcec6..773a03f 100755 --- a/import-and-analyse.py +++ b/import-and-analyse.py @@ -90,9 +90,17 @@ for build in buildfarm.get_new_builds(): continue if not opts.dry_run: -build = buildfarm.builds.upload_build(build) +try: +build = buildfarm.builds.upload_build(build) +except data.MissingRevisionInfo: +print No revision info in %r, skipping % build +continue -(rev, rev_timestamp) = build.revision_details() +try: +(rev, rev_timestamp) = build.revision_details() +except data.MissingRevisionInfo: +print No revision info in %r, skipping % build +continue if opts.verbose = 2: print str(build.status()) -- build.samba.org
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 7f9ad8f Fix formatting. from 2ec5988 Cope with missing revision info in logs. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 7f9ad8f7fd4ba40d144db6e1215b4bb36f8e2a1f Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 19:31:50 2010 +0100 Fix formatting. --- Summary of changes: import-and-analyse.py |4 +--- 1 files changed, 1 insertions(+), 3 deletions(-) Changeset truncated at 500 lines: diff --git a/import-and-analyse.py b/import-and-analyse.py index 773a03f..ec61fba 100755 --- a/import-and-analyse.py +++ b/import-and-analyse.py @@ -83,9 +83,6 @@ The build may have been broken by one of the following commits: for build in buildfarm.get_new_builds(): -if opts.verbose = 2: -print Processing %s... % build, - if build in buildfarm.builds: continue @@ -103,6 +100,7 @@ for build in buildfarm.get_new_builds(): continue if opts.verbose = 2: +print %s... % build, print str(build.status()) try: -- build.samba.org
[SCM] build.samba.org - branch master updated
The branch, master has been updated via c3b0245 Try a bit harder to not fetch from the upload directory when looking for old builds. from 7f9ad8f Fix formatting. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit c3b0245cc29002085ff3c1e8cdcd6a565ebb5bb9 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 19:42:52 2010 +0100 Try a bit harder to not fetch from the upload directory when looking for old builds. --- Summary of changes: buildfarm/__init__.py |2 +- import-and-analyse.py |4 +++- 2 files changed, 4 insertions(+), 2 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/__init__.py b/buildfarm/__init__.py index 58b66fb..8d593c4 100644 --- a/buildfarm/__init__.py +++ b/buildfarm/__init__.py @@ -130,7 +130,7 @@ class BuildFarm(object): return perc def get_build(self, tree, host, compiler, rev=None): -if rev: +if rev is not None: return self.builds.get_build(tree, host, compiler, rev) else: return self.upload_builds.get_build(tree, host, compiler) diff --git a/import-and-analyse.py b/import-and-analyse.py index ec61fba..ed854ec 100755 --- a/import-and-analyse.py +++ b/import-and-analyse.py @@ -115,11 +115,13 @@ for build in buildfarm.get_new_builds(): # Can't send a nastygram until there are 2 builds.. else: try: -prev_build = buildfarm.get_build(build.tree, build.host, build.compiler, prev_rev) +assert prev_rev is not None +prev_build = buildfarm.builds.get_build(build.tree, build.host, build.compiler, prev_rev) except data.NoSuchBuildError: if opts.verbose = 1: print Previous build %s has disappeared % prev_build else: +assert build.log_checksum() != prev_build.log_checksum() check_and_send_mails(build.tree, build.host, build.compiler, build, prev_build) if not opts.dry_run: -- build.samba.org
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 52a2144 Remove now obsolete hostdb.pm. via 6cccf14 Switch over to using python import-and-analyse. via 90150b8 Fix blame mail sending. from c3b0245 Try a bit harder to not fetch from the upload directory when looking for old builds. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 52a21444c38370270d31dd604934e56727aa8128 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 20:00:40 2010 +0100 Remove now obsolete hostdb.pm. commit 6cccf1436f7a41a8f63051bccade9f4a2ab309ee Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 19:57:47 2010 +0100 Switch over to using python import-and-analyse. commit 90150b8b7b08904b4ec880822f3208a34c2467ec Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 19:56:28 2010 +0100 Fix blame mail sending. --- Summary of changes: build.pm |9 - hostdb.pm | 236 import-and-analyse.pl | 407 - import-and-analyse.py |5 +- revs.sh |2 +- tests/hostdb.pl | 74 - web/trees.conf|2 +- 7 files changed, 4 insertions(+), 731 deletions(-) delete mode 100644 build.pm delete mode 100644 hostdb.pm delete mode 100755 import-and-analyse.pl delete mode 100755 tests/hostdb.pl Changeset truncated at 500 lines: diff --git a/build.pm b/build.pm deleted file mode 100644 index 7de9cc7..000 --- a/build.pm +++ /dev/null @@ -1,9 +0,0 @@ -package BuildFarm; - -use strict; - -use vars qw ( $VERSION ); - -$VERSION = '0.01'; - -1; diff --git a/hostdb.pm b/hostdb.pm deleted file mode 100644 index 372483d..000 --- a/hostdb.pm +++ /dev/null @@ -1,236 +0,0 @@ -#!/usr/bin/perl - -# Samba.org buildfarm -# Copyright (C) 2008 Andrew Bartlett abart...@samba.org -# Copyright (C) 2008 Jelmer Vernooij jel...@samba.org -# -# This program is free software; you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation; either version 3 of the License, or -# (at your option) any later version. -# -# This program is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with this program. If not, see http://www.gnu.org/licenses/. -# - -package hostdb; - -use DBI; -use warnings; -use strict; - -sub new($) - { -my ($class, $filename) = @_; - -my $dbh = DBI-connect(dbi:SQLite:$filename, , , {RaiseError = 1, PrintError = 0, -ShowErrorStatement = 1, AutoCommit = 0}) or return undef; - -my $self = { filename = $filename, dbh = $dbh }; - -bless($self, $class); -} - -sub provision($) -{ - my ($self) = @_; - eval { - $self-{dbh}-do(CREATE TABLE host ( name text, owner text, owner_email text, password text, ssh_access int, fqdn text, platform text, permission text, last_dead_mail int, join_time int );); - - $self-{dbh}-do(CREATE UNIQUE INDEX unique_hostname ON host (name);); - - $self-{dbh}-do(CREATE TABLE build ( id integer primary key autoincrement, tree text, revision text, host text, compiler text, checksum text, age int, status text, commit_revision text);); - $self-{dbh}-do(CREATE UNIQUE INDEX unique_checksum ON build (checksum);); - - $self-{dbh}-do(CREATE TABLE test_run ( build int, test text, result text, output text);); - $self-{dbh}-commit(); - }; - if ($@) { - local $self-{dbh}-{RaiseError} = 0; - $self-{dbh}-rollback(); - print DB Failure: $@; - return 0; - } - return 1; -} - -sub createhost($$) -{ - my ($self, $name, $platform, $owner, $owner_email, $password, $permission) = @_; - my $sth = $self-{dbh}-prepare(INSERT INTO host (name, platform, owner, owner_email, password, permission, join_time) VALUES (?,?,?,?,?,?,?)); - - eval { - $sth-execute($name, $platform, $owner, $owner_email, $password, $permission, time()); - $self-{dbh}-commit(); - }; - if ($@) { - local $self-{dbh}-{RaiseError} = 0; - $self-{dbh}-rollback(); - print DB Failure: $@; - return 0; - } - return 1; -} - -sub deletehost($$) -{ - my ($self, $name) = @_; - my $ret; - my $sth = $self-{dbh}-prepare(DELETE FROM host WHERE name = ?); - - eval { - $ret = $sth-execute($name); -
[SCM] build.samba.org - branch master updated
The branch, master has been updated via cfdeca5 Properly encode dictionary elements. from 52a2144 Remove now obsolete hostdb.pm. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit cfdeca50120dceb9f0bb3e5d161a185c93159def Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 20:09:11 2010 +0100 Properly encode dictionary elements. --- Summary of changes: buildfarm/tests/test_data.py |5 - import-and-analyse.py| 14 -- 2 files changed, 16 insertions(+), 3 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/tests/test_data.py b/buildfarm/tests/test_data.py index 5c233c2..e79224b 100755 --- a/buildfarm/tests/test_data.py +++ b/buildfarm/tests/test_data.py @@ -97,7 +97,10 @@ bla BUILD COMMIT TIME: 3 August 2010 ) build = self.x.get_build(tdb, charis, cc, 12) -self.assertEquals((43, 3 August 2010), build.revision_details()) +(rev, timestamp) = build.revision_details() +self.assertIsInstance(rev, str) +self.assertIsInstance(timestamp, str) +self.assertEquals((43, 3 August 2010), (rev, timestamp)) def test_revision_details_no_timestamp(self): self.create_mock_logfile(tdb, charis, cc, rev=12, contents= diff --git a/import-and-analyse.py b/import-and-analyse.py index 8ce42b2..3891f1e 100755 --- a/import-and-analyse.py +++ b/import-and-analyse.py @@ -69,8 +69,18 @@ See http://build.samba.org/?function=View+Build;host=%(host)s;tree=%(tree)s;comp The build may have been broken by one of the following commits: %(change_log)s - % {tree: tree, host: host, compiler: compiler, change_log: change_log, scm: t.scm, branch: t.branch, -cur_rev: cur_rev, old_rev: old_rev, cur_status: cur_status, old_status: old_status } + % { +tree: tree.encode(utf-8), +host: host.encode(utf-8), +compiler: compiler.encode(utf-8), +change_log: change_log, +scm: t.scm, +branch: t.branch, +cur_rev: cur_rev, +old_rev: old_rev, +cur_status: cur_status, +old_status: old_status, +} msg = MIMEText(body) msg[Subject] = BUILD of %s:%s BROKEN on %s with %s AT REVISION %s % (tree, t.branch, host, compiler, cur_rev) -- build.samba.org
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 4bfd615 Fix nasty gram sending. from cfdeca5 Properly encode dictionary elements. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 4bfd6157b74a824ed22549cf1f22be31ca6f5243 Author: Jelmer Vernooij jel...@samba.org Date: Sat Nov 13 20:11:15 2010 +0100 Fix nasty gram sending. --- Summary of changes: import-and-analyse.py |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) Changeset truncated at 500 lines: diff --git a/import-and-analyse.py b/import-and-analyse.py index 3891f1e..d88ef43 100755 --- a/import-and-analyse.py +++ b/import-and-analyse.py @@ -87,7 +87,7 @@ The build may have been broken by one of the following commits: msg[From] = \Build Farm\ bu...@samba.org msg[To] = ,.join(recipients) if not opts.dry_run: -smtp.send(msg[From], [msg[To]], msg.as_string()) +smtp.sendmail(msg[From], [msg[To]], msg.as_string()) else: print msg.as_string() -- build.samba.org
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated via 7cb0f95 s4-auth: fixed infinite loop in krb5 auth via 6582d47 s4-auth: fixed crash in krb5 auth from f198c90 s3/doc: add missing documentation for vfs_time_audit http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit 7cb0f95bf28c24c45955d6df52f505019341ef31 Author: Andrew Tridgell tri...@samba.org Date: Sun Nov 14 14:26:06 2010 +1100 s4-auth: fixed infinite loop in krb5 auth we were continually trying the first address returned, instead of moving to the next address Autobuild-User: Andrew Tridgell tri...@samba.org Autobuild-Date: Sun Nov 14 04:11:28 UTC 2010 on sn-devel-104 commit 6582d4739c25a879c58fc768290b3f97ecf795c7 Author: Andrew Tridgell tri...@samba.org Date: Sun Nov 14 13:40:02 2010 +1100 s4-auth: fixed crash in krb5 auth remote_addr was used after free --- Summary of changes: source4/auth/kerberos/krb5_init_context.c |5 ++--- 1 files changed, 2 insertions(+), 3 deletions(-) Changeset truncated at 500 lines: diff --git a/source4/auth/kerberos/krb5_init_context.c b/source4/auth/kerberos/krb5_init_context.c index 6727de0..78f3e1f 100644 --- a/source4/auth/kerberos/krb5_init_context.c +++ b/source4/auth/kerberos/krb5_init_context.c @@ -207,7 +207,6 @@ krb5_error_code smb_krb5_send_and_recv_func(krb5_context context, { krb5_error_code ret; NTSTATUS status; - struct socket_address *remote_addr; const char *name; struct addrinfo *ai, *a; struct smb_krb5_socket *smb_krb5; @@ -239,7 +238,8 @@ krb5_error_code smb_krb5_send_and_recv_func(krb5_context context, return ret; } - for (a = ai; a; a = ai-ai_next) { + for (a = ai; a; a = a-ai_next) { + struct socket_address *remote_addr; smb_krb5 = talloc(tmp_ctx, struct smb_krb5_socket); if (!smb_krb5) { talloc_free(tmp_ctx); @@ -291,7 +291,6 @@ krb5_error_code smb_krb5_send_and_recv_func(krb5_context context, talloc_free(smb_krb5); continue; } - talloc_free(remote_addr); /* Setup the FDE, start listening for read events * from the start (otherwise we may miss a socket -- Samba Shared Repository
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated via 52f2520 Fix the unexpected.tdb database problem. Change nmbd to store the transaction id of packets it was requested to send via a client, and only store replies that match these ids. On the client side change clients to always attempt to ask nmbd first for name_query and node_status calls, and then fall back to doing socket calls if we can't talk to nmbd (either nmbd is not running, or we're not root and cannot open the messaging tdb's). Fix readers of unexpected.tdb to delete packets they've successfully read. via 781c4aa Move error reporting of messaging context creation fail into the daemons themselves. Allows client utilities to silently fail to create a messaging context due to access denied on the messaging tdb (which I need for the following patch). from 7cb0f95 s4-auth: fixed infinite loop in krb5 auth http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit 52f252064817b4be4d45d9fdbb0ad07748a0f317 Author: Jeremy Allison j...@samba.org Date: Sat Nov 13 20:32:36 2010 -0800 Fix the unexpected.tdb database problem. Change nmbd to store the transaction id of packets it was requested to send via a client, and only store replies that match these ids. On the client side change clients to always attempt to ask nmbd first for name_query and node_status calls, and then fall back to doing socket calls if we can't talk to nmbd (either nmbd is not running, or we're not root and cannot open the messaging tdb's). Fix readers of unexpected.tdb to delete packets they've successfully read. This should fix a long standing problem of unexpected.tdb growing out of control in noisy NetBIOS envioronments with lots of bradcasts, yet still allow unprivileged client apps to work mostly as well as they already did (nmblookup for example) in an environment when nmbd isn't running. Jeremy. Autobuild-User: Jeremy Allison j...@samba.org Autobuild-Date: Sun Nov 14 05:22:45 UTC 2010 on sn-devel-104 commit 781c4aabb87e63df77c76a360b6ed5f6a20e6d58 Author: Jeremy Allison j...@samba.org Date: Sat Nov 13 20:28:41 2010 -0800 Move error reporting of messaging context creation fail into the daemons themselves. Allows client utilities to silently fail to create a messaging context due to access denied on the messaging tdb (which I need for the following patch). Jeremy. --- Summary of changes: source3/include/proto.h|2 + source3/lib/messages.c |4 +- source3/lib/messages_local.c |2 +- source3/lib/server_contexts.c |3 - source3/libsmb/clidgram.c | 14 +++-- source3/libsmb/clidgram.h |4 +- source3/libsmb/dsgetdcname.c |8 ++- source3/libsmb/namequery.c | 35 +-- source3/libsmb/unexpected.c| 133 +++- source3/nmbd/nmbd.c| 25 +++- source3/nmbd/nmbd_packets.c|7 ++- source3/smbd/globals.c |6 ++- source3/winbindd/winbindd.c| 10 +++ source3/winbindd/winbindd.h|1 - source3/winbindd/winbindd_cm.c |7 ++- 15 files changed, 219 insertions(+), 42 deletions(-) Changeset truncated at 500 lines: diff --git a/source3/include/proto.h b/source3/include/proto.h index 10409a4..e31d30d 100644 --- a/source3/include/proto.h +++ b/source3/include/proto.h @@ -2907,6 +2907,8 @@ NTSTATUS change_trust_account_password( const char *domain, const char *remote_m /* The following definitions come from libsmb/unexpected.c */ +bool is_requested_send_packet(struct packet_struct *p); +bool store_outstanding_send_packet(struct packet_struct *p); void unexpected_packet(struct packet_struct *p); void clear_unexpected(time_t t); struct packet_struct *receive_unexpected(enum packet_type packet_type, int id, diff --git a/source3/lib/messages.c b/source3/lib/messages.c index 11dc931..41f67d0 100644 --- a/source3/lib/messages.c +++ b/source3/lib/messages.c @@ -194,7 +194,7 @@ struct messaging_context *messaging_init(TALLOC_CTX *mem_ctx, status = messaging_tdb_init(ctx, ctx, ctx-local); if (!NT_STATUS_IS_OK(status)) { - DEBUG(0, (messaging_tdb_init failed: %s\n, + DEBUG(2, (messaging_tdb_init failed: %s\n, nt_errstr(status))); TALLOC_FREE(ctx); return NULL; @@ -205,7 +205,7 @@ struct messaging_context *messaging_init(TALLOC_CTX *mem_ctx, status = messaging_ctdbd_init(ctx, ctx, ctx-remote); if (!NT_STATUS_IS_OK(status)) { - DEBUG(1, (messaging_ctdb_init failed: %s\n, + DEBUG(2, (messaging_ctdb_init failed: %s\n, nt_errstr(status)));
Build status as of Sun Nov 14 07:00:01 2010
URL: http://build.samba.org/ --- /home/build/master/cache/broken_results.txt.old 2010-11-13 00:00:04.0 -0700 +++ /home/build/master/cache/broken_results.txt 2010-11-14 00:00:03.0 -0700 @@ -1,22 +1,22 @@ -Build status as of Sat Nov 13 07:00:02 2010 +Build status as of Sun Nov 14 07:00:01 2010 Build counts: Tree Total Broken Panic build_farm 0 0 0 -ccache 37 5 0 -ccache-maint 36 4 0 -ldb 37 26 0 -libreplace 37 37 0 +ccache 36 5 0 +ccache-maint 35 5 0 +ldb 36 26 0 +libreplace 36 36 0 lorikeet 0 0 0 -pidl 21 21 0 -ppp 15 0 0 -rsync37 15 0 +pidl 20 20 0 +ppp 14 0 0 +rsync36 14 0 samba-docs 0 0 0 samba-web0 0 0 -samba_3_current 37 27 1 -samba_3_master 37 20 0 -samba_3_next 37 36 0 -samba_4_0_test 42 37 2 -talloc 37 15 0 -tdb 35 14 0 +samba_3_current 36 26 0 +samba_3_master 36 19 0 +samba_3_next 36 35 0 +samba_4_0_test 42 36 0 +talloc 36 15 0 +tdb 34 15 0