Re: [Samba] Multi samba domain in one LDAP Backend with multi-site authentication
Thanks Gaiseric for your answer, I know this things about trust relashionship even if i still don't have setup one, but we need to have only one LDAP backend, to allow others applications to authenticate user with LDAP. We can't specifie more than one backend in our application. I've thought that i could create different OU with each domain, and configure smbldap-tools and pam to work with this OU, with a base like : dc=DOMAIN, dc=company, dc=com , and replicate this LDAP on other site. But is it possible to use trust relationship with this kind of LDAP structure ? will i need IDMAP ? Thanks, Thibault -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] I can not create Home DIR
Claudio Guzman wrote: when users enter the system can not see your home folder, or it asks the username and password denuevo. My configuration is Server + Samba + LDAP PDC Samba server that only has the shared folders and are accessed via winbind and pam # Global settings [global] display charset = LOCALE passwd chat = *New*password* %n\n *Retype*new*password* %n\n *passwd:*all*authentication*tokens*updated*successfully* idmap gid = 1000-33554431 passwd program = /usr/bin/passwd %u netbios name = srvsamba idmap uid = 1000-33554431 dos charset = CP850 local master = no workgroup = SERVER debug level = 9 os level = 0 security = domain log file = /var/log/samba/%m.log guest account = nobody smb passwd file = /etc/samba/smbpasswd load printers = no socket options = TCP_NODELAY SO_RCVBUF=8192 SO_SNDBUF=8192 wins server = 192.168.1.252 map to guest = Bad User domain master = no encrypt passwords = yes realm = template shell = /bin/false server string = srvsamba winbind enum users = Yes password server = 192.168.1.252 template homedir = /mnt/samba/home/%U winbind enum groups = Yes unix charset = UTF-8 preferred master = no pam password change = yes winbind use default domain = no [homes] comment = Home Directories browseable = no writeable = yes What OS is Samba running on? I have seen this sort of thing when the user's home directory from the Unix side (passwd or LDAP) is a symlink rather than a real directory, on CentOS 5. SELinux prevents Samba from using the symlink, so I have to make sure all users' home directory entries point to the real thing. To check for SELinux errors use aureport --avc, or see if there's anything useful in the samba logs. Moray. To err is human. To purr, feline -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] Can't connect to printers from Win 2008 and Win 7
Thank You for making that clear! New problem. I have been trying to set up a temporary Samba server to handle printing until this problem gets fixed. I am running 3.3.10 on a Sun Solaris 10 machine. From a Win 2003 (32 bit) server I can install print drivers and print ok but from Win 2008 R2 and Win 7 it´s not working. In Win 2008 R2, if I try to connect to a printer with no drivers installed on the server I get a dialog box stating that there are no drivers installed on the server. This is as expected. However, if I go to properties there is NO dialog about missing drivers. Clicking on the Advanced tab and then New Driver starts the Add Printer wizard but when I get to the Printer Driver Selection there are no drivers to choose from. All I can see is the text (Unable to find any drivers for this device) If I try to connect to a printer that has a driver installed, I get error 0x06d1 Any ideas? Åke Holmlund --On 29 januari 2010 10.09.57 -0800 Jeremy Allison j...@samba.org wrote: On Fri, Jan 29, 2010 at 05:46:14PM +0100, Åke Holmlund wrote: Hello! I have sent this a few days ago but I don't think it made it to the mailinglist. We have problems connecting to printers (spoolss) from Windows Server 2008 (64-bit, domain member) and Windows 7 (64-bit, NOT domain member). When trying to connect to the printers we get an error message saying Windows cannot connect to the printer. Operation could not be completed (error 0x000d). Connecting as a standard user or a domain administrator makes no difference. If I try to add a printer on the server from Windows 2008, there are no ports to choose from in the Choose a printer port dialog. Connecting to the printers is no problem from Win XP, Win Vista and Win 2003 (all 32-bit). When trying to add a printer from Windows 2003, the Samba Printer Port shows up in the Choose a printer port dialog. The Samba (domain) server is 3.4.4 running on Solaris 10 (Sparc). Any ideas what the problem can be or where to start looking? Ok, here's the deal. If you have 64-bit Windows clients at the moment you need to be using 3.3.10, not anything later. If you're using 32-bit Windows clients, you can use 3.4.5 or later. The reason (and Guenther can correct me if I'm wrong), is that in 3.4.x we changed from the hand-marshalled SPOOLSS RPC we used in 3.3.x, which was mostly correct after being worked on for many years, to pidl-generated SPOOLSS RPC directly from the idl files. Now our idl files are correct, but it turns out that the Windows idl parser for SPOOLSS is itself custom, and won't accept the normally marshalled RPC packets that the pidl-generated code creates. Guenter and Metze have been doing a lot of work on fixing this, and this is why 32-bit Windows clients work in 3.4.5. However there are still some changes needed for 64-bit Windows clients, and this will probably make it into an early patch for 3.5.x (it won't make 3.5.0, as it's too late to test the changes needed before we ship next month). I'm going to write this up as a tech-note for the next 3.4.x release and for 3.5.0 and get it into the release notes. Hope this hasn't caused you too many problems. Jeremy. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] [Announce] Samba 3.5.0rc2 Available for Download
On Sun, Jan 31, 2010 at 11:05 PM, Christian PERRIER bubu...@debian.orgwrote: Quoting Karolin Seeger (ksee...@samba.org): Release Announcements = This is the second release candidate of Samba 3.5. This is *not* Debian packages for 3.5.0rc2 were uploaded to Debian experimental as of Sunday Jan 31st. Many special thanks to Michael Adam for bringing a new configure option for the *.dat files location. That allowed us, Debian maintainers, to drop the very last bit of code changes meant to deal with file locations that were not fitting the Debian view of the FHS. The Debian diff is now very minimally restricted to Debian specific changes that are not suitable for upstream inclusion (mostly documentation stuff). The Debian packaging team for Samba will now discuss the opportunity to bring 3.5 packages in Debian unstable: - release Debian squeeze with 3.5 instead of the now quite rock solid 3.4? - consequences for Ubuntu? Anyway, we again renew public thanks to Karolin's for her tremendous job in release management. The entire Samba Team also deserves thanks for their commitment to follow their release manager. A predictable upstream release policy is the best that can happen to distro packagers. This is very exciting, thank you all for your hard work. We would like to see 3.5 in Squeeze soon as we will most likely be deploying Windows 7 2nd quarter this year. Thanks, Robert LeBlanc Life Sciences Undergraduate Education Computer Support Brigham Young University -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] Samba Load Balancing with DNS round robin
I want to add another server to my existing Samba environment and use DNS round robin to do load balancing. Both will use same smb.conf. Is anyone out there using this kind of setup? I have few quick questions: 1. Can interfaces statement be used in DNS round robin. If yes, then how? 2. Should preferred master= yes be defined on both boxes? Thanks. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] Multi samba domain in one LDAP Backend with multi-site authentication
You should be able to have separate OU's in LDAP for each domain. From the point-of-view of samba, each samba PDC would only know about its own section of the LDAP tree- and in effect would be the same as separate LDAP servers. You would want to make sure that the underlying unix authentication on that server would also only used the domain-specific section of LDAP for user authentication. You would still want to use idmap. Do you need all domains trusting each other- or is it a series of trusts between each remote office and the central office.A full mesh could get really messy, although I haven't tried this yet. If you have six domains, potentially each domain OU is could contain 5 ou's for idmap entries. I think you could have one, enterprise-wide idmap ldap section, with a ou underneath for each domain.You might even want to make them read-only for most domains so that the entries stay consistent. By default, winbind allocates uid and gid's from a dynamic range that does NOT overlap the uid and gid ranges for local users.I personally would either modify entries as they are created, or actually prepopulate entries, so that the uid/gid entries in the idmap are the same as the real uid/gid values for the actual unix account.I used to do this when using idmap with member servers with in the domain to make sure that when a windows user created or modified a file on an NFS shared, a consistent unix uid was used between windows and nfs on all servers. So your LDAP structure would be similar to ou=DomainA ou=people,ou=DomainA ou=DomainB ou=people,ou=DomainB ... ou=DomainF ou=people,ou=DomainF ... ou=IDMAP ou=DomainA,ou=IDMAP ou=DomainB,ou=IDMAP ... ou=DomainF,ou=IDMAP I don't know if OpenLDAP handles multi-master replication (I am using Sun Directory Server.) Assuming it did, each site could have its own LDAP server but in the same LDAP tree. the other approach would be to have a separate LDAP server and structure for each site BUT configure referrals between each LDAP server (i.e. on ServerA, ou=DomainB points to ou=DomainB on ServerB) to create the appearance of a single LDAP tree. On 02/01/10 04:02, Thibault Vançon wrote: Thanks Gaiseric for your answer, I know this things about trust relashionship even if i still don't have setup one, but we need to have only one LDAP backend, to allow others applications to authenticate user with LDAP. We can't specifie more than one backend in our application. I've thought that i could create different OU with each domain, and configure smbldap-tools and pam to work with this OU, with a base like : dc=DOMAIN, dc=company, dc=com , and replicate this LDAP on other site. But is it possible to use trust relationship with this kind of LDAP structure ? will i need IDMAP ? Thanks, Thibault -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] Printer sharing with 64-bit Windows clients
Jeremy Allison wrote: Ok, here's the deal. If you have 64-bit Windows clients at the moment you need to be using 3.3.10, not anything later. If you're using 32-bit Windows clients, you can use 3.4.5 or later. I'm going to write this up as a tech-note for the next 3.4.x release and for 3.5.0 and get it into the release notes. Hope this hasn't caused you too many problems. Just to add my 2cents. We're going to see alot more of this problem as Win7 64bit becomes the standard oem install for new systems. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] How can i build samba server on uClinux platform
Hi, all How can i build samba server on uClinux platform? My hardware platform is samsung s3c44b0 chip (nommu). Does anyone has idea for that? Thanks a lots~! -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] Printer sharing with 64-bit Windows clients
On Mon, Feb 01, 2010 at 12:53:12PM -0500, Ryan Suarez wrote: Jeremy Allison wrote: Ok, here's the deal. If you have 64-bit Windows clients at the moment you need to be using 3.3.10, not anything later. If you're using 32-bit Windows clients, you can use 3.4.5 or later. I'm going to write this up as a tech-note for the next 3.4.x release and for 3.5.0 and get it into the release notes. Hope this hasn't caused you too many problems. Just to add my 2cents. We're going to see alot more of this problem as Win7 64bit becomes the standard oem install for new systems. I know. We're working on it as fast as we can... -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] [Announce] Samba 3.5.0rc2 Available for Download
On 02/01/2010 12:05 AM, Christian PERRIER wrote: Quoting Karolin Seeger (ksee...@samba.org): Release Announcements = This is the second release candidate of Samba 3.5. This is *not* Debian packages for 3.5.0rc2 were uploaded to Debian experimental as of Sunday Jan 31st. Many special thanks to Michael Adam for bringing a new configure option for the *.dat files location. That allowed us, Debian maintainers, to drop the very last bit of code changes meant to deal with file locations that were not fitting the Debian view of the FHS. The Debian diff is now very minimally restricted to Debian specific changes that are not suitable for upstream inclusion (mostly documentation stuff). The Debian packaging team for Samba will now discuss the opportunity to bring 3.5 packages in Debian unstable: - release Debian squeeze with 3.5 instead of the now quite rock solid 3.4? - consequences for Ubuntu? Anyway, we again renew public thanks to Karolin's for her tremendous job in release management. The entire Samba Team also deserves thanks for their commitment to follow their release manager. A predictable upstream release policy is the best that can happen to distro packagers. Any chance that there will be included in this release, a fix for the libkrb5-3 1.8 problem with winbind? Or will that fix have to come from the libkrb5-3 maintainers? -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] [Announce] Samba 3.5.0rc2 Available for Download
Quoting Dale Schroeder (d...@briannassaladdressing.com): On 02/01/2010 12:05 AM, Christian PERRIER wrote: Quoting Karolin Seeger (ksee...@samba.org): Release Announcements = This is the second release candidate of Samba 3.5. This is *not* Debian packages for 3.5.0rc2 were uploaded to Debian experimental as of Sunday Jan 31st. Many special thanks to Michael Adam for bringing a new configure option for the *.dat files location. That allowed us, Debian maintainers, to drop the very last bit of code changes meant to deal with file locations that were not fitting the Debian view of the FHS. The Debian diff is now very minimally restricted to Debian specific changes that are not suitable for upstream inclusion (mostly documentation stuff). The Debian packaging team for Samba will now discuss the opportunity to bring 3.5 packages in Debian unstable: - release Debian squeeze with 3.5 instead of the now quite rock solid 3.4? - consequences for Ubuntu? Anyway, we again renew public thanks to Karolin's for her tremendous job in release management. The entire Samba Team also deserves thanks for their commitment to follow their release manager. A predictable upstream release policy is the best that can happen to distro packagers. Any chance that there will be included in this release, a fix for the libkrb5-3 1.8 problem with winbind? Or will that fix have to come from the libkrb5-3 maintainers? ** CRM114 Whitelisted by: lists.samba.org ** -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] [Announce] Samba 3.5.0rc2 Available for Download
On Mon, Feb 01, 2010 at 07:05:02AM +0100, Christian PERRIER wrote: Anyway, we again renew public thanks to Karolin's for her tremendous job in release management. Hear hear ! I'd also like to publicly thank Karolin for the wonderful job she is doing as release manager ! -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] DOS dir command on Samba share
Hi, I was wondering if anyone could shed some light on a problem I am having with the DOS DIR command on a SMB share. I have a batch script that copies files from one mapped network drive to another. To check that the copy has been sucsessful it creates two text files with the output of the command dir /s directory | find File on both directories. We have recently upgraded one of the servers from Windows to Samba running on RHEL 5. For some reason now the output of the dir command comes in a different order and the total byte size is different on the samba share than the native Windows share thus breaking the check. All the file appear to by copied correctly though. I have attached a sample of what happens. Any help is greatly appreciated. Regards, Braxton Neate 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 4 File(s)484,509 bytes 9 File(s) 15,456,617 bytes 32 File(s)103,890,559 bytes 9 File(s) 9,493,444 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 1 File(s)501 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes Total Files Listed: 55 File(s)129,325,630 bytes 0 File(s) 0 bytes 4 File(s)488,605 bytes 32 File(s)103,890,559 bytes 9 File(s) 15,456,617 bytes 9 File(s) 9,493,444 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 0 File(s) 0 bytes 1 File(s)501 bytes Total Files Listed: 55 File(s)129,329,726 bytes-- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] DOS dir command on Samba share
Braxton Neate put forth on 2/1/2010 9:37 PM: Hi, I was wondering if anyone could shed some light on a problem I am having with the DOS DIR command on a SMB share. I have a batch script that copies files from one mapped network drive to another. To check that the copy has been sucsessful it creates two text files with the output of the command dir /s directory | find File on both directories. We have recently upgraded one of the servers from Windows to Samba running on RHEL 5. For some reason now the output of the dir command comes in a different order and the total byte size is different on the samba share than the native Windows share thus breaking the check. All the file appear to by copied correctly though. I have attached a sample of what happens. Any help is greatly appreciated. You should be able to fix the directory order issue by setting the environment variable set dircmd=/ogn on the Windows machine running the script. The file size discrepancy probably has to do with the fact that you changed the underlying filesystem from (I assume) NTFS to one of EXT2/3/4, ReiserFS, XFS, or JFS. Each filesystem may report slightly different sizes for files and directories. This problem may be more difficult to solve. Are you currently using? fstype = NTFS or fstype = Samba You may try swapping those and see if it fixes the file size issue. I have no idea if this will help, but you might try it. However, keep in mind that any software that attempts to store information in NTFS alternate data streams will be unable to do so. For example, the Windows 2000 file viewer, Windows Explorer, stores image file thumbnail data in ADS on NTFS filesystems. Thus, if you enable fstype = NTFS and the underlying FS is say, ext3 or XFS, thumbnail data will not be written to the AFS as it doesn't exist, and on top of that, Explorer will not automatically fall back and create a thumbs.db file, which is how thumbnail data is handled on FAT32 and all filesystems *other than* NTFS. Thus, if one has thumbnail view enabled on a Samba share containing hundreds or thousands of photos, the thumbnail data must be regenerated, for every file, each time the share is viewed, and this is _painfully_ slow. I took me more a few hours to figure this out. So, again, exporting a *nix Samba share as NTFS will cause unforeseen problems with applications that expect the AFS to exist on the filesystem. It seems your current verification method may not work in the future. Have you looked into a fully automated, purpose built tool for this task, such as rsync? It may give you some more options and flexibility to accomplish your goals. http://samba.anu.edu.au/rsync/ -- Stan -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] DOS dir command on Samba share
Stan Hoeppner put forth on 2/1/2010 10:48 PM: Braxton Neate put forth on 2/1/2010 9:37 PM: Hi, I was wondering if anyone could shed some light on a problem I am having with the DOS DIR command on a SMB share. I have a batch script that copies files from one mapped network drive to another. To check that the copy has been sucsessful it creates two text files with the output of the command dir /s directory | find File on both directories. We have recently upgraded one of the servers from Windows to Samba running on RHEL 5. For some reason now the output of the dir command comes in a different order and the total byte size is different on the samba share than the native Windows share thus breaking the check. All the file appear to by copied correctly though. I have attached a sample of what happens. Any help is greatly appreciated. You should be able to fix the directory order issue by setting the environment variable set dircmd=/ogn on the Windows machine running the script. The file size discrepancy probably has to do with the fact that you changed the underlying filesystem from (I assume) NTFS to one of EXT2/3/4, ReiserFS, XFS, or JFS. Each filesystem may report slightly different sizes for files and directories. This problem may be more difficult to solve. Are you currently using? fstype = NTFS or fstype = Samba You may try swapping those and see if it fixes the file size issue. I have no idea if this will help, but you might try it. However, keep in mind that any software that attempts to store information in NTFS alternate data streams will be unable to do so. For example, the Windows 2000 file viewer, Windows Explorer, stores image file thumbnail data in ADS on NTFS filesystems. Thus, if you enable fstype = NTFS and the underlying FS is say, ext3 or XFS, thumbnail data will not be written to the AFS as it doesn't exist, and on top of that, Explorer will not automatically fall back and create a thumbs.db file, which is how thumbnail data is handled on FAT32 and all filesystems *other than* NTFS. Thus, if one has thumbnail view enabled on a Samba share containing hundreds or thousands of photos, the thumbnail data must be regenerated, for every file, each time the share is viewed, and this is _painfully_ slow. I took me more a few hours to figure this out. So, again, exporting a *nix Samba share as NTFS will cause unforeseen problems with applications that expect the AFS to exist on the filesystem. It seems your current verification method may not work in the future. Have you looked into a fully automated, purpose built tool for this task, such as rsync? It may give you some more options and flexibility to accomplish your goals. http://samba.anu.edu.au/rsync/ Come to think of it, these file size differences being reported may be directly related to existing alternate data streams on your NTFS volume. -- Stan -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] Different file permissions on new file
Alexander Busam schrieb: Hello list :-) I got different file permissions on a new created/copied file. If I copy the file to \\192.168.1.25\data I got -rwx--1 advantage advantage 0 Jan 18 11:16 test.txt If I copy the file to \\192.168.1.25\data\pix I got -rwxrwxr--+ 1 advantage advantage 0 Jan 18 11:18 test.txt The permissions of the path to the share are: drwxrwxr-x+ 14 advantage advantage 4096 Jan 18 11:16 data What is wrong? Best regards, Alex The smb.conf: # Samba config file created using SWAT # from 127.0.0.1 (127.0.0.1) # Date: 2010/01/18 11:22:35 [global] workgroup = HMDBSRV server string = map to guest = Bad Password printcap name = cups add machine script = /usr/sbin/useradd -c Machine -d /var/lib/nobody -s /bin/false %m$ logon script = scripts\login.bat logon path = \\%L\profiles\.msprofile logon drive = L: logon home = \\%L\%U\.9xprofile domain master = No ldap admin dn = cn=xxx,dc=,dc=de ldap group suffix = ou=groups ldap machine suffix = ou=hosts ldap passwd sync = Yes ldap suffix = dc=,dc=de ldap ssl = no ldap user suffix = ou=users usershare allow guests = Yes printing = cups cups options = raw print command = lpq command = %p lprm command = oplocks = No [data] comment = Datenbank path = /data force user = advantage force group = advantage read only = No guest ok = Yes preserve case = No level2 oplocks = No Can anybody tell me what's wrong ? -- 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 5310596... tdb: fix an early release of the global lock that can
cause data corruption
from 42f5125... s4-smbtorture: check for RouterReplyPrinterEx packets
inside backchannel in RPC-SPOOLSS-NOTIFY.
http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master
- Log -
commit 531059696e17d1ee538310d81af309c107d08e3e
Author: Volker Lendecke v...@samba.org
Date: Fri Jan 29 18:21:09 2010 +0100
tdb: fix an early release of the global lock that can cause data corruption
There was a bug in tdb where the
tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
(ending the transaction-mutex) was done before the
/* remove the recovery marker */
This means that when a transaction is committed there is a window where
another
opener of the file sees the transaction marker while the transaction
committer
is still fully functional and working on it. This led to transaction being
rolled back by that second opener of the file while transaction_commit()
gave
no error to the caller.
This patch moves the F_UNLCK to after the recovery marker was removed,
closing
this window.
---
Summary of changes:
lib/tdb/common/transaction.c | 15 ++-
1 files changed, 10 insertions(+), 5 deletions(-)
Changeset truncated at 500 lines:
diff --git a/lib/tdb/common/transaction.c b/lib/tdb/common/transaction.c
index 20f2bfc..b8988ea 100644
--- a/lib/tdb/common/transaction.c
+++ b/lib/tdb/common/transaction.c
@@ -135,6 +135,9 @@ struct tdb_transaction {
bool prepared;
tdb_off_t magic_offset;
+ /* set when the GLOBAL_LOCK has been taken */
+ bool global_lock_taken;
+
/* old file size before transaction */
tdb_len_t old_map_size;
@@ -603,6 +606,11 @@ int _tdb_transaction_cancel(struct tdb_context *tdb)
}
}
+ if (tdb-transaction-global_lock_taken) {
+ tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
+ tdb-transaction-global_lock_taken = false;
+ }
+
/* remove any global lock created during the transaction */
if (tdb-global_lock.count != 0) {
tdb_brlock(tdb, FREELIST_TOP, F_UNLCK, F_SETLKW, 0,
4*tdb-header.hash_size);
@@ -947,11 +955,12 @@ static int _tdb_transaction_prepare_commit(struct
tdb_context *tdb)
return -1;
}
+ tdb-transaction-global_lock_taken = true;
+
if (!(tdb-flags TDB_NOSYNC)) {
/* write the recovery data to the end of the file */
if (transaction_setup_recovery(tdb,
tdb-transaction-magic_offset) == -1) {
TDB_LOG((tdb, TDB_DEBUG_FATAL,
tdb_transaction_prepare_commit: failed to setup recovery data\n));
- tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
_tdb_transaction_cancel(tdb);
return -1;
}
@@ -966,7 +975,6 @@ static int _tdb_transaction_prepare_commit(struct
tdb_context *tdb)
tdb-transaction-old_map_size) ==
-1) {
tdb-ecode = TDB_ERR_IO;
TDB_LOG((tdb, TDB_DEBUG_FATAL,
tdb_transaction_prepare_commit: expansion failed\n));
- tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
_tdb_transaction_cancel(tdb);
return -1;
}
@@ -1056,7 +1064,6 @@ int tdb_transaction_commit(struct tdb_context *tdb)
tdb_transaction_recover(tdb);
_tdb_transaction_cancel(tdb);
- tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
TDB_LOG((tdb, TDB_DEBUG_FATAL, tdb_transaction_commit:
write failed\n));
return -1;
@@ -1072,8 +1079,6 @@ int tdb_transaction_commit(struct tdb_context *tdb)
return -1;
}
- tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
-
/*
TODO: maybe write to some dummy hdr field, or write to magic
offset without mmap, before the last sync, instead of the
--
Samba Shared Repository
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated
via 93142e4... s4/ldif: Handle Schema:prefixMap blobs in W2K3 and W2K8
via 83068b4... s4/ldif: Better control on ldif_write_NDR() errors
processing
via a342e3a... s4/idl: drsblobs IDL regeneration
via 20db321... s4/drsblobs: Custom ndr_print_ implementation for
drsuapi_MSPrefixMap_Entry
via cd3d165... s4/idl: PrefixMap description for W2K3 and W2K8
Schema:prefixMap attribute
from 5310596... tdb: fix an early release of the global lock that can
cause data corruption
http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master
- Log -
commit 93142e4a3a0a7dbe5f8ba737bd45f01c61d997da
Author: Kamen Mazdrashki kamen.mazdras...@postpath.com
Date: Wed Jan 20 18:10:05 2010 +0200
s4/ldif: Handle Schema:prefixMap blobs in W2K3 and W2K8
Signed-off-by: Stefan Metzmacher me...@samba.org
commit 83068b42ae22af18c5a06dc05a86021db0e25634
Author: Kamen Mazdrashki kamen.mazdras...@postpath.com
Date: Wed Jan 20 17:58:39 2010 +0200
s4/ldif: Better control on ldif_write_NDR() errors processing
Current implementation mask NDR_ errors implicitly.
Thus the caller has no opportunity handle such an error.
Signed-off-by: Stefan Metzmacher me...@samba.org
commit a342e3a419058e28f073a6c5b96ede15a8bdec91
Author: Kamen Mazdrashki kamen.mazdras...@postpath.com
Date: Tue Jan 19 16:54:09 2010 +0200
s4/idl: drsblobs IDL regeneration
Signed-off-by: Stefan Metzmacher me...@samba.org
commit 20db3211ae406004425dd2e2b4294b03b4c4ecbf
Author: Kamen Mazdrashki kamen.mazdras...@postpath.com
Date: Tue Jan 19 16:53:32 2010 +0200
s4/drsblobs: Custom ndr_print_ implementation for drsuapi_MSPrefixMap_Entry
Signed-off-by: Stefan Metzmacher me...@samba.org
commit cd3d165778e51987f185a58ff49c124a555e14e7
Author: Kamen Mazdrashki kamen.mazdras...@postpath.com
Date: Tue Jan 19 16:52:10 2010 +0200
s4/idl: PrefixMap description for W2K3 and W2K8 Schema:prefixMap attribute
Signed-off-by: Stefan Metzmacher me...@samba.org
---
Summary of changes:
librpc/gen_ndr/drsblobs.h | 12
librpc/gen_ndr/ndr_drsblobs.c | 103 +
librpc/gen_ndr/ndr_drsblobs.h |5 ++
librpc/idl/drsblobs.idl | 22 +++
librpc/ndr/ndr_drsblobs.c | 25
librpc/ndr/ndr_drsblobs.h |2 +-
source4/lib/ldb-samba/ldif_handlers.c | 45 +++---
7 files changed, 203 insertions(+), 11 deletions(-)
Changeset truncated at 500 lines:
diff --git a/librpc/gen_ndr/drsblobs.h b/librpc/gen_ndr/drsblobs.h
index 57712a4..7fa201d 100644
--- a/librpc/gen_ndr/drsblobs.h
+++ b/librpc/gen_ndr/drsblobs.h
@@ -134,6 +134,18 @@ struct partialAttributeSetBlob {
union partialAttributeSetCtr ctr;/* [switch_is(version)] */
}/* [public] */;
+struct drsuapi_MSPrefixMap_Entry {
+ uint16_t entryID;
+ uint16_t length;
+ uint8_t *binary_oid;
+}/* [noprint,flag(LIBNDR_FLAG_NOALIGN)] */;
+
+struct drsuapi_MSPrefixMap_Ctr {
+ uint32_t num_entries;
+ uint32_t __ndr_size;/*
[value(ndr_size_drsuapi_MSPrefixMap_Ctr(r,ndr-iconv_convenience,ndr-flags))]
*/
+ struct drsuapi_MSPrefixMap_Entry *entries;
+}/* [gensize,public] */;
+
enum prefixMapVersion
#ifndef USE_UINT_ENUMS
{
diff --git a/librpc/gen_ndr/ndr_drsblobs.c b/librpc/gen_ndr/ndr_drsblobs.c
index 3102251..4290ce5 100644
--- a/librpc/gen_ndr/ndr_drsblobs.c
+++ b/librpc/gen_ndr/ndr_drsblobs.c
@@ -1213,6 +1213,109 @@ _PUBLIC_ void ndr_print_partialAttributeSetBlob(struct
ndr_print *ndr, const cha
ndr-depth--;
}
+static enum ndr_err_code ndr_push_drsuapi_MSPrefixMap_Entry(struct ndr_push
*ndr, int ndr_flags, const struct drsuapi_MSPrefixMap_Entry *r)
+{
+ {
+ uint32_t _flags_save_STRUCT = ndr-flags;
+ ndr_set_flags(ndr-flags, LIBNDR_FLAG_NOALIGN);
+ if (ndr_flags NDR_SCALARS) {
+ NDR_CHECK(ndr_push_align(ndr, 2));
+ NDR_CHECK(ndr_push_uint16(ndr, NDR_SCALARS,
r-entryID));
+ NDR_CHECK(ndr_push_uint16(ndr, NDR_SCALARS, r-length));
+ NDR_CHECK(ndr_push_array_uint8(ndr, NDR_SCALARS,
r-binary_oid, r-length));
+ NDR_CHECK(ndr_push_trailer_align(ndr, 2));
+ }
+ if (ndr_flags NDR_BUFFERS) {
+ }
+ ndr-flags = _flags_save_STRUCT;
+ }
+ return NDR_ERR_SUCCESS;
+}
+
+static enum ndr_err_code ndr_pull_drsuapi_MSPrefixMap_Entry(struct ndr_pull
*ndr, int ndr_flags, struct drsuapi_MSPrefixMap_Entry *r)
+{
+ {
+ uint32_t _flags_save_STRUCT = ndr-flags;
+ ndr_set_flags(ndr-flags, LIBNDR_FLAG_NOALIGN);
+ if (ndr_flags
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated
via 431d75f... s4:NBT-WINS: test large scopes
via 70779a4... s4:NBT-WINS: pass the expected rcode of the name
registration to the test code
via c1d8dc0... s4:NBT-WINSREPLICATION: test replication with names
including scopes
via 331505f... s4:NBT-WINSREPLICATION: fix compiler warnings
via 07b06e5... s4:NBT-WINSREPLICATION: use an array of nbt_names to
loop over different names
via ea7ec4f... s4:winsserver: reject name registrations with a scope
length 237
via 7a02a2b... s4:wrepl_server: truncate the scope of a netbios name to
237 bytes as Windows 2008 does
via e37dc56... libcli/nbt: fix ndr_push_nbt_string() string labels with
a length of 63 (0x3F) are allowed
from 93142e4... s4/ldif: Handle Schema:prefixMap blobs in W2K3 and W2K8
http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master
- Log -
commit 431d75fc9c7f942dcbd31f80380dda59887ac229
Author: Stefan Metzmacher me...@samba.org
Date: Mon Feb 1 15:32:37 2010 +0100
s4:NBT-WINS: test large scopes
metze
commit 70779a46a7b7f6e5c00eba048dbf653d86bc07d4
Author: Stefan Metzmacher me...@samba.org
Date: Mon Feb 1 14:55:14 2010 +0100
s4:NBT-WINS: pass the expected rcode of the name registration to the test
code
metze
commit c1d8dc01fc6f0fa3a22a405419917d880b832761
Author: Stefan Metzmacher me...@samba.org
Date: Sat Jan 30 10:50:33 2010 +0100
s4:NBT-WINSREPLICATION: test replication with names including scopes
metze
commit 331505f09d61cdc5c920293fc36ac1853235177b
Author: Stefan Metzmacher me...@samba.org
Date: Fri Jan 29 16:42:24 2010 +0100
s4:NBT-WINSREPLICATION: fix compiler warnings
metze
commit 07b06e51bbf95168899081ba2aa92a4e6a5bac12
Author: Stefan Metzmacher me...@samba.org
Date: Fri Jan 29 16:33:58 2010 +0100
s4:NBT-WINSREPLICATION: use an array of nbt_names to loop over different
names
metze
commit ea7ec4fb30a261cedb882c9325815679e5e20d57
Author: Stefan Metzmacher me...@samba.org
Date: Mon Feb 1 14:39:13 2010 +0100
s4:winsserver: reject name registrations with a scope length 237
This matches Windows 2008 behavior. Name releases are just ignored.
metze
commit 7a02a2ba80c2997947a2b2c48c33f8c7647f7471
Author: Stefan Metzmacher me...@samba.org
Date: Sun Jan 31 18:59:41 2010 +0100
s4:wrepl_server: truncate the scope of a netbios name to 237 bytes as
Windows 2008 does
metze
commit e37dc56e971f44c00791529a39144d1a972e3ca2
Author: Stefan Metzmacher me...@samba.org
Date: Mon Feb 1 15:18:15 2010 +0100
libcli/nbt: fix ndr_push_nbt_string() string labels with a length of 63
(0x3F) are allowed
metze
---
Summary of changes:
libcli/nbt/nbtname.c |4 +-
source4/nbt_server/wins/winsserver.c |9 +++
source4/torture/nbt/wins.c | 94 ++-
source4/torture/nbt/winsreplication.c | 93 ++--
source4/wrepl_server/wrepl_apply_records.c | 14
5 files changed, 175 insertions(+), 39 deletions(-)
Changeset truncated at 500 lines:
diff --git a/libcli/nbt/nbtname.c b/libcli/nbt/nbtname.c
index 792b340..136379a 100644
--- a/libcli/nbt/nbtname.c
+++ b/libcli/nbt/nbtname.c
@@ -173,9 +173,9 @@ _PUBLIC_ enum ndr_err_code ndr_push_nbt_string(struct
ndr_push *ndr, int ndr_fla
complen = strcspn(s, .);
/* we need to make sure the length fits into 6 bytes */
- if (complen = 0x3F) {
+ if (complen 0x3F) {
return ndr_push_error(ndr, NDR_ERR_STRING,
- component length %u[%08X]
0x3F,
+ component length %u[%08X]
0x003F,
(unsigned)complen,
(unsigned)complen);
}
diff --git a/source4/nbt_server/wins/winsserver.c
b/source4/nbt_server/wins/winsserver.c
index 4b8b8cc..ca8daed 100644
--- a/source4/nbt_server/wins/winsserver.c
+++ b/source4/nbt_server/wins/winsserver.c
@@ -496,6 +496,11 @@ static void nbtd_winsserver_register(struct
nbt_name_socket *nbtsock,
goto done;
}
+ if (name-scope strlen(name-scope) 237) {
+ rcode = NBT_RCODE_SVR;
+ goto done;
+ }
+
duplicate_packet = wins_check_wack_queue(iface, packet, src);
if (duplicate_packet) {
/* just ignore the packet */
@@ -874,6 +879,10 @@ static void nbtd_winsserver_release(struct nbt_name_socket
*nbtsock,
goto done;
}
+ if (name-scope strlen(name-scope) 237) {
+ goto done;
+ }
+
status = winsdb_lookup(winssrv-wins_db, name, packet, rec);
svn commit: samba-web r1362 - in trunk/support: .
Author: jra Date: 2010-02-01 11:57:40 -0700 (Mon, 01 Feb 2010) New Revision: 1362 WebSVN: http://websvn.samba.org/cgi-bin/viewcvs.cgi?view=revroot=samba-webrev=1362 Log: Update Dentar phone number. Modified: trunk/support/us.html Changeset: Modified: trunk/support/us.html === --- trunk/support/us.html 2010-01-26 15:10:26 UTC (rev 1361) +++ trunk/support/us.html 2010-02-01 18:57:40 UTC (rev 1362) @@ -934,8 +934,7 @@ Dentar, Inc. Thomas Glascock, President UNIX, Linux, and Samba expertise. -513.985.0964 -513.312.6673 +513-312-6673 den...@dentar.com /small/pre
[SCM] CTDB repository - branch 1.0.112 updated - ctdb-1.0.111-18-g04e40de
The branch, 1.0.112 has been updated
via 04e40deac8d0c7edf907135ae81ac961c23135c3 (commit)
via 72d6ae64ab5ece7645b02054d617b71e231d4741 (commit)
from 3e2b1839a9f8419eeeb7f22ea5925f6c42f32a65 (commit)
http://gitweb.samba.org/?p=sahlberg/ctdb.git;a=shortlog;h=1.0.112
- Log -
commit 04e40deac8d0c7edf907135ae81ac961c23135c3
Author: Ronnie Sahlberg ronniesahlb...@gmail.com
Date: Tue Feb 2 08:03:37 2010 +1100
Version 1.0.112-5
commit 72d6ae64ab5ece7645b02054d617b71e231d4741
Author: Volker Lendecke v...@samba.org
Date: Fri Jan 29 18:21:09 2010 +0100
tdb: fix an early release of the global lock that can cause data corruption
There was a bug in tdb where the
tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
(ending the transaction-mutex) was done before the
/* remove the recovery marker */
This means that when a transaction is committed there is a window where
another
opener of the file sees the transaction marker while the transaction
committer
is still fully functional and working on it. This led to transaction being
rolled back by that second opener of the file while transaction_commit()
gave
no error to the caller.
This patch moves the F_UNLCK to after the recovery marker was removed,
closing
this window.
---
Summary of changes:
lib/tdb/common/transaction.c | 15 ++-
packaging/RPM/ctdb.spec.in |4 +++-
2 files changed, 13 insertions(+), 6 deletions(-)
Changeset truncated at 500 lines:
diff --git a/lib/tdb/common/transaction.c b/lib/tdb/common/transaction.c
index 20f2bfc..b8988ea 100644
--- a/lib/tdb/common/transaction.c
+++ b/lib/tdb/common/transaction.c
@@ -135,6 +135,9 @@ struct tdb_transaction {
bool prepared;
tdb_off_t magic_offset;
+ /* set when the GLOBAL_LOCK has been taken */
+ bool global_lock_taken;
+
/* old file size before transaction */
tdb_len_t old_map_size;
@@ -603,6 +606,11 @@ int _tdb_transaction_cancel(struct tdb_context *tdb)
}
}
+ if (tdb-transaction-global_lock_taken) {
+ tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
+ tdb-transaction-global_lock_taken = false;
+ }
+
/* remove any global lock created during the transaction */
if (tdb-global_lock.count != 0) {
tdb_brlock(tdb, FREELIST_TOP, F_UNLCK, F_SETLKW, 0,
4*tdb-header.hash_size);
@@ -947,11 +955,12 @@ static int _tdb_transaction_prepare_commit(struct
tdb_context *tdb)
return -1;
}
+ tdb-transaction-global_lock_taken = true;
+
if (!(tdb-flags TDB_NOSYNC)) {
/* write the recovery data to the end of the file */
if (transaction_setup_recovery(tdb,
tdb-transaction-magic_offset) == -1) {
TDB_LOG((tdb, TDB_DEBUG_FATAL,
tdb_transaction_prepare_commit: failed to setup recovery data\n));
- tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
_tdb_transaction_cancel(tdb);
return -1;
}
@@ -966,7 +975,6 @@ static int _tdb_transaction_prepare_commit(struct
tdb_context *tdb)
tdb-transaction-old_map_size) ==
-1) {
tdb-ecode = TDB_ERR_IO;
TDB_LOG((tdb, TDB_DEBUG_FATAL,
tdb_transaction_prepare_commit: expansion failed\n));
- tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
_tdb_transaction_cancel(tdb);
return -1;
}
@@ -1056,7 +1064,6 @@ int tdb_transaction_commit(struct tdb_context *tdb)
tdb_transaction_recover(tdb);
_tdb_transaction_cancel(tdb);
- tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
TDB_LOG((tdb, TDB_DEBUG_FATAL, tdb_transaction_commit:
write failed\n));
return -1;
@@ -1072,8 +1079,6 @@ int tdb_transaction_commit(struct tdb_context *tdb)
return -1;
}
- tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
-
/*
TODO: maybe write to some dummy hdr field, or write to magic
offset without mmap, before the last sync, instead of the
diff --git a/packaging/RPM/ctdb.spec.in b/packaging/RPM/ctdb.spec.in
index 93f5e22..9b74d49 100644
--- a/packaging/RPM/ctdb.spec.in
+++ b/packaging/RPM/ctdb.spec.in
@@ -5,7 +5,7 @@ Vendor: Samba Team
Packager: Samba Team sa...@samba.org
Name: ctdb
Version: 1.0.112
-Release: 4
+Release: 5
Epoch: 0
License: GNU GPL version 3
Group: System Environment/Daemons
@@ -123,6 +123,8 @@ rm -rf
[SCM] CTDB repository - branch master updated - ctdb-1.0.113-66-g898b5ed
The branch, master has been updated
via 898b5edfe757cb145960b8f3631029bfd5592119 (commit)
from 522fbb012524fe41a67dbe43589a282dda6bcbe2 (commit)
http://gitweb.samba.org/?p=sahlberg/ctdb.git;a=shortlog;h=master
- Log -
commit 898b5edfe757cb145960b8f3631029bfd5592119
Author: Volker Lendecke v...@samba.org
Date: Fri Jan 29 18:21:09 2010 +0100
tdb: fix an early release of the global lock that can cause data corruption
There was a bug in tdb where the
tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
(ending the transaction-mutex) was done before the
/* remove the recovery marker */
This means that when a transaction is committed there is a window where
another
opener of the file sees the transaction marker while the transaction
committer
is still fully functional and working on it. This led to transaction being
rolled back by that second opener of the file while transaction_commit()
gave
no error to the caller.
This patch moves the F_UNLCK to after the recovery marker was removed,
closing
this window.
---
Summary of changes:
lib/tdb/common/transaction.c | 15 ++-
1 files changed, 10 insertions(+), 5 deletions(-)
Changeset truncated at 500 lines:
diff --git a/lib/tdb/common/transaction.c b/lib/tdb/common/transaction.c
index 20f2bfc..b8988ea 100644
--- a/lib/tdb/common/transaction.c
+++ b/lib/tdb/common/transaction.c
@@ -135,6 +135,9 @@ struct tdb_transaction {
bool prepared;
tdb_off_t magic_offset;
+ /* set when the GLOBAL_LOCK has been taken */
+ bool global_lock_taken;
+
/* old file size before transaction */
tdb_len_t old_map_size;
@@ -603,6 +606,11 @@ int _tdb_transaction_cancel(struct tdb_context *tdb)
}
}
+ if (tdb-transaction-global_lock_taken) {
+ tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
+ tdb-transaction-global_lock_taken = false;
+ }
+
/* remove any global lock created during the transaction */
if (tdb-global_lock.count != 0) {
tdb_brlock(tdb, FREELIST_TOP, F_UNLCK, F_SETLKW, 0,
4*tdb-header.hash_size);
@@ -947,11 +955,12 @@ static int _tdb_transaction_prepare_commit(struct
tdb_context *tdb)
return -1;
}
+ tdb-transaction-global_lock_taken = true;
+
if (!(tdb-flags TDB_NOSYNC)) {
/* write the recovery data to the end of the file */
if (transaction_setup_recovery(tdb,
tdb-transaction-magic_offset) == -1) {
TDB_LOG((tdb, TDB_DEBUG_FATAL,
tdb_transaction_prepare_commit: failed to setup recovery data\n));
- tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
_tdb_transaction_cancel(tdb);
return -1;
}
@@ -966,7 +975,6 @@ static int _tdb_transaction_prepare_commit(struct
tdb_context *tdb)
tdb-transaction-old_map_size) ==
-1) {
tdb-ecode = TDB_ERR_IO;
TDB_LOG((tdb, TDB_DEBUG_FATAL,
tdb_transaction_prepare_commit: expansion failed\n));
- tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
_tdb_transaction_cancel(tdb);
return -1;
}
@@ -1056,7 +1064,6 @@ int tdb_transaction_commit(struct tdb_context *tdb)
tdb_transaction_recover(tdb);
_tdb_transaction_cancel(tdb);
- tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
TDB_LOG((tdb, TDB_DEBUG_FATAL, tdb_transaction_commit:
write failed\n));
return -1;
@@ -1072,8 +1079,6 @@ int tdb_transaction_commit(struct tdb_context *tdb)
return -1;
}
- tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
-
/*
TODO: maybe write to some dummy hdr field, or write to magic
offset without mmap, before the last sync, instead of the
--
CTDB repository
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated
via b300981... Fix bug #7080 - Quota only shown when logged as root.
from 431d75f... s4:NBT-WINS: test large scopes
http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master
- Log -
commit b3009819c55043aca81d5b53c805696dd4c88aac
Author: Jeremy Allison j...@samba.org
Date: Mon Feb 1 15:55:55 2010 -0800
Fix bug #7080 - Quota only shown when logged as root.
conn-server_info-utok.uid == 0
isn't the correct check to see if we're root anymore. As rpc_samr_nt.c does,
the correct check is :
geteuid() == sec_initial_uid()
Jeremy.
---
Summary of changes:
source3/smbd/fake_file.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
Changeset truncated at 500 lines:
diff --git a/source3/smbd/fake_file.c b/source3/smbd/fake_file.c
index 6898793..9917481 100644
--- a/source3/smbd/fake_file.c
+++ b/source3/smbd/fake_file.c
@@ -126,7 +126,7 @@ NTSTATUS open_fake_file(struct smb_request *req,
connection_struct *conn,
NTSTATUS status;
/* access check */
- if (conn-server_info-utok.uid != 0) {
+ if (geteuid() == sec_initial_uid()) {
DEBUG(3, (open_fake_file_shared: access_denied to
service[%s] file[%s] user[%s]\n,
lp_servicename(SNUM(conn)),
--
Samba Shared Repository
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated
via a35be8c... Torture test for bug 7084 - Create time on directories
not stored properly in an EA in new create time code.
from 4e08d0d... Arg. Got the sense of the test reversed to fix bug #7080
- Quota only shown when logged as root.. Doh !
http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master
- Log -
commit a35be8c0c2d804638d9eda5afb6c96ef2f090549
Author: Jeremy Allison j...@samba.org
Date: Mon Feb 1 18:49:50 2010 -0800
Torture test for bug 7084 - Create time on directories not stored properly
in an EA in new create time code.
Ensure new files in a directory don't reset the create time.
Jeremy.
---
Summary of changes:
source3/script/tests/test_smbtorture_s3.sh |2 +-
source3/torture/torture.c | 87
2 files changed, 88 insertions(+), 1 deletions(-)
Changeset truncated at 500 lines:
diff --git a/source3/script/tests/test_smbtorture_s3.sh
b/source3/script/tests/test_smbtorture_s3.sh
index 1cbfc23..3e0e3de 100755
--- a/source3/script/tests/test_smbtorture_s3.sh
+++ b/source3/script/tests/test_smbtorture_s3.sh
@@ -28,7 +28,7 @@ tests=FDPASS LOCK1 LOCK2 LOCK3 LOCK4 LOCK5 LOCK6 LOCK7 LOCK9
#tests=$tests UNLINK BROWSE ATTR TRANS2 MAXFID TORTURE
tests=$tests UNLINK BROWSE ATTR TRANS2 TORTURE
tests=$tests OPLOCK1 OPLOCK2 OPLOCK3 STREAMERROR
-tests=$tests DIR DIR1 TCON TCONDEV RW1 RW2 RW3
+tests=$tests DIR DIR1 DIR-CREATETIME TCON TCONDEV RW1 RW2 RW3
tests=$tests OPEN XCOPY RENAME DELETE PROPERTIES W2K
tests=$tests TCON2 IOCTL CHKPATH FDSESS LOCAL-SUBSTITUTE CHAIN1
tests=$tests GETADDRINFO POSIX UID-REGRESSION-TEST SHORTNAME-TEST
diff --git a/source3/torture/torture.c b/source3/torture/torture.c
index 758bb60..07cf5f9 100644
--- a/source3/torture/torture.c
+++ b/source3/torture/torture.c
@@ -6340,6 +6340,92 @@ static bool run_tldap(int dummy)
return true;
}
+/* Torture test to ensure no regression of :
+https://bugzilla.samba.org/show_bug.cgi?id=7084
+*/
+
+static bool run_dir_createtime(int dummy)
+{
+ struct cli_state *cli;
+ const char *dname = \\testdir;
+ const char *fname = \\testdir\\testfile;
+ NTSTATUS status;
+ struct timespec create_time;
+ struct timespec create_time1;
+ uint16_t fnum;
+ bool ret = false;
+
+ if (!torture_open_connection(cli, 0)) {
+ return false;
+ }
+
+ cli_unlink(cli, fname, aSYSTEM | aHIDDEN);
+ cli_rmdir(cli, dname);
+
+ status = cli_mkdir(cli, dname);
+ if (!NT_STATUS_IS_OK(status)) {
+ printf(mkdir failed: %s\n, nt_errstr(status));
+ goto out;
+ }
+
+ if (!cli_qpathinfo2(cli,
+ dname,
+ create_time,
+ NULL,
+ NULL,
+ NULL,
+ NULL,
+ NULL,
+ NULL)) {
+ status = cli_nt_error(cli);
+ printf(cli_qpathinfo2 returned %s\n,
+ nt_errstr(status));
+ goto out;
+ }
+
+ /* Sleep 3 seconds, then create a file. */
+ sleep(3);
+
+ status = cli_open(cli, fname, O_RDWR | O_CREAT | O_EXCL,
+ DENY_NONE, fnum);
+ if (!NT_STATUS_IS_OK(status)) {
+ printf(cli_open failed: %s\n, nt_errstr(status));
+ goto out;
+ }
+
+ if (!cli_qpathinfo2(cli,
+ dname,
+ create_time1,
+ NULL,
+ NULL,
+ NULL,
+ NULL,
+ NULL,
+ NULL)) {
+ status = cli_nt_error(cli);
+ printf(cli_qpathinfo2 (2) returned %s\n,
+ nt_errstr(status));
+ goto out;
+ }
+
+ if (timespec_compare(create_time1, create_time)) {
+ printf(run_dir_createtime: create time was updated (error)\n);
+ } else {
+ printf(run_dir_createtime: create time was not updated
(correct)\n);
+ ret = true;
+ }
+
+ out:
+
+ cli_unlink(cli, fname, aSYSTEM | aHIDDEN);
+ cli_rmdir(cli, dname);
+ if (!torture_close_connection(cli)) {
+ ret = false;
+ }
+ return ret;
+}
+
+
static bool run_streamerror(int dummy)
{
struct cli_state *cli;
@@ -7195,6 +7281,7 @@ static struct {
{OPLOCK3, run_oplock3, 0},
{DIR, run_dirtest, 0},
{DIR1, run_dirtest1, 0},
+ {DIR-CREATETIME, run_dir_createtime, 0},
{DENY1, torture_denytest1, 0},
{DENY2, torture_denytest2, 0},
{TCON, run_tcon_test, 0},
--
Samba Shared Repository
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated
via f3ee151... Fix bug #7084 - Create time on directories not stored
properly in an EA in new create time code.
from a35be8c... Torture test for bug 7084 - Create time on directories
not stored properly in an EA in new create time code.
http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master
- Log -
commit f3ee1516e631274531439e0ab7c384d02aff789c
Author: Jeremy Allison j...@samba.org
Date: Mon Feb 1 18:50:43 2010 -0800
Fix bug #7084 - Create time on directories not stored properly in an EA in
new create time code.
Remove erroneous optimisation that caused no EA to be set
if calculated btime matched st_ex btime, and calculated DOS
attribute matched existing file attribute.
Jeremy.
---
Summary of changes:
source3/smbd/dosmode.c |7 ---
1 files changed, 0 insertions(+), 7 deletions(-)
Changeset truncated at 500 lines:
diff --git a/source3/smbd/dosmode.c b/source3/smbd/dosmode.c
index e5d6bc3..10249f4 100644
--- a/source3/smbd/dosmode.c
+++ b/source3/smbd/dosmode.c
@@ -739,13 +739,6 @@ int file_set_dosmode(connection_struct *conn, struct
smb_filename *smb_fname,
dosmode = ~FILE_ATTRIBUTE_OFFLINE;
old_mode = ~FILE_ATTRIBUTE_OFFLINE;
- if (old_mode == dosmode
- (timespec_compare(new_create_timespec,
- smb_fname-st.st_ex_btime) == 0)) {
- smb_fname-st.st_ex_mode = unixmode;
- return(0);
- }
-
smb_fname-st.st_ex_btime = new_create_timespec;
#ifdef HAVE_STAT_DOS_FLAGS
--
Samba Shared Repository
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated via 86bf5ea... Add debug to make it clear when EA dosmode set is invoked. from f3ee151... Fix bug #7084 - Create time on directories not stored properly in an EA in new create time code. http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit 86bf5eac678d54037d8fcf81bade1f464dd97299 Author: Jeremy Allison j...@samba.org Date: Mon Feb 1 19:21:35 2010 -0800 Add debug to make it clear when EA dosmode set is invoked. Jeremy. --- Summary of changes: source3/smbd/dosmode.c |5 + 1 files changed, 5 insertions(+), 0 deletions(-) Changeset truncated at 500 lines: diff --git a/source3/smbd/dosmode.c b/source3/smbd/dosmode.c index 10249f4..4365563 100644 --- a/source3/smbd/dosmode.c +++ b/source3/smbd/dosmode.c @@ -372,6 +372,11 @@ static bool set_ea_dos_attribute(connection_struct *conn, unix_timespec_to_nt_time(dosattrib.info.info3.create_time, smb_fname-st.st_ex_btime); + DEBUG(10,(set_ea_dos_attributes: set attribute 0x%x, btime = %s on file %s\n, + (unsigned int)dosmode, + time_to_asc(convert_timespec_to_time_t(smb_fname-st.st_ex_btime)), + smb_fname_str_dbg(smb_fname) )); + ndr_err = ndr_push_struct_blob( blob, talloc_tos(), NULL, dosattrib, (ndr_push_flags_fn_t)ndr_push_xattr_DOSATTRIB); -- Samba Shared Repository
Build status as of Tue Feb 2 07:00:03 2010
URL: http://build.samba.org/ --- /home/build/master/cache/broken_results.txt.old 2010-02-01 00:00:16.0 -0700 +++ /home/build/master/cache/broken_results.txt 2010-02-02 00:00:30.0 -0700 @@ -1,10 +1,10 @@ -Build status as of Mon Feb 1 07:00:06 2010 +Build status as of Tue Feb 2 07:00:03 2010 Build counts: Tree Total Broken Panic build_farm 0 0 0 -ccache 3 0 0 -ldb 28 28 0 +ccache 1 0 0 +ldb 20 20 0 libreplace 1 0 0 lorikeet 0 0 0 pidl 0 0 0 @@ -13,9 +13,9 @@ samba-docs 0 0 0 samba-web0 0 0 samba_3_current 0 0 0 -samba_3_master 25 24 3 -samba_3_next 17 16 1 -samba_4_0_test 28 28 1 +samba_3_master 17 16 1 +samba_3_next 0 0 0 +samba_4_0_test 20 20 1 talloc 1 0 0 -tdb 1 1 0 +tdb 18 8 0
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated via d8071e7... Change uint_t to unsigned int in source4 via a47b6eb... Change uint_t to unsigned int in source3 via 3ee9d1a... Change uint_t to unsigned int in pidl via aa6a507... Change uint_t to unsigned int in libcli via a7250eb... Change uint_t to unsigned int in lib/util via 2cb2d53... Change uint_t to unsigned int in lib/talloc via d4fc2e8... Change uint_t to unsigned int in lib/crypto via a2c95a5... Change uint_t to unsigned int in lib/replace via e3aeea2... s4:NBT-WINSREPLICATION: don't mark a local variable as static from 86bf5ea... Add debug to make it clear when EA dosmode set is invoked. http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit d8071e7ed7c57e8b0f2d80420e27f9ea7ab63022 Author: Matt Kraai mkr...@beckman.com Date: Tue Jan 5 09:42:54 2010 -0800 Change uint_t to unsigned int in source4 Signed-off-by: Stefan Metzmacher me...@samba.org commit a47b6ebb3036f5073fdb2320b584345d3604c38d Author: Matt Kraai mkr...@beckman.com Date: Tue Jan 5 09:42:23 2010 -0800 Change uint_t to unsigned int in source3 Signed-off-by: Stefan Metzmacher me...@samba.org commit 3ee9d1a3d1e934eccd0846b4dac30aae40dc70f8 Author: Matt Kraai mkr...@beckman.com Date: Tue Jan 5 09:41:53 2010 -0800 Change uint_t to unsigned int in pidl Signed-off-by: Stefan Metzmacher me...@samba.org commit aa6a507e76685f2a8d1f3bd33cbf0fbdeb0d69ce Author: Matt Kraai mkr...@beckman.com Date: Tue Jan 5 09:41:24 2010 -0800 Change uint_t to unsigned int in libcli Signed-off-by: Stefan Metzmacher me...@samba.org commit a7250eb14270101d4ca580acd8fdcaf7e308e515 Author: Matt Kraai mkr...@beckman.com Date: Tue Jan 5 09:40:54 2010 -0800 Change uint_t to unsigned int in lib/util Signed-off-by: Stefan Metzmacher me...@samba.org commit 2cb2d5317f8428b5eaef2365c815aa34739a80b5 Author: Matt Kraai mkr...@beckman.com Date: Tue Jan 5 09:40:26 2010 -0800 Change uint_t to unsigned int in lib/talloc Signed-off-by: Stefan Metzmacher me...@samba.org commit d4fc2e82012277e5e8570da5d08d62c6847a99c0 Author: Matt Kraai mkr...@beckman.com Date: Tue Jan 5 09:39:56 2010 -0800 Change uint_t to unsigned int in lib/crypto Signed-off-by: Stefan Metzmacher me...@samba.org commit a2c95a5d961594f2accbb032cd7e8da9767e2e47 Author: Matt Kraai mkr...@beckman.com Date: Tue Jan 5 09:43:24 2010 -0800 Change uint_t to unsigned int in lib/replace Signed-off-by: Stefan Metzmacher me...@samba.org commit e3aeea29a4e730ccbe7367645703cdc9ee8e7a7e Author: Stefan Metzmacher me...@samba.org Date: Mon Feb 1 17:30:30 2010 +0100 s4:NBT-WINSREPLICATION: don't mark a local variable as static This was somehow introduced in commit 8773e743c518578584d07d35ffdafdd598af88b0. metze --- Summary of changes: lib/crypto/md5.c |8 ++-- lib/replace/README |1 - lib/replace/getifaddrs.c |2 +- lib/replace/libreplace_cc.m4 |1 - lib/replace/test/testsuite.c |1 - lib/talloc/talloc.3.xml|6 ++-- lib/talloc/talloc_guide.txt|6 ++-- lib/util/asn1.c|4 +- lib/util/byteorder.h |4 +- lib/util/charset/iconv.c |2 +- lib/util/util_str.c|2 +- libcli/auth/msrpc_parse.c |2 +- libcli/auth/proto.h|2 +- libcli/auth/smbdes.c |2 +- libcli/nbt/nbtname.c |2 +- libcli/security/dom_sid.c |2 +- pidl/tests/Util.pm |1 - source3/include/proto.h|8 ++-- source3/libsmb/clirap2.c |8 ++-- source4/client/client.c|4 +- source4/dsdb/common/util.c | 16 source4/dsdb/samdb/ldb_modules/password_hash.c |4 +- source4/lib/messaging/messaging.c |2 +- source4/libcli/clifile.c |6 ++-- source4/libcli/raw/interfaces.h| 20 +- source4/libcli/raw/libcliraw.h | 34 +- source4/libcli/raw/raweas.c| 44 source4/libcli/raw/rawfile.c |4 +- source4/libcli/raw/rawfileinfo.c |2 +- source4/libcli/raw/rawrequest.c| 38 ++-- source4/libcli/raw/rawsearch.c | 12 +++---
