Re: [Samba] Problems with ACL jumbo patch
Okay, I search the apt repositories and found three Kerberos libraries: libpam-krb5, krb5-auth-dialog and libkrb5-dev. 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. I looked through the options for configure.developer, can I disable Kerberos in the configure step and bypass the problem? -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] vfs_acl_xattr issue
Le 08/10/2010 17:58, zorgman a écrit : Hello, I'm using samba 3.4.8~dfsg-2 with vfs objects = acl_xattr. [share] path = /home/users/ valid users = read only = No create mask = 0666 directory mask = 0770 vfs objects = acl_xattr Here is my problem, i'm connecting to my share, which have a folder 'f01' with some acl. I created a folder and broke inherit 'f02'. Now if i create a folder in 'f02' without change it (so it's called New folder, the checkbox inherit acl is checked). If i rename New folder as something the checkbox inherit acl is becoming unchecked. F01 = tutu,toto,titi (inherited checked) \_F02 = tutu,toto (inherited unchecked) \_New folder = tutu,toto (inherited checked) \_something = tutu,toto (inherited unchecked) Can you help me? thanks I tried different thing, first, update samba to 3.5.6. All seems work but unless I check de Replace permission entries on all child objects with entries shown here that apply to child objects. Include these with entries explicitly defined here, after that, if i create a child folder, the inherited checkbox is unchecked unless i don't rename it. The Replace permission entries on all child objects with entries shown here that apply to child objects. Include these with entries explicitly defined here seems also have some problems because if I try it on somes folders, i have Windows erros like : An error occured when applied security information to \Device\Lanman\Redirector\nb_servername\shere. Is there a link between the probs? Please help me! -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] samba on SLES 10: binding problem
Greets, samba-users, I try to find a solution for the following situation: We run samba on a SLES10 SP3 server (32bit). All the SLES-updates are in ... The server started to show problems with CUPS, cupsd just stopped working out of the blue and had to be restarted. No specific errors were logged. I wrote a small script to watchdog that daemon and restart it (people want to print ...) while also tar-ing the logs aside. Now we also had samba stopping 3 or 4 times. In the beginning it was an old release (3.0.37 or so), I gradually upgraded, my last step yesterday was from 3.4.9 to 3.5.6 (using the rpms from sernet.de, thanks), after the last stopped samba on friday. The problem was that smbd seemed not to be able to bind to the network interface. I issued /etc/init.d/smb restart and netstat did not show an smbd bound to the IP of eth0 after this. Only a rcnetwork restart helped sometimes (or a reboot). We are trying to reproduce the problem, but aren't successful so far. Maybe 3.5.6 solves it? Additional info: the server is the PDC for the NT4-domain there, and we also think of the windows-7-clients to maybe do something nasty ... ? Ah, and one specific thing is that the server is also hosting VMware Server 2 (maybe the networking-part of that beast is cross-firing). I'd be thankful for any thoughts or advice ... thanks! Stefan -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] [ob...@samba.org: 3.6:idmap:Q2: get rid of (all/most) idmap alloc parameters for idmap_ldap ?]
Hi Samba Users, I forgot to send this mail to the samba ML, too. The question is whether anyone is using the idmap alloc config options for idmap_ldap. I would like to remove them for Samba 3.6. Details below. Thanks - Michael - Forwarded message from Michael Adam ob...@samba.org - Date: Wed, 10 Nov 2010 11:19:56 +0100 From: Michael Adam ob...@samba.org To: samba-techni...@lists.samba.org Subject: 3.6:idmap:Q2: get rid of (all/most) idmap alloc parameters for idmap_ldap ? Hi, a second question about config-visible idmap changes I propose for samba 3.6: In my idmap rewrite, I kept the alloc related parameters for the LDAP idmap backend for now: - idmap alloc config : ldap_url - idmap alloc config : ldap_base_dn - idmap alloc config : ldap_user_dn and the related idmap alloc secret. I would like to get rid of these. Therefore, I am asking here, if there is anyone out there using these? I can not imagine a reason why one would want to use different server and/or user+password for storing the uid/gid counter. The only option that I would attest a certain, though minimal, right to exist is the ldap_base_dn. But usually, it should imho ok to store the uid/gid counter in the same location as the mappings. So, again: Are these options needed/used at all? Or can I remove them for 3.6.0 ? Cheers - Michael Note: If we need to keep any of the options, the current form (idmap alloc config : option = ...) would reference the default config, but my idmap rewrite would enable us to set these on a per-domain basis, which would call for options like this idmap config DOMAIN : alloc_option) - End forwarded message - pgpSmg0piPeLl.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] vfs_acl_xattr issue
I am also under the impression that inheritance still does not work properly, even after applying the acl jumbo patch to samba 3.5.6. I don't want to be alarmist, though, so I am still testing. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] Samba try to Map and receive message Not enough server storage is availble to process this command
Jeremy Allison wrote: On Mon, Nov 08, 2010 at 12:20:28PM -0500, Thomas Horton wrote: Hello, Samba was working fine as of 4pm last Friday for us. Now we are getting the space message. Does anyone have any idea what would randomly cause this. We looked at log files and they don't appear large and consuming all our drive space. Nothing should randomly cause this. Turn up the debug log level and see if you can spot where the DISK FULL messages are being returned to the client. Running out of i-nodes rather than disk space? Use df -i to check. 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
[Samba] [ob...@samba.org: 3.6:idmap:Q1: add idmap range ?]
Hi Samba Users, I forgot to send this mail to the samba ML, too. The question is whether there are serious objections to replacing idmap uid and idmap gid by a new option idmap range. Details below. Thanks - Michael - Forwarded message from Michael Adam ob...@samba.org - Date: Wed, 10 Nov 2010 09:58:26 +0100 From: Michael Adam ob...@samba.org To: samba-techni...@lists.samba.org Subject: 3.6:idmap:Q1: add idmap range ? Hi List, I have recently rewritten the idmap code in winbindd to some extent. The code has been unified under the hood. And one thing I would really like to do would be to unify the two configuration options idmap uid and idmap gid into one option called idmap range for the sake of consistency of the default idmap config with the explicit domain configurations (idmap config DOMAIN : range = ...). Currently, the uid and gid range are expected to be identical anyways. The RID module errors out if they aren't and otherwise the intersection of both ranges is built. So de facto we do have one single range already. After discussing this with Volker and Jeremy, the procedure I have in mind is this: For 3.6, I would like to introduce idmap range in addition and mark idmap uid and idmap gid as deprecated. At the place where now the intersection of idmap uid and idmap gid is built, I would build the intersection of all three ranges (possibly with a warning). This way old and new configs would work without problems. In 3.7 or any other future major new version, I would then drop idmap uid and idmap gid. Does that sound ok? Or are there any major concerns with this? I would really love to reach a stage where we have a sane id mapping system and configuration. :-) Cheers - Michael - End forwarded message - pgphB5Puekyyj.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] Fwd: Windows 7 problems
-Message d'origine- De : samba-boun...@lists.samba.org [mailto:samba-boun...@lists.samba.org] De la part de Cain, Marc Envoyé : mardi 9 novembre 2010 21:34 À : samba@lists.samba.org Objet : [Samba] Fwd: Windows 7 problems On Nov 9, 2010, at 7:29 AM, henri wrote: I have a couple of questions related to Windows 7 : - opening a domain session on the Win7 client takes a long time. At least 40 seconds, while there is no delay on XP client. Is it a kown problem ? A bug ? What can be done to avoid such a delay ? There are a least two timeouts that I've found that can affect domain logons -- Roaming profile logon timeout: When a user with a roaming profile attempts to logon to a Samba domain Windows will display the Welcome screen for 30 seconds before enabling the user's desktop. This bug does not affect Active Directory logons. Setting the GPO below to 0 seconds will work around this timeout. \\Computer Configuration\Administrative Templates\System\Set maximum wait time for the network if a user has a roaming user profile or remote home directory Set to: 0 -- Synchronous Logon script timeout: When the local Computer GPO is set to Run logon scripts synchronously Windows 7 displays the Welcome screen for 30 seconds before enabling the user's desktop. NOTE: this bug may only be extant in Windows 7 Professional, not Enterprise or may have been addressed in service pack. Creating the following REG_DWORD registry key will work around this bug: \\HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion \Policies\System\DelayedDesktopSwitchTimeout Set to: 0001 There is also the solid color background bug (again, it may have since been addressed in a service pack): Change the background color to any .jpg image. Thanks for that. After some tests I got it work. - in smb.conf , the %a value for Windows 7 is Vista , right ? Don't know. The %a switch hasn't worked in our shop for some time. Windows 7 will tag the profile folder listed with a .V2 extension or will expect a .V2 extension. I was asking for the %a substitution in order to call the right script file at logon . For me , logon%a.cmd is substitued by logonVista.cmd . I assume it's ok. - NT Default Profile no longer works with Windows 7 . I guess it is normal , but is there a workaround ? Default User profiles do work with Windows 7 and Samba. See the article for Windows 7 default profile details: http://support.microsoft.com/kb/973289 Thanks for the link. I will try the procedure. - Same question for NT policies : I assume it is normal that windows 7 does not support NTconfig.pol files any longer. Nevertheless, is there a way to get the same features ? For these kind of functionality (having some control of user profiles on Win7 clients) , is AD the only and inescapable solution ? We create a local group policy that we copy from the server via logon script at logon time. Works great for bot WinXP and Windows 7. How do you generate the policy file and how do you copy it from server to client via logon script ? XP was automaticaly loading ntconfig.pol (generated by the old poledit.exe) in the netlogon directory , but what about Win 7 ? Thanks for your help. Henri For my needs (200 domain users), NT domain was a sufficient solution and worked perfectly with XP. I am affraid that with Windows 7, I will have to move to something much more and unnecessarily complex :-( . Thanks in advance for your help. Henri -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] switching winbind's ad without reboot?
The IT department is phasing in new Active Directory servers, during work ours, and I need to get all my systems using winbind for authentication to switch to the new servers, and hopefully without a reboot. I tried to redirect a machine currently authenticating via the current AD to the new AD, but even after restarting all the winbind, smbd, nmbd processes, I see it still pointing to the old AD. After rebooting, it points to the new server. Trying to fool samba by adding a line to /etc/hosts didn't work either. Is there a slick non-reboot method? -BEFORE CHANGE krb5.conf contains: [realms] MY.DOMAIN.COM = { kdc = 192.168.0.1 admin_server = 192.168.0.1 default_domain = my.domain.com } smb.conf contains: security = ADS password server = ad1.my.domain.com AFTER CHANGE [realms] MY.DOMAIN.COM = { kdc = 192.168.0.2 admin_server = 192.168.0.2 default_domain = my.domain.com } smb.conf contains: security = ADS password server = ad2.my.domain.com The way I know the system is pointing to the old ad server is by doing wbinfo -g followed by lsof -i tcp -nP |grep winbind -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] Windows 7 problems
On Tue, 9 Nov 2010 19:29:23 +0400 henri ciradhb.forw...@laposte.net wrote: h Same question for NT policies : I assume it is normal that windows 7 does h not support NTconfig.pol files any longer. Nevertheless, is there a way to h get the same features ? For these kind of functionality (having some control h of user profiles on Win7 clients) , is AD the only and inescapable solution h ? We use a combination of a logon script and Wpkg (wpkg.org) to pull the machine and user registry entries to our Win7 clients we add a registry value with version number of the changes and have a check of this number before updating the hives, but that's not really necessary and you could overwrite the registry every time. (if your users have admin rights on their machine a logon script will suffice.) machine changes Wpkg definition package id=vreg name=V2 Registry settings revision=10 reboot=false priority=900 check type=registry condition=equals path=HKLM\Software\Wpkg\RegistryRevision value=10 / install cmd='regedit /s %WpkgRoot%\Files\V2.reg' / upgrade cmd='regedit /s %WpkgRoot%\Files\V2.reg' / /package -V2.reg-- Windows Registry Editor Version 5.00 [HKEY_LOCAL_MACHINE\SOFTWARE\WPKG] RegistryRevision=dword:000a ;Fix for slow tab opening IE7/8 [-HKEY_LOCAL_MACHINE\SOFTWARE\Classes\CLSID\{DBC80044-A445-435b-BC74-9C25C1C588A9}] [-HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Explorer\Browser Helper Objects\{DBC80044-A445-435b-BC74-9C25C1C588A9}] [HKEY_LOCAL_MACHINE\SOFTWARE\JavaSoft\Java Update\Policy] EnableJavaUpdate=dword: [HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\policies\Explorer] NoDriveTypeAutoRun=dword:00ff [HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System] VerboseStatus=dword:0001 RunLogonScriptSync=dword:0001 DelayedDesktopSwitchTimeout=dword:0005 [HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon] RunLogonScriptSync=dword:0001 SyncForegroundPolicy=dword:0001 [HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Microsoft\Windows\System] CompatibleRUPSecurity=dword:0001 [HKEY_LOCAL_MACHINE\Software\Policies\Microsoft\Windows\NetCache] NoConfigCache=dword: DisableFRAdminPin=dword:0001 [HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Microsoft\windows\WindowsUpdate\AU] NoAutoUpdate=dword: AUOptions=dword:0004 ScheduledInstallDay=dword: ScheduledInstallTime=dword:0006 RescheduleWaitTime=dword:0005 NoAutoRebootWithLoggedOnUsers=dword:0001 [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\LSA] NoLMHash=dword:0001 RestrictAnonymous=dword:0002 [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Terminal Server] fDenyTSConnections=dword: [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\LanmanWorkstation\Parameters] DomainCompatibilityMode=dword:0001 DNSNameResolutionRequired=dword: [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip6\Parameters] DisabledComponents=dword: [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager] CWDIllegalInDllSearch=dword:0002 -- User changes ---Logon.cmd- .. @if not %allusersprofile%==%SystemDrive%\ProgramData goto Step_4 @reg query HKCU\Software\Ersatz /v UserShellFolders | find 0x2 NUL @if errorlevel 1 reg import %Logonserver%\netlogon\hkcu_user_shell_folders_v2.reg @reg query HKCU\Software\Ersatz /v SomeOtherFunction | find 0x9 NUL @if errorlevel 1 reg import %Logonserver%\netlogon\Someotherfunction_v2.reg :Step_4 .. - --hkcu_user_shell_folders_v2.reg--- Windows Registry Editor Version 5.00 [HKEY_CURRENT_USER\Software\Ersatz] UserShellFolders=dword:0002 [HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Explorer\User Shell Folders] Favorites=hex(2):25,00,48,00,4f,00,4d,00,45,00,44,00,52,00,49,00,56,00,45,00,\ 25,00,5c,00,50,00,72,00,6f,00,66,00,69,00,6c,00,65,00,5c,00,46,00,61,00,76,\ 00,6f,00,72,00,69,00,74,00,65,00,73,00,00,00 My Music=hex(2):25,00,48,00,4f,00,4d,00,45,00,44,00,52,00,49,00,56,00,45,00,\ 25,00,5c,00,4d,00,79,00,20,00,44,00,6f,00,63,00,75,00,6d,00,65,00,6e,00,74,\ 00,73,00,5c,00,4d,00,79,00,20,00,4d,00,75,00,73,00,69,00,63,00,00,00 My Pictures=hex(2):25,00,48,00,4f,00,4d,00,45,00,44,00,52,00,49,00,56,00,45,\ 00,25,00,5c,00,4d,00,79,00,20,00,44,00,6f,00,63,00,75,00,6d,00,65,00,6e,00,\ 74,00,73,00,5c,00,4d,00,79,00,20,00,50,00,69,00,63,00,74,00,75,00,72,00,65,\ 00,73,00,00,00 My Video=hex(2):25,00,48,00,4f,00,4d,00,45,00,44,00,52,00,49,00,56,00,45,00,\ 25,00,5c,00,4d,00,79,00,20,00,44,00,6f,00,63,00,75,00,6d,00,65,00,6e,00,74,\
Re: [Samba] Fwd: Windows 7 problems
On 10/11/2010 15:03, henri wrote: -Message d'origine- De : samba-boun...@lists.samba.org [mailto:samba-boun...@lists.samba.org] De la part de Cain, Marc Envoyé : mardi 9 novembre 2010 21:34 À : samba@lists.samba.org Objet : [Samba] Fwd: Windows 7 problems On Nov 9, 2010, at 7:29 AM, henri wrote: I have a couple of questions related to Windows 7 : - opening a domain session on the Win7 client takes a long time. At least 40 seconds, while there is no delay on XP client. Is it a kown problem ? A bug ? What can be done to avoid such a delay ? There are a least two timeouts that I've found that can affect domain logons -- Roaming profile logon timeout: When a user with a roaming profile attempts to logon to a Samba domain Windows will display the Welcome screen for 30 seconds before enabling the user's desktop. This bug does not affect Active Directory logons. Setting the GPO below to 0 seconds will work around this timeout. \\Computer Configuration\Administrative Templates\System\Set maximum wait time for the network if a user has a roaming user profile or remote home directory Set to: 0 -- Synchronous Logon script timeout: When the local Computer GPO is set to Run logon scripts synchronously Windows 7 displays the Welcome screen for 30 seconds before enabling the user's desktop. NOTE: this bug may only be extant in Windows 7 Professional, not Enterprise or may have been addressed in service pack. Creating the following REG_DWORD registry key will work around this bug: \\HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion \Policies\System\DelayedDesktopSwitchTimeout Set to: 0001 There is also the solid color background bug (again, it may have since been addressed in a service pack): Change the background color to any .jpg image. Thanks for that. After some tests I got it work. - in smb.conf , the %a value for Windows 7 is Vista , right ? Don't know. The %a switch hasn't worked in our shop for some time. Windows 7 will tag the profile folder listed with a .V2 extension or will expect a .V2 extension. I was asking for the %a substitution in order to call the right script file at logon . For me , logon%a.cmd is substitued by logonVista.cmd . I assume it's ok. - NT Default Profile no longer works with Windows 7 . I guess it is normal , but is there a workaround ? Default User profiles do work with Windows 7 and Samba. See the article for Windows 7 default profile details: http://support.microsoft.com/kb/973289 Thanks for the link. I will try the procedure. - Same question for NT policies : I assume it is normal that windows 7 does not support NTconfig.pol files any longer. Nevertheless, is there a way to get the same features ? For these kind of functionality (having some control of user profiles on Win7 clients) , is AD the only and inescapable solution ? We create a local group policy that we copy from the server via logon script at logon time. Works great for bot WinXP and Windows 7. How do you generate the policy file and how do you copy it from server to client via logon script ? XP was automaticaly loading ntconfig.pol (generated by the old poledit.exe) in the netlogon directory , but what about Win 7 ? Henri, Yeah, since Vista ntconfig.pol does not work. We have a tool that translates contents of ntconfig.pol (which essentially is a registry hive) into local group policy - it feteches the .pol file from the netlogon share and does the translation on the fly. I have not fully tested it on windows 7 (since we moved to samba4) but it worked fine on Windows XP. You are more than welcome to have it. Note that the script only supports machine policy. If you after proper group policy you should consider samba4 Regrds Luk Thanks for your help. Henri For my needs (200 domain users), NT domain was a sufficient solution and worked perfectly with XP. I am affraid that with Windows 7, I will have to move to something much more and unnecessarily complex :-( . Thanks in advance for your help. Henri -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] switching winbind's ad without reboot?
I found a way to change the AD for winbind without rebooting on Debian and Gentoo, however I wonder if this is a bug. First I stop smbd, nmbd, windbindd if they are running. Then I remove some files. On Debian, I removed: /var/run/samba/smb_krb5/krb5.conf.MS /var/run/samba/gencache.tdb On Gentoo, I removed: /var/cache/samba/smb_krb5/krb5.conf.MS /var/cache/samba/gencache.tdb Finally I restart winbindd, and consistently see my change using: lsof -i tcp -nP |grep winbindd Shouldn't a config change and winbindd process restart be enough? I would like to know more about what is really going on here. On Wed, 2010-11-10 at 08:43 -0800, John Stile wrote: The IT department is phasing in new Active Directory servers, during work ours, and I need to get all my systems using winbind for authentication to switch to the new servers, and hopefully without a reboot. I tried to redirect a machine currently authenticating via the current AD to the new AD, but even after restarting all the winbind, smbd, nmbd processes, I see it still pointing to the old AD. After rebooting, it points to the new server. Trying to fool samba by adding a line to /etc/hosts didn't work either. Is there a slick non-reboot method? -BEFORE CHANGE krb5.conf contains: [realms] MY.DOMAIN.COM = { kdc = 192.168.0.1 admin_server = 192.168.0.1 default_domain = my.domain.com } smb.conf contains: security = ADS password server = ad1.my.domain.com AFTER CHANGE [realms] MY.DOMAIN.COM = { kdc = 192.168.0.2 admin_server = 192.168.0.2 default_domain = my.domain.com } smb.conf contains: security = ADS password server = ad2.my.domain.com The way I know the system is pointing to the old ad server is by doing wbinfo -g followed by lsof -i tcp -nP |grep winbind -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] switching winbind's ad without reboot?
On Wed, Nov 10, 2010 at 2:00 PM, John Stile j...@stilen.com wrote: I found a way to change the AD for winbind without rebooting on Debian and Gentoo, however I wonder if this is a bug. First I stop smbd, nmbd, windbindd if they are running. Then I remove some files. On Debian, I removed: /var/run/samba/smb_krb5/krb5.conf.MS /var/run/samba/gencache.tdb On Gentoo, I removed: /var/cache/samba/smb_krb5/krb5.conf.MS /var/cache/samba/gencache.tdb was similar for NT4 domains, even a `net rpc join` doesn't reset the local secrets.tdb so the join to a new domain won't actually work unless you delete it first. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] idmap trouble
Hi, I am hoping someone can point out what I am doing wrong, I am upgrading samba from 3.0.37 to 3.5.6 and running into trouble with idmapping using ADS security. I have multiple linux boxes running 3.0.37 and when I execute getent passwd I get: # getent passwd DOMAIN+gregorcy gregorcy:*:2933:1013:Brian Gregorcy:/home/DOMAIN/gregorcy:/bin/bash on all the boxes running 3.0.37. On my new box running 3.5.6 I get: # getent passwd DOMAIN+gregorcy gregorcy:*:502:506::/home/DOMAIN/gregorcy:/bin/bash Which is not what I had hoped for. Both machines are joined to my domain and allow me to ssh into them using my AD cred, just the uid gid are not lining up. My 3.0.37 smb.conf [global] workgroup = DOMAIN netbios name= harley realm = DOMAIN.UTAH.EDU server string = harley security= ADS preferred master= no client use spnego = yes server signing = auto encrypt passwords = yes nt acl support = yes acl map full control= yes socket options = TCP_NODELAY SO_RCVBUF=8192 SO_SNDBUF=8192 template shell = /bin/false password server = * log level = 3 log file= /var/log/samba/%m max log size= 100 preferred master= No dns proxy = No wins server = 192.168.1.100 192.168.1.101 winbind cache time = 0 winbind nested groups = yes allow trusted domains = No idmap backend = rid:KPAK=500-1 idmap uid = 500-1 idmap gid = 500-1 template shell = /bin/bash winbind use default domain = Yes winbind separator = + winbind enum users = yes winbind enum groups = yes winbind use default domain = yes obey pam restrictions = yes My 3.5.6 smb.conf [global] workgroup = DOMAIN netbios name= vwww3 realm = DOMAIN.UTAH.EDU server string = web3 security= ADS preferred master= no client use spnego = yes server signing = auto encrypt passwords = yes nt acl support = yes acl map full control= yes wide links = no password server = * log level = 3 log file= /var/log/samba/%m max log size= 100 wins server = 192.168.1.100 192.168.1.101 winbind offline logon= yes idmap domains = default, domain.utah.edu idmap config default: default = yes idmap config domain.utah.edu: range = 500-1 idmap config domain.utah.edu: backend = ad idmap alloc backend = tdb idmap uid = 500-1 idmap gid = 500-1 winbind separator = + winbind use default domain = yes winbind enum users = yes winbind enum groups = yes winbind refresh tickets = yes winbind nested groups = yes client ntlmv2 auth = yes encrypt passwords = yes template shell = /bin/bash allow trusted domains = yes Thanks for any help, Brian Gregorcy -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] idmap trouble
Hi Brian, the id mapping configuration hat changed (again) between 3.2 and 3.3. The idmap domains setting has vanished. Could you try and use the id mapping config from your 3.0 setup, i.e. simply this idmap backend = rid:KPAK=500-1 idmap uid = 500-1 idmap gid = 500-1 instead of this: idmap domains = default, domain.utah.edu idmap config default: default = yes idmap config domain.utah.edu: range = 500-1 idmap config domain.utah.edu: backend = ad idmap alloc backend = tdb idmap uid = 500-1 idmap gid = 500-1 Er, btw, can it be that KPAK should actually read DOMAIN in the first section? ;-) Then, for debugging, please paste the output of wbinfo -n DOMAIN+gregorcy I should list the sid of gegorcy. I expect the rid to be 2433. What I think happened is that your 3.5 config fell back to tdb id mapping because the domain DOMAIN did not match the domain name domain.utah.edu that you gave in the idmap config... The recommended setting would be to have a tdb backend default idmap range and explicit and disjoint ranges with rid backend vor various domains like this: idmap backend = tdb idmap uid = 10-19 idmap gid = 10-19 idmap config DOMAIN : backend = rid idmap config DOMAIN : range = 20-29 (see man idmap_rid) But you should not change your config in that way if you are updating a production server, since it would change the uids/gids that users are accessing the disk with, and so, permissions and ownerships would be broken. Hope this helps, Michael gregorcy wrote: Hi, I am hoping someone can point out what I am doing wrong, I am upgrading samba from 3.0.37 to 3.5.6 and running into trouble with idmapping using ADS security. I have multiple linux boxes running 3.0.37 and when I execute getent passwd I get: # getent passwd DOMAIN+gregorcy gregorcy:*:2933:1013:Brian Gregorcy:/home/DOMAIN/gregorcy:/bin/bash on all the boxes running 3.0.37. On my new box running 3.5.6 I get: # getent passwd DOMAIN+gregorcy gregorcy:*:502:506::/home/DOMAIN/gregorcy:/bin/bash Which is not what I had hoped for. Both machines are joined to my domain and allow me to ssh into them using my AD cred, just the uid gid are not lining up. My 3.0.37 smb.conf [global] workgroup = DOMAIN netbios name= harley realm = DOMAIN.UTAH.EDU server string = harley security= ADS preferred master= no client use spnego = yes server signing = auto encrypt passwords = yes nt acl support = yes acl map full control= yes socket options = TCP_NODELAY SO_RCVBUF=8192 SO_SNDBUF=8192 template shell = /bin/false password server = * log level = 3 log file= /var/log/samba/%m max log size= 100 preferred master= No dns proxy = No wins server = 192.168.1.100 192.168.1.101 winbind cache time = 0 winbind nested groups = yes allow trusted domains = No idmap backend = rid:KPAK=500-1 idmap uid = 500-1 idmap gid = 500-1 template shell = /bin/bash winbind use default domain = Yes winbind separator = + winbind enum users = yes winbind enum groups = yes winbind use default domain = yes obey pam restrictions = yes My 3.5.6 smb.conf [global] workgroup = DOMAIN netbios name= vwww3 realm = DOMAIN.UTAH.EDU server string = web3 security= ADS preferred master= no client use spnego = yes server signing = auto encrypt passwords = yes nt acl support = yes acl map full control= yes wide links = no password server = * log level = 3 log file= /var/log/samba/%m max log size
Re: [Samba] idmap trouble
That fixed it thanks, Er, btw, can it be that KPAK should actually read DOMAIN in the first section? ;-) Yep typo on my part I was trying different configs and made that error. 3.5 wbinfo -n DOMAIN+gregorcy S-1-5-21-3579845861-693198353-1953747050-2433 SID_USER (1) 3.0.37 wbinfo -n CHEMENG+gregorcy S-1-5-21-3579845861-693198353-1953747050-2433 User (1) Works :) Thanks for the help, --Brian On 11/10/10 14:38, Michael Adam wrote: Hi Brian, the id mapping configuration hat changed (again) between 3.2 and 3.3. The idmap domains setting has vanished. Could you try and use the id mapping config from your 3.0 setup, i.e. simply this idmap backend = rid:KPAK=500-1 idmap uid = 500-1 idmap gid = 500-1 instead of this: idmap domains = default, domain.utah.edu idmap config default: default = yes idmap config domain.utah.edu: range = 500-1 idmap config domain.utah.edu: backend = ad idmap alloc backend = tdb idmap uid = 500-1 idmap gid = 500-1 Er, btw, can it be that KPAK should actually read DOMAIN in the first section? ;-) Then, for debugging, please paste the output of wbinfo -n DOMAIN+gregorcy I should list the sid of gegorcy. I expect the rid to be 2433. What I think happened is that your 3.5 config fell back to tdb id mapping because the domain DOMAIN did not match the domain name domain.utah.edu that you gave in the idmap config... The recommended setting would be to have a tdb backend default idmap range and explicit and disjoint ranges with rid backend vor various domains like this: idmap backend = tdb idmap uid = 10-19 idmap gid = 10-19 idmap config DOMAIN : backend = rid idmap config DOMAIN : range = 20-29 (see man idmap_rid) But you should not change your config in that way if you are updating a production server, since it would change the uids/gids that users are accessing the disk with, and so, permissions and ownerships would be broken. Hope this helps, Michael gregorcy wrote: Hi, I am hoping someone can point out what I am doing wrong, I am upgrading samba from 3.0.37 to 3.5.6 and running into trouble with idmapping using ADS security. I have multiple linux boxes running 3.0.37 and when I execute getent passwd I get: # getent passwd DOMAIN+gregorcy gregorcy:*:2933:1013:Brian Gregorcy:/home/DOMAIN/gregorcy:/bin/bash on all the boxes running 3.0.37. On my new box running 3.5.6 I get: # getent passwd DOMAIN+gregorcy gregorcy:*:502:506::/home/DOMAIN/gregorcy:/bin/bash Which is not what I had hoped for. Both machines are joined to my domain and allow me to ssh into them using my AD cred, just the uid gid are not lining up. My 3.0.37 smb.conf [global] workgroup = DOMAIN netbios name= harley realm = DOMAIN.UTAH.EDU server string = harley security= ADS preferred master= no client use spnego = yes server signing = auto encrypt passwords = yes nt acl support = yes acl map full control= yes socket options = TCP_NODELAY SO_RCVBUF=8192 SO_SNDBUF=8192 template shell = /bin/false password server = * log level = 3 log file= /var/log/samba/%m max log size= 100 preferred master= No dns proxy = No wins server = 192.168.1.100 192.168.1.101 winbind cache time = 0 winbind nested groups = yes allow trusted domains = No idmap backend = rid:KPAK=500-1 idmap uid = 500-1 idmap gid = 500-1 template shell = /bin/bash winbind use default domain = Yes winbind separator = + winbind enum users = yes winbind enum groups = yes winbind use default domain = yes obey pam restrictions = yes My 3.5.6 smb.conf [global] workgroup = DOMAIN netbios name= vwww3 realm = DOMAIN.UTAH.EDU server string = web3 security= ADS preferred master= no client use spnego
[Samba] tree connect failed: NT_STATUS_BAD_NETWORK_NAME
Let me start off by apologizing for this rambling post. I have been working on this problem all day and have not gotten anywhere. This is my first time working with samba and I have been tasked to set up a share that does not require a password for access. I have been tinkering with the configuration file but I cannot get it to work. Here is my current smb.conf file that I have: global] security = share workgroup = somedomain.com guest account = murex log file = /var/log/samba/%m.log [Murex] comment = Murex pickup path = /apps/murex/pickup writable = yes read only = no guest ok = yes [testdir] comment = Troubleshooting area path = /sambatest writable = yes read only = no guest ok = yes guest only = yes When I try to access it either of the shares (Murex or testdir) using smbclient, I get the following: # smbclient //localhost/murex Password: Domain=[SOMEDOMAIN.COM] OS=[Unix] Server=[Samba 3.0.28-0.el5.8] Server not using user level security and no password supplied. tree connect failed: NT_STATUS_BAD_NETWORK_NAME # Not sure why it is stating that no password was supplied because I entered a password. I tried both the root password since that is what I was logged in as and the password for the user murex. The murex user is a valid unix account. I also created a samba password for the murex user (I used smbpasswd -a murex). This was done recently but it did not change anything. I can see the shares from our windows clients but when I try and open up a share (by double clicking on it) or map a drive to the share I get the following errors in the log files: [2010/11/10 18:49:35, 0] smbd/service.c:make_connection_snum(1003) '/apps/murex/pickup' does not exist or permission denied when connecting to [Murex] Error was Permission denied [2010/11/10 18:49:36, 0] smbd/service.c:make_connection_snum(1003) '/sambatest' does not exist or permission denied when connecting to [testdir] Error was Permission denied I also see these same errors in the log files when I run the smbclient command from above. The unix permissions on these directories are set at 777. This is driving me nuts because I think it has got be a pretty minor configuration error but I can't figure out what it is. Does anyone have any ideas of what I am missing or doing wrong? Thanks, Ayo Adeyeri InfoHedge Technologies LLC 747 Third Avenue, 4th Floor, New York, NY 10017 ayo.adey...@infohedge.netmailto:ayo.adey...@infohedge.net Office: (212) 531-5835 This e-mail message, and any enclosures, is intended only for the persons to whom it is addressed, and may contain confidential information. Any use, distribution, modification, copying or disclosure by any other person is strictly prohibited. If you receive this message in error, please notify the sender by return e-mail and delete this message from your computer. InfoHedge disclaims all responsibility from and accepts no liability whatsoever for any incorrect, misleading or altered information contained herein, or for the consequences of any unauthorized person acting, or refraining from acting, on any information contained in this message. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
[Samba] samba over internet slow with images/thumbnails
Hello, Iam providing access to many images on a share to windows users from my linux samba over the internet. The problem is that I need thumbnail viewing because of the nature of this library and thumbnails take to much time to load it seems as though it is actually loading the entire image to then display the small thumbnail. Is this true? Also isnt there some kind of caching that can be activated? Anything to increase this speed. Each image weights about 800kb, and the user is sitting on a 1mbit line but it takes several seconds for each image. Simply not the best experience. Thanks Mark -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
Re: [Samba] samba over internet slow with images/thumbnails
JP CR put forth on 11/10/2010 11:42 PM: Hello, Iam providing access to many images on a share to windows users from my linux samba over the internet. The problem is that I need thumbnail viewing because of the nature of this library and thumbnails take to much time to load it seems as though it is actually loading the entire image to then display the small thumbnail. Is this true? Probably yes, especially if there is not thumbs.db file. The remote Windows client, when in thumbnail view in Explorer, will download the entire image and create the thumbnail. In a directory with hundreds or thousands of image files, this will take a long time, possibly hours, over a 1Mb/s broadband connection, for a directory with say 5000 image files. Also isnt there some kind of caching that can be activated? The caching would have to take place on the client side. AFAIK Explorer performs no file caching. Any web browser will, unless it's been disabled. Anything to increase this speed. Each image weights about 800kb, and the user is sitting on a 1mbit line but it takes several seconds for each image. It should take about 6.4 seconds to load each file, assuming no overhead. You will have overhead, so you're looking at 6.4 to 10 or more seconds per file. How many files are in this directory? Simply not the best experience. Well of course not. You're using the wrong tool for the job. You should be hosting/serving up these files via http. I find lighttpd + curator is a good combination for serving lots of static image files. It works great, as long as the filenames have no spaces in them. Other special characters seem to work fine. If you use this method, and the contents of the image directory change frequently, simply cron curator to run on that directory each night. Using lighttpd (or any web server) and curator, you can achieve something like this: http://www.hardwarefreak.com/construction/ This image page is served from a 500 Kbit/s upstream channel ADSL line. This should give you an idea of the performance you might expect. Click on each thumbnail image to see the full size image. Click on the links in the top right corner to see alternative views and sorting of the thumbnails by file name. You can also click the left and right arrows for a manual slide show, or simply click the top left/right corner of each full size image. To do all of this, simply install curator and the dependencies, cd into the image directory, and type curator ./. It's that simple. This command created everything you see at the link above. Your *nix distribution may or may not have a curator package. If not, try here: http://curator.sourceforge.net/local/ Also note curator does not work well, if at all, with GIF files. JPGs work great. -- Stan -- 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 fa203f4 s3/time_audit: fix a change that was just for debuggin purposeÑ from 7e3419f s4-pyldb: Handle internal errors in py_ldb_contains() properly http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit fa203f4541190b01b1f82f1ff9a47c31152c2412 Author: Björn Jacke b...@sernet.de Date: Wed Nov 10 12:39:41 2010 +0100 s3/time_audit: fix a change that was just for debuggin purposeÑ Autobuild-User: Björn Jacke bja...@samba.org Autobuild-Date: Wed Nov 10 12:28:41 UTC 2010 on sn-devel-104 --- Summary of changes: source3/modules/vfs_time_audit.c |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) Changeset truncated at 500 lines: diff --git a/source3/modules/vfs_time_audit.c b/source3/modules/vfs_time_audit.c index 4e6d65e..84e4175 100644 --- a/source3/modules/vfs_time_audit.c +++ b/source3/modules/vfs_time_audit.c @@ -838,10 +838,10 @@ static int smb_time_audit_chdir(vfs_handle_struct *handle, const char *path) clock_gettime_mono(ts1); result = SMB_VFS_NEXT_CHDIR(handle, path); clock_gettime_mono(ts2); - timediff = nsec_time_diff(ts2,ts1); + timediff = nsec_time_diff(ts2,ts1)*1.0e-9; if (timediff audit_timeout) { - smb_time_audit_log(chdir, timediff*1.0e-9); + smb_time_audit_log(chdir, timediff); } return result; -- Samba Shared Repository
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 7851c49 Split up further. via 2f489ce Split up BuildResultStore further. via 1300d88 Move more functionality to buildfarm.BuildFarm. via cea251c Remove trees from BuildResultStore. via ecd6140 Add BuildFarm object, remove compilers and host_age from BuildResultStore. via 0669b47 Fix smtplib.Connection - smtplib.SMTP. via 69c7574 Add python version of import-and-analyse.pl. from 6b3372f Remove unused variable. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 7851c49eaa94c31e8e89cef7d5551722bbb4e451 Author: Jelmer Vernooij jel...@samba.org Date: Wed Nov 10 09:31:49 2010 +0100 Split up further. commit 2f489ce85384cc5a81237187db1f39dac12d771b Author: Jelmer Vernooij jel...@samba.org Date: Wed Nov 10 09:26:07 2010 +0100 Split up BuildResultStore further. commit 1300d8845dd004d73d8c0e2cc780093ba442b8c6 Author: Jelmer Vernooij jel...@samba.org Date: Wed Nov 10 08:32:38 2010 +0100 Move more functionality to buildfarm.BuildFarm. commit cea251c95ae5c166f6b40da896f3b395d1d3ca6a Author: Jelmer Vernooij jel...@samba.org Date: Wed Nov 10 08:23:49 2010 +0100 Remove trees from BuildResultStore. commit ecd61407c8e7c6e59a18c810a3038493f54e37e9 Author: Jelmer Vernooij jel...@samba.org Date: Wed Nov 10 08:19:01 2010 +0100 Add BuildFarm object, remove compilers and host_age from BuildResultStore. commit 0669b471176c6204afa1cb75b37e7c7d314f690e Author: Jelmer Vernooij jel...@samba.org Date: Wed Nov 10 08:09:05 2010 +0100 Fix smtplib.Connection - smtplib.SMTP. commit 69c7574774c799c6a2ee3c11b6dd51640e1b1eb7 Author: Jelmer Vernooij jel...@samba.org Date: Wed Nov 10 08:08:48 2010 +0100 Add python version of import-and-analyse.pl. --- Summary of changes: admin.py |6 +- buildfarm/__init__.py| 107 - buildfarm/data.py| 274 +++--- buildfarm/tests/__init__.py | 69 ++- buildfarm/tests/test_data.py | 122 +++ import-and-analyse.py| 181 mail-dead-hosts.py |8 +- web/build.py | 17 ++-- 8 files changed, 562 insertions(+), 222 deletions(-) create mode 100644 import-and-analyse.py Changeset truncated at 500 lines: diff --git a/admin.py b/admin.py index d94b572..909d303 100755 --- a/admin.py +++ b/admin.py @@ -19,7 +19,7 @@ from buildfarm import ( hostdb, -open_hostdb, +BuildFarm, ) import commands import os @@ -28,7 +28,9 @@ import sys import time from email.MIMEText import MIMEText -db = open_hostdb() +buildfarm = BuildFarm() + +db = buildfarm.hostdb dry_run = False diff --git a/buildfarm/__init__.py b/buildfarm/__init__.py index e133cda..30c6005 100644 --- a/buildfarm/__init__.py +++ b/buildfarm/__init__.py @@ -17,9 +17,108 @@ # along with this program; if not, write to the Free Software # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. +import ConfigParser import os -def open_hostdb(): -from buildfarm import hostdb -return hostdb.HostDatabase( -os.path.join(os.path.dirname(__file__), .., hostdb.sqlite)) + +class Tree(object): +A tree to build. + +def __init__(self, name, scm, repo, branch, subdir=, srcdir=): +self.name = name +self.repo = repo +self.scm = scm +self.branch = branch +self.subdir = subdir +self.srcdir = srcdir +self.scm = scm + +def __repr__(self): +return %s %r % (self.__class__.__name__, self.name) + + +def read_trees_from_conf(path): +Read trees from a configuration file. +ret = {} +cfp = ConfigParser.ConfigParser() +cfp.readfp(open(path)) +for s in cfp.sections(): +ret[s] = Tree(name=s, **dict(cfp.items(s))) +return ret + + +def lcov_extract_percentage(text): +m = re.search('\td class=headerItem.*?\Code\nbsp\;covered\:\\/td\.*?\n.*?\td class=headerValue.*?\([0-9.]+) \%', text) +if m: +return m.group(1) +else: +return None + + +class BuildFarm(object): + +LCOVHOST = magni +OLDAGE = 60*60*4, +DEADAGE = 60*60*24*4 + +def __init__(self, path=None): +if path is None: +path = os.path.abspath(os.path.join(os.path.dirname(__file__), ..)) +self.path = path +self.webdir = os.path.join(self.path, web) +if not os.path.isdir(path): +raise Exception(web directory %s does not exist % self.webdir) +self.trees = read_trees_from_conf(os.path.join(self.webdir, trees.conf)) +self.builds = self._open_build_results() +self.hostdb = self._open_hostdb() +self.compilers = self._load_compilers() +self.lcovdir =
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 211abe8 More improvements to import-and-analyse. from ee296a4 Simplify some code. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 211abe8c5db88a77017077495fd37f3a7d7cf019 Author: Jelmer Vernooij jel...@samba.org Date: Wed Nov 10 11:23:57 2010 +0100 More improvements to import-and-analyse. --- Summary of changes: buildfarm/__init__.py |4 +- buildfarm/data.py | 96 ++--- buildfarm/util.py |3 +- import-and-analyse.py | 18 - 4 files changed, 38 insertions(+), 83 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/__init__.py b/buildfarm/__init__.py index b12fb7b..2370940 100644 --- a/buildfarm/__init__.py +++ b/buildfarm/__init__.py @@ -71,7 +71,7 @@ class BuildFarm(object): raise Exception(web directory %s does not exist % self.webdir) self.trees = read_trees_from_conf(os.path.join(self.webdir, trees.conf)) self.builds = self._open_build_results() -self.upload_builds = self._open_upload_builds_results() +self.upload_builds = self._open_upload_build_results() self.hostdb = self._open_hostdb() self.compilers = self._load_compilers() self.lcovdir = os.path.join(self.path, lcov/data) @@ -131,7 +131,7 @@ class BuildFarm(object): def get_new_builds(self): from buildfarm import data -for host in self.hostsdb.hosts(): +for host in self.hostdb.hosts(): for tree in self.trees: for compiler in self.compilers: # By building the log file name this way, using only the list of diff --git a/buildfarm/data.py b/buildfarm/data.py index 7062431..493badb 100644 --- a/buildfarm/data.py +++ b/buildfarm/data.py @@ -175,6 +175,10 @@ class Build(object): self.host = host self.compiler = compiler self.rev = rev +if rev is None: +self.basename = self._store.build_fname(self.tree, self.host, self.compiler) +else: +self.basename = self._store.build_fname(self.tree, self.host, self.compiler, self.rev) ### # the mtime age is used to determine if builds are still happening @@ -183,26 +187,22 @@ class Build(object): def age_mtime(self): get the age of build from mtime -file = self._store.build_fname(self.tree, self.host, self.compiler, self.rev) - -st = os.stat(%s.log % file) +st = os.stat(%s.log % self.basename) return time.time() - st.st_mtime def age_ctime(self): get the age of build from ctime -file = self._store.build_fname(self.tree, self.host, self.compiler, self.rev) - -st = os.stat(%s.log % file) +st = os.stat(%s.log % self.basename) return time.time() - st.st_ctime def read_log(self): read full log file -return open(self._store.build_fname(self.tree, self.host, self.compiler, self.rev)+.log, r) +return open(self.basename+.log, r) def read_err(self): read full err file try: -return open(self._store.build_fname(self.tree, self.host, self.compiler, self.rev)+.err, 'r') +return open(self.basename+.err, 'r') except IOError: # No such file return StringIO() @@ -232,7 +232,7 @@ class Build(object): timestamp = None f = self.read_log() try: -for l in f.readlines(): +for l in f: if l.startswith(BUILD COMMIT REVISION: ): commit_revid = l.split(:, 1)[1].strip() elif l.startswith(BUILD REVISION: ): @@ -270,9 +270,8 @@ class CachingBuild(Build): to calculate. def revision_details(self): -file = self._store.build_fname(self.tree, self.host, self.compiler, self.rev) cachef = self._store.cache_fname(self.tree, self.host, self.compiler, self.rev) -st1 = os.stat(%s.log % file) +st1 = os.stat(%s.log % self.basename) try: st2 = os.stat(%s.revision % cachef) @@ -297,9 +296,8 @@ class CachingBuild(Build): return (revid, commit_revid, timestamp) def err_count(self): -file = self._store.build_fname(self.tree, self.host, self.compiler, self.rev) cachef = self._store.cache_fname(self.tree, self.host, self.compiler, self.rev) -st1 = os.stat(%s.err % file) +st1 = os.stat(%s.err % self.basename) try: st2 = os.stat(%s.errcount % cachef) @@ -318,10 +316,9 @@ class CachingBuild(Build): return ret def status(self): -file = self._store.build_fname(self.tree, self.host, self.compiler, self.rev)
[SCM] build.samba.org - branch master updated
The branch, master has been updated via ee296a4 Simplify some code. via 1459e8f Simplify new build iteration. via 5bd8f3b More work on import and analyse. from 7851c49 Split up further. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit ee296a4ceeb7338d3f5d572f2f591ba4ab3a0f07 Author: Jelmer Vernooij jel...@samba.org Date: Wed Nov 10 10:05:34 2010 +0100 Simplify some code. commit 1459e8f65ea41886db18c6387e7b4968f3a2723b Author: Jelmer Vernooij jel...@samba.org Date: Wed Nov 10 09:44:57 2010 +0100 Simplify new build iteration. commit 5bd8f3b6c81edcc765b0c57c460a94c7bed6ebde Author: Jelmer Vernooij jel...@samba.org Date: Wed Nov 10 09:38:26 2010 +0100 More work on import and analyse. --- Summary of changes: buildfarm/__init__.py| 19 buildfarm/data.py| 14 ++- buildfarm/tests/test_util.py |8 --- buildfarm/util.py|5 import-and-analyse.py| 48 ++ 5 files changed, 38 insertions(+), 56 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/__init__.py b/buildfarm/__init__.py index 30c6005..b12fb7b 100644 --- a/buildfarm/__init__.py +++ b/buildfarm/__init__.py @@ -19,6 +19,7 @@ import ConfigParser import os +import re class Tree(object): @@ -70,6 +71,7 @@ class BuildFarm(object): raise Exception(web directory %s does not exist % self.webdir) self.trees = read_trees_from_conf(os.path.join(self.webdir, trees.conf)) self.builds = self._open_build_results() +self.upload_builds = self._open_upload_builds_results() self.hostdb = self._open_hostdb() self.compilers = self._load_compilers() self.lcovdir = os.path.join(self.path, lcov/data) @@ -81,6 +83,10 @@ class BuildFarm(object): from buildfarm import data return data.BuildResultStore(os.path.join(self.path, data, oldrevs)) +def _open_upload_build_results(self): +from buildfarm import data +return data.UploadBuildResultStore(os.path.join(self.path, data, upload)) + def _open_hostdb(self): from buildfarm import hostdb return hostdb.HostDatabase( @@ -122,3 +128,16 @@ class BuildFarm(object): if self.readonly: util.FileSave(cachefile, ret) return perc + +def get_new_builds(self): +from buildfarm import data +for host in self.hostsdb.hosts(): +for tree in self.trees: +for compiler in self.compilers: +# By building the log file name this way, using only the list of +# hosts, trees and compilers as input, we ensure we +# control the inputs +try: +yield self.upload_builds.get_build(host, tree, compiler) +except data.NoSuchBuildError: +continue diff --git a/buildfarm/data.py b/buildfarm/data.py index 7ff8c92..7062431 100644 --- a/buildfarm/data.py +++ b/buildfarm/data.py @@ -226,12 +226,11 @@ class Build(object): :return: Tuple with revision id and timestamp (if available) -file = self._store.build_fname(self.tree, self.host, self.compiler, self.rev) revid = None commit_revid = None timestamp = None -f = open(%s.log % file, 'r') +f = self.read_log() try: for l in f.readlines(): if l.startswith(BUILD COMMIT REVISION: ): @@ -262,15 +261,8 @@ class Build(object): def err_count(self): get status of build -file = self._store.build_fname(self.tree, self.host, self.compiler, self.rev) - -try: -err = util.FileLoad(%s.err % file) -except OSError: -# File does not exist -return 0 - -return util.count_lines(err) +file = self.read_err() +return len(file.readlines()) class CachingBuild(Build): diff --git a/buildfarm/tests/test_util.py b/buildfarm/tests/test_util.py index 3a1d5a4..79a311d 100755 --- a/buildfarm/tests/test_util.py +++ b/buildfarm/tests/test_util.py @@ -22,14 +22,6 @@ import unittest from buildfarm import util -class CountLinesTests(unittest.TestCase): - -def test_simple(self): -self.assertEquals(2, util.count_lines(foo\nbar)) -self.assertEquals(1, util.count_lines(bar)) -self.assertEquals(1, util.count_lines()) - - class ChangeExtensionTests(unittest.TestCase): def test_simple(self): diff --git a/buildfarm/util.py b/buildfarm/util.py index 83346c1..60b263e 100644 --- a/buildfarm/util.py +++ b/buildfarm/util.py @@ -77,11 +77,6 @@ def ChangeExtension(fname, ext): return %s.%s % (base, ext)
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated via 7e3419f s4-pyldb: Handle internal errors in py_ldb_contains() properly via 38585a7 s4-pydsdb: py_dsdb_am_rodc() may call samdb_rodc() only once to get the job done :) via d073c5f s4-pydsdb: py_dsdb_load_partition_usn() - simplify error handling in via 89440dd s4-pydsdb-py_samdb_ntds_objectGUID(): Avoid potential memory leak via 06b0596 s4-pydsdb-py_dsdb_get_oid_from_attid(): Avoid potential memory leak via ad5a399 s4-pydsdb.c: Fix small memory leak in py_samdb_set_domain_sid() from 4b978b3 s4/syntax: Add tests for DN+String and DN+Binary http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit 7e3419f32f0a46cbd4926cbc98bf54da27749d31 Author: Kamen Mazdrashki kame...@samba.org Date: Tue Nov 9 00:21:57 2010 +0200 s4-pyldb: Handle internal errors in py_ldb_contains() properly It is an exceptional condition for ldb_search() to return more than one results during SCOPE_BASE search on DN Autobuild-User: Kamen Mazdrashki kame...@samba.org Autobuild-Date: Wed Nov 10 09:02:00 UTC 2010 on sn-devel-104 commit 38585a74905185e87494c7326d248495b5e01e49 Author: Kamen Mazdrashki kame...@samba.org Date: Mon Nov 8 23:57:31 2010 +0200 s4-pydsdb: py_dsdb_am_rodc() may call samdb_rodc() only once to get the job done :) commit d073c5f23ae6563a65af73fd6dc6099011da Author: Kamen Mazdrashki kame...@samba.org Date: Mon Nov 8 23:50:23 2010 +0200 s4-pydsdb: py_dsdb_load_partition_usn() - simplify error handling in and print on which partition error has occured commit 89440dd617eb4ff64cda8bf97f1f22ddf94bf717 Author: Kamen Mazdrashki kame...@samba.org Date: Mon Nov 8 23:40:14 2010 +0200 s4-pydsdb-py_samdb_ntds_objectGUID(): Avoid potential memory leak in case py_ldb is not a valid LDB commit 06b0596537e66ed928b24af059ada0472a375f3b Author: Kamen Mazdrashki kame...@samba.org Date: Mon Nov 8 23:36:09 2010 +0200 s4-pydsdb-py_dsdb_get_oid_from_attid(): Avoid potential memory leak in case py_ldb is not a valid LDB commit ad5a399d54fba103822ba9cbea4515d52eafd1fe Author: Kamen Mazdrashki kame...@samba.org Date: Mon Nov 8 23:09:44 2010 +0200 s4-pydsdb.c: Fix small memory leak in py_samdb_set_domain_sid() --- Summary of changes: source4/dsdb/pydsdb.c | 26 ++ source4/lib/ldb/pyldb.c | 13 + 2 files changed, 23 insertions(+), 16 deletions(-) Changeset truncated at 500 lines: diff --git a/source4/dsdb/pydsdb.c b/source4/dsdb/pydsdb.c index 2471548..16a4cd4 100644 --- a/source4/dsdb/pydsdb.c +++ b/source4/dsdb/pydsdb.c @@ -134,6 +134,7 @@ static PyObject *py_samdb_set_domain_sid(PyLdbObject *self, PyObject *args) sid = dom_sid_parse_talloc(NULL, PyString_AsString(py_sid)); ret = samdb_set_domain_sid(ldb, sid); + talloc_free(sid); if (!ret) { PyErr_SetString(PyExc_RuntimeError, set_domain_sid failed); return NULL; @@ -244,14 +245,14 @@ static PyObject *py_dsdb_get_oid_from_attid(PyObject *self, PyObject *args) if (!PyArg_ParseTuple(args, Oi, py_ldb, attid)) return NULL; + PyErr_LDB_OR_RAISE(py_ldb, ldb); + mem_ctx = talloc_new(NULL); if (mem_ctx == NULL) { PyErr_NoMemory(); return NULL; } - PyErr_LDB_OR_RAISE(py_ldb, ldb); - schema = dsdb_get_schema(ldb, NULL); if (!schema) { @@ -413,19 +414,18 @@ static PyObject *py_samdb_ntds_objectGUID(PyObject *self, PyObject *args) TALLOC_CTX *mem_ctx; const struct GUID *guid; - mem_ctx = talloc_new(NULL); - if (mem_ctx == NULL) { - PyErr_NoMemory(); - return NULL; - } - if (!PyArg_ParseTuple(args, O, py_ldb)) { - talloc_free(mem_ctx); return NULL; } PyErr_LDB_OR_RAISE(py_ldb, ldb); + mem_ctx = talloc_new(NULL); + if (mem_ctx == NULL) { + PyErr_NoMemory(); + return NULL; + } + guid = samdb_ntds_objectGUID(ldb); if (guid == NULL) { PyErr_SetString(PyExc_RuntimeError, Failed to find NTDS GUID); @@ -483,8 +483,10 @@ static PyObject *py_dsdb_load_partition_usn(PyObject *self, PyObject *args) ret = dsdb_load_partition_usn(ldb, dn, highest_uSN, urgent_uSN); if (ret != LDB_SUCCESS) { - char *errstr = talloc_asprintf(mem_ctx, Failed to load partition uSN - %s, ldb_errstring(ldb)); - PyErr_SetString(PyExc_RuntimeError, errstr); + PyErr_Format(PyExc_RuntimeError, + Failed to load partition [%s] uSN - %s, + ldb_dn_get_linearized(dn), +
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 4db119b We no longer have CVS-based projects, and should not introduce any newer; remove cvslog.pl. from 211abe8 More improvements to import-and-analyse. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 4db119b01c2f86fbff7829d40c2ed45309f2c6a9 Author: Jelmer Vernooij jel...@samba.org Date: Wed Nov 10 11:26:30 2010 +0100 We no longer have CVS-based projects, and should not introduce any newer; remove cvslog.pl. --- Summary of changes: cvslog.pl | 282 - 1 files changed, 0 insertions(+), 282 deletions(-) delete mode 100755 cvslog.pl Changeset truncated at 500 lines: diff --git a/cvslog.pl b/cvslog.pl deleted file mode 100755 index 11d559e..000 --- a/cvslog.pl +++ /dev/null @@ -1,282 +0,0 @@ -#!/usr/bin/perl -w -# -# Extract information about recent CVS commits -# -# tri...@samba.org, April 2001 - -use strict; -use util; -use POSIX; -use Data::Dumper; -use File::stat; -use Date::Parse; - - -# parse a CVS date -sub cvs_parse_date($) -{ -my $s = shift; - -if (! ($s =~ /@/)) { - return str2time($s); -} - -if ($s =~ /(.*) (.*) ([0-9]+), ([0-9]+) @ ([0-9]+):([0-9]+)/) { - my $day = $1; - my $month = $2; - my $mday = $3; - my $year = $4; - my $hour = $5; - my $min = $6; - my (%months) = ('January' = 1, 'February' = 2, 'March' = 3, 'April' = 4, - 'May' = 5, 'June' = 6, 'July' = 7, 'August' = 8, 'September' = 9, - 'October' = 10, 'November' = 11, 'December' = 12); - my $t = mktime(0, $min, $hour, $mday, $months{$month}-1, $year-1900); - return $t; -} - -print ERROR: bad date format $s\n; -return 0; -} - - - -# push an entry onto the array - -# FIXME: This incorrectly handles multiple commits with the same log -# message. The cvslog output only shows the revisions for the last -# commit, but actually we want to either not coalesce, or show the -# overall delta. - -sub push_entry($) -{ - my $entry = shift; - my $log = shift; - my $days = shift; - my $tree = shift; - my $tag = shift; - my $lastentry = $log-[$#{$log}]; - - if ($lastentry-{DATE} $lastentry-{AUTHOR} - ($lastentry-{DATE} $entry-{DATE}-600) - ($lastentry-{AUTHOR} eq $entry-{AUTHOR}) - ((!$lastentry-{TAG} !$entry-{TAG}) || - ($lastentry-{TAG} eq $entry-{TAG})) - ((!$lastentry-{MESSAGE} !$entry-{MESSAGE}) || - ($lastentry-{MESSAGE} eq $entry-{MESSAGE}))) { - -if (exists $lastentry-{FILES}) { - $lastentry-{FILES} .= $entry-{FILES}; -} else { - $lastentry-{FILES} = $entry-{FILES}; -} - -if (exists $lastentry-{ADDED}) { - $lastentry-{ADDED} .= $entry-{ADDED}; -} else { - $lastentry-{ADDED} = $entry-{ADDED}; -} - -if (exists $lastentry-{REMOVED}) { - $lastentry-{REMOVED} .= $entry-{REMOVED}; -} else { - $lastentry-{REMOVED} = $entry-{REMOVED}; -} - -if (exists $lastentry-{REVISIONS}) { - $lastentry-{REVISIONS} = {%{$lastentry-{REVISIONS}},%{$entry-{REVISIONS}}}; -} else { - $lastentry-{REVISIONS} = $entry-{REVISIONS}; -} - } else { -if (($entry-{DATE} time() - $days*24*60*60) - ((!$entry-{TAG} !$tag) || -($entry-{TAG} eq $tag)) - ($entry-{TREE} eq $tree)) { - push(@{$log}, $entry); -} - } - return $log; -} - - -# return an array of logfile entries given a cvs log file. -# Only return entries newer than $days old -sub cvs_parse() -{ -my $file = shift; -my $days = shift; -my $tree = shift; -my $tag = shift; -my $log; -my $entry; - -open(FILE, $file); -while (FILE) { - my $line = $_; - - # ignore separator lines - if ($line =~ /^[*]+/) { next; } - - if ($line =~ /^Date:\s*(.*)/) { - if ($entry-{DATE}) { - $log = push_entry($entry, $log, $days, $tree, $tag); - } - $entry = {}; - $entry-{DATE} = cvs_parse_date($1); - $entry-{DIR} = ; - next; - } - - if ($line =~ /^Author:\s*(.*)/) { - $entry-{AUTHOR} = $1; - next; - } - - if ($line =~ /^Update of (.*)/) { - $entry-{TREE} = $1; - if ($entry-{TREE} =~ /\/(data|home)\/cvs\/([^\/]*)\/?(.*)/) { - $entry-{TREE} = $2; - $entry-{DIR} = $3; - if ($entry-{DIR}) { $entry-{DIR} .= /; } - } elsif ($entry-{TREE} =~ /\/home\/tridge\/cvstest\/([^\/]*)\/?(.*)/) { - $entry-{TREE} = $1; - $entry-{DIR} = $2; - if ($entry-{DIR})
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated via ec50db6 s4-smbtorture: use test_GetPrinterData{Ex}_checktype calls in RPC-SPOOLSS tests. via 1c57931 s3-spoolss: fix spoolss GetPrinterData behaviour via fa0b44e s4-smbtorture: allow to test returned type after specific error code in PrinterData tests. from fa203f4 s3/time_audit: fix a change that was just for debuggin purposeÑ http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit ec50db61fa6f3cf026c1f0f40528587b3dec8da3 Author: Günther Deschner g...@samba.org Date: Wed Nov 10 15:21:15 2010 +0100 s4-smbtorture: use test_GetPrinterData{Ex}_checktype calls in RPC-SPOOLSS tests. Guenther Autobuild-User: Günther Deschner g...@samba.org Autobuild-Date: Wed Nov 10 15:07:12 UTC 2010 on sn-devel-104 commit 1c579318ae2d2480ee4cc998443c0d1661b39846 Author: David Disseldorp dd...@suse.de Date: Fri Nov 5 15:24:22 2010 +0100 s3-spoolss: fix spoolss GetPrinterData behaviour Windows sends spoolss GetPrinterData requests with an offered buffer size of zero, Model and TrayFormTable data is commonly requested in this way. Samba's GetPrinterData response for the above case includes the correct error code (WERR_MORE_DATA), however the type field is set to REG_NONE. This causes Windows (seen on XP and 2k3) to give up on the request. If the type field is retained (not set to REG_NONE) when responding with WERR_MORE_DATA, Windows reissues the GetPrinterData request with an increased offered buffer size. Signed-off-by: Günther Deschner g...@samba.org commit fa0b44ed9c31baf19c70aa47a25233c947c0ae05 Author: Günther Deschner g...@samba.org Date: Wed Nov 10 10:19:01 2010 +0100 s4-smbtorture: allow to test returned type after specific error code in PrinterData tests. Guenther --- Summary of changes: source3/rpc_server/srv_spoolss_nt.c |2 +- source4/torture/rpc/spoolss.c | 69 ++- 2 files changed, 52 insertions(+), 19 deletions(-) Changeset truncated at 500 lines: diff --git a/source3/rpc_server/srv_spoolss_nt.c b/source3/rpc_server/srv_spoolss_nt.c index aafba01..c0f953b 100644 --- a/source3/rpc_server/srv_spoolss_nt.c +++ b/source3/rpc_server/srv_spoolss_nt.c @@ -9222,7 +9222,7 @@ WERROR _spoolss_GetPrinterDataEx(struct pipes_struct *p, } done: - *r-out.type= SPOOLSS_BUFFER_OK(*r-out.type, REG_NONE); + /* retain type when returning WERR_MORE_DATA */ r-out.data = SPOOLSS_BUFFER_OK(r-out.data, r-out.data); return SPOOLSS_BUFFER_OK(WERR_OK, WERR_MORE_DATA); diff --git a/source4/torture/rpc/spoolss.c b/source4/torture/rpc/spoolss.c index 8154cd2..6530a85 100644 --- a/source4/torture/rpc/spoolss.c +++ b/source4/torture/rpc/spoolss.c @@ -3474,13 +3474,14 @@ static bool test_ResumePrinter(struct torture_context *tctx, return true; } -static bool test_GetPrinterData(struct torture_context *tctx, - struct dcerpc_binding_handle *b, - struct policy_handle *handle, - const char *value_name, - enum winreg_Type *type_p, - uint8_t **data_p, - uint32_t *needed_p) +static bool test_GetPrinterData_checktype(struct torture_context *tctx, + struct dcerpc_binding_handle *b, + struct policy_handle *handle, + const char *value_name, + enum winreg_Type *expected_type, + enum winreg_Type *type_p, + uint8_t **data_p, + uint32_t *needed_p) { NTSTATUS status; struct spoolss_GetPrinterData r; @@ -3501,6 +3502,9 @@ static bool test_GetPrinterData(struct torture_context *tctx, torture_assert_ntstatus_ok(tctx, status, GetPrinterData failed); if (W_ERROR_EQUAL(r.out.result, WERR_MORE_DATA)) { + if (expected_type) { + torture_assert_int_equal(tctx, type, *expected_type, unexpected type); + } r.in.offered = needed; r.out.data = talloc_zero_array(tctx, uint8_t, r.in.offered); status = dcerpc_spoolss_GetPrinterData_r(b, tctx, r); @@ -3527,14 +3531,27 @@ static bool test_GetPrinterData(struct torture_context *tctx, return true; } -static bool test_GetPrinterDataEx(struct torture_context *tctx, - struct dcerpc_pipe *p, - struct policy_handle *handle, -
[SCM] build.samba.org - branch master updated
The branch, master has been updated via 94ae43a Serve in standalone mode only files that are local the current folder from 4db119b We no longer have CVS-based projects, and should not introduce any newer; remove cvslog.pl. http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit 94ae43ae63388a10385e4d07f66cd8f15a36d38a Author: Matthieu Patou m...@matws.net Date: Thu Nov 11 01:00:40 2010 +0300 Serve in standalone mode only files that are local the current folder This should avoid security problem --- Summary of changes: web/build.py | 25 + 1 files changed, 13 insertions(+), 12 deletions(-) Changeset truncated at 500 lines: diff --git a/web/build.py b/web/build.py index 67546a9..5c342ea 100755 --- a/web/build.py +++ b/web/build.py @@ -912,18 +912,19 @@ def buildApp(environ, start_response): if standalone and environ['PATH_INFO']: dir = os.path.join(os.path.dirname(__file__)) -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 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')]) -- build.samba.org
[SCM] build.samba.org - branch master updated
The branch, master has been updated via a6951b8 desactivate history for the moment (it's broken) via cc86506 use buildfarm get_build function instead of the one of the buildstore via 4e12cf0 implement get_build on the buildfarm object so that it can dispatch to the adhoc object via bb1a424 use buildstore function to get the cached status via 1c09aa4 add function to get the lcov cache status on a build store from 94ae43a Serve in standalone mode only files that are local the current folder http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit a6951b866d22bedef2450d25fdedf67a56457071 Author: Matthieu Patou m...@matws.net Date: Thu Nov 11 02:42:46 2010 +0300 desactivate history for the moment (it's broken) commit cc865062d99b19523c564a90383ae6f94323153c Author: Matthieu Patou m...@matws.net Date: Thu Nov 11 02:42:24 2010 +0300 use buildfarm get_build function instead of the one of the buildstore commit 4e12cf023151dc138ccd007de85846fecf92943e Author: Matthieu Patou m...@matws.net Date: Thu Nov 11 02:39:16 2010 +0300 implement get_build on the buildfarm object so that it can dispatch to the adhoc object commit bb1a424d137f7d153dfd5c1b278c22e1f2593634 Author: Matthieu Patou m...@matws.net Date: Thu Nov 11 02:37:48 2010 +0300 use buildstore function to get the cached status commit 1c09aa46acf31df23417076e1ca1be3e28ddbfb6 Author: Matthieu Patou m...@matws.net Date: Thu Nov 11 02:41:16 2010 +0300 add function to get the lcov cache status on a build store --- Summary of changes: buildfarm/__init__.py | 21 +++-- buildfarm/data.py |5 + web/build.py | 10 +- 3 files changed, 29 insertions(+), 7 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/__init__.py b/buildfarm/__init__.py index 2370940..118a729 100644 --- a/buildfarm/__init__.py +++ b/buildfarm/__init__.py @@ -99,8 +99,7 @@ class BuildFarm(object): def lcov_status(self, tree): get status of build from buildfarm import data, util -cachefile = os.path.join(self.builds.cachedir, lcov.%s.%s.status % ( -self.LCOVHOST, tree)) +cachefile = self.builds.get_lcov_cached_status(self.LCOVHOST, tree) file = os.path.join(self.lcovdir, self.LCOVHOST, tree, index.html) try: st1 = os.stat(file) @@ -129,6 +128,24 @@ class BuildFarm(object): util.FileSave(cachefile, ret) return perc +def get_build(self, tree, host, compiler, rev=None): +if rev: +if host in self.hostdb.hosts() and\ +tree in self.trees and\ +compiler in self.compilers: +return self.builds.get_build(tree, host, compiler) +else: +from buildfarm import data +raise data.NoSuchBuildError(tree, host, compiler) +else: +if host in [h.name for h in self.hostdb.hosts()] and\ +tree in self.trees and\ +compiler in self.compilers: +return self.upload_builds.get_build(tree, host, compiler) +else: +from buildfarm import data +raise data.NoSuchBuildError(tree, host, compiler) + def get_new_builds(self): from buildfarm import data for host in self.hostdb.hosts(): diff --git a/buildfarm/data.py b/buildfarm/data.py index 493badb..f709422 100644 --- a/buildfarm/data.py +++ b/buildfarm/data.py @@ -396,6 +396,9 @@ class BuildResultStore(object): self.path = path +def get_lcov_cached_status(self, host, tree): +return None + def get_build(self, tree, host, compiler, rev): logf = self.build_fname(tree, host, compiler, rev) + .log if not os.path.exists(logf): @@ -477,5 +480,7 @@ class CachingBuildResultStore(BuildResultStore): def cache_fname(self, tree, host, compiler, rev): return os.path.join(self.cachedir, build.%s.%s.%s-%s % (tree, host, compiler, rev)) +def get_lcov_cached_status(self, host, tree): +return os.path.join(self.cachedir, lcov.%s.%s.status % (host, tree)) diff --git a/web/build.py b/web/build.py index 5c342ea..6d06c3d 100755 --- a/web/build.py +++ b/web/build.py @@ -48,7 +48,7 @@ basedir = os.path.abspath(os.path.join(webdir, ..)) buildfarm = BuildFarm() db = data.BuildResultStore(basedir) -history = history.History(db) +#history = history.History(db) hostsdb = buildfarm.hostdb compilers = buildfarm.compilers @@ -173,7 +173,7 @@ def view_summary(myself, output_type): for compiler in compilers: for tree in trees: try: -build = db.get_build(tree, host.name.encode(utf-8), compiler) +
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated via c4c230b autobuild: perfer to use git mail address in autobuild comment from ec50db6 s4-smbtorture: use test_GetPrinterData{Ex}_checktype calls in RPC-SPOOLSS tests. http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit c4c230bd0fb25a0433e36c4e4a315268ff44fbe0 Author: Björn JACKE b...@sernet.de Date: Thu Nov 11 10:12:07 2010 +1100 autobuild: perfer to use git mail address in autobuild comment Signed-off-by: Andrew Tridgell tri...@samba.org Autobuild-User: Andrew Tridgell tri...@samba.org Autobuild-Date: Wed Nov 10 23:56:37 UTC 2010 on sn-devel-104 --- Summary of changes: script/commit_mark.sh |6 +- 1 files changed, 5 insertions(+), 1 deletions(-) Changeset truncated at 500 lines: diff --git a/script/commit_mark.sh b/script/commit_mark.sh index 93a345b..8495808 100755 --- a/script/commit_mark.sh +++ b/script/commit_mark.sh @@ -7,9 +7,13 @@ if grep -q '^Autobuild.User' $1; then fi fullname=$(getent passwd $USER | cut -d: -f5| cut -d',' -f1) +mailaddr=$(git config user.email) +if test -z $mailaddr ; then + mailaddr=$u...@samba.org +fi cat EOF $1 -Autobuild-User: $fullname $u...@samba.org +Autobuild-User: $fullname $mailaddr Autobuild-Date: $(date) on $(hostname) EOF exit 0 -- Samba Shared Repository
[SCM] build.samba.org - branch master updated
The branch, master has been updated via e7f5f8b Move lcov cache onto BuildFarm, simplify get_build. from a6951b8 desactivate history for the moment (it's broken) http://gitweb.samba.org/?p=build-farm.git;a=shortlog;h=master - Log - commit e7f5f8bd5485ef4b1a742b6b64aa0cb49008ce01 Author: Jelmer Vernooij jel...@samba.org Date: Thu Nov 11 01:50:54 2010 +0100 Move lcov cache onto BuildFarm, simplify get_build. --- Summary of changes: buildfarm/__init__.py | 20 +--- buildfarm/data.py |8 2 files changed, 5 insertions(+), 23 deletions(-) Changeset truncated at 500 lines: diff --git a/buildfarm/__init__.py b/buildfarm/__init__.py index 118a729..85690e3 100644 --- a/buildfarm/__init__.py +++ b/buildfarm/__init__.py @@ -66,6 +66,7 @@ class BuildFarm(object): if path is None: path = os.path.abspath(os.path.join(os.path.dirname(__file__), ..)) self.path = path +self.cachedir = os.path.join(self.path, cache) self.webdir = os.path.join(self.path, web) if not os.path.isdir(path): raise Exception(web directory %s does not exist % self.webdir) @@ -99,7 +100,8 @@ class BuildFarm(object): def lcov_status(self, tree): get status of build from buildfarm import data, util -cachefile = self.builds.get_lcov_cached_status(self.LCOVHOST, tree) +cachefile = os.path.join(self.cachedir, lcov.%s.%s.status % ( +self.LCOVHOST, tree)) file = os.path.join(self.lcovdir, self.LCOVHOST, tree, index.html) try: st1 = os.stat(file) @@ -130,21 +132,9 @@ class BuildFarm(object): def get_build(self, tree, host, compiler, rev=None): if rev: -if host in self.hostdb.hosts() and\ -tree in self.trees and\ -compiler in self.compilers: -return self.builds.get_build(tree, host, compiler) -else: -from buildfarm import data -raise data.NoSuchBuildError(tree, host, compiler) +return self.builds.get_build(tree, host, compiler, rev) else: -if host in [h.name for h in self.hostdb.hosts()] and\ -tree in self.trees and\ -compiler in self.compilers: -return self.upload_builds.get_build(tree, host, compiler) -else: -from buildfarm import data -raise data.NoSuchBuildError(tree, host, compiler) +return self.upload_builds.get_build(tree, host, compiler) def get_new_builds(self): from buildfarm import data diff --git a/buildfarm/data.py b/buildfarm/data.py index f709422..9f7708c 100644 --- a/buildfarm/data.py +++ b/buildfarm/data.py @@ -396,9 +396,6 @@ class BuildResultStore(object): self.path = path -def get_lcov_cached_status(self, host, tree): -return None - def get_build(self, tree, host, compiler, rev): logf = self.build_fname(tree, host, compiler, rev) + .log if not os.path.exists(logf): @@ -479,8 +476,3 @@ class CachingBuildResultStore(BuildResultStore): def cache_fname(self, tree, host, compiler, rev): return os.path.join(self.cachedir, build.%s.%s.%s-%s % (tree, host, compiler, rev)) - -def get_lcov_cached_status(self, host, tree): -return os.path.join(self.cachedir, lcov.%s.%s.status % (host, tree)) - - -- build.samba.org
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated via 0afb299 Fix bug #7791 - gvfsd-smb (Gnome vfs) fails to copy files from a SMB share using SMB signing. via ac0ffba Add test that detects problems in the SMB signing code when run through cli_write() (which doesn't use the new async methods). from c4c230b autobuild: perfer to use git mail address in autobuild comment http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit 0afb2995a2177afa2eb7b8f99887a39cdaf23a15 Author: Jeremy Allison j...@samba.org Date: Wed Nov 10 17:59:05 2010 -0800 Fix bug #7791 - gvfsd-smb (Gnome vfs) fails to copy files from a SMB share using SMB signing. The underlying problem is that the old code invoked by cli_write() increments cli-mid directly when issuing outstanding writes. This should now be done only in libsmb/clientgen.c to make metze's new signing engine works correctly. Just deleting this code fixes the problem. Jeremy. Autobuild-User: Jeremy Allison j...@samba.org Autobuild-Date: Thu Nov 11 02:50:08 UTC 2010 on sn-devel-104 commit ac0ffba004160693ea5829775d732146ef261c47 Author: Jeremy Allison j...@samba.org Date: Wed Nov 10 17:54:57 2010 -0800 Add test that detects problems in the SMB signing code when run through cli_write() (which doesn't use the new async methods). Unfortunately, and I think due to the socket wrapper code, this doesn't detect the failure on the build farm, but running the RW-SIGNING test separately against a Samba or Windows server using signing does. Jeremy. --- Summary of changes: selftest/target/Samba3.pm |2 + source3/libsmb/clireadwrite.c |7 +--- source3/script/tests/test_smbtorture_s3.sh |2 +- source3/torture/torture.c | 46 +--- 4 files changed, 40 insertions(+), 17 deletions(-) Changeset truncated at 500 lines: diff --git a/selftest/target/Samba3.pm b/selftest/target/Samba3.pm index f1387d9..1326949 100644 --- a/selftest/target/Samba3.pm +++ b/selftest/target/Samba3.pm @@ -525,6 +525,8 @@ sub provision($$) max protocol = SMB2 read only = no + server signing = auto + smbd:sharedelay = 10 # smbd:writetimeupdatedelay = 50 map hidden = no diff --git a/source3/libsmb/clireadwrite.c b/source3/libsmb/clireadwrite.c index 370c068..cd95b17 100644 --- a/source3/libsmb/clireadwrite.c +++ b/source3/libsmb/clireadwrite.c @@ -703,8 +703,7 @@ static bool cli_issue_write(struct cli_state *cli, off_t offset, uint16 mode, const char *buf, - size_t size, - int i) + size_t size) { char *p; bool large_writex = false; @@ -778,8 +777,6 @@ static bool cli_issue_write(struct cli_state *cli, cli_setup_bcc(cli, p+size); } - SSVAL(cli-outbuf,smb_mid,cli-mid + i); - show_msg(cli-outbuf); if (direct_writes) { /* For direct writes we now need to write the data @@ -828,7 +825,7 @@ ssize_t cli_write(struct cli_state *cli, if (!cli_issue_write(cli, fnum, offset + bsent, write_mode, buf + bsent, - size1, issued)) + size1)) return -1; issued++; } diff --git a/source3/script/tests/test_smbtorture_s3.sh b/source3/script/tests/test_smbtorture_s3.sh index 1abf7b6..4680f64 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 DIR-CREATETIME TCON TCONDEV RW1 RW2 RW3 +tests=$tests DIR DIR1 DIR-CREATETIME TCON TCONDEV RW1 RW2 RW3 RW-SIGNING 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 161c6f9..f69d734 100644 --- a/source3/torture/torture.c +++ b/source3/torture/torture.c @@ -52,6 +52,7 @@ static fstring multishare_conn_fname; static bool use_multishare_conn = False; static bool do_encrypt; static const char *local_path = NULL; +static int signing_state = Undefined; bool torture_showall = False; @@ -178,7
[SCM] Samba Shared Repository - branch master updated
The branch, master has been updated via ee50bdd s4-loadparm: fix the FLAG_DEFAULT settings on specially handled parameters via 21fe96e s4-provision: include command line provision options in the generated smb.conf via 60cb372 s4-devel: useful script for testing join with empty smb.conf via e908c26 s4-test: use IPC$ instead of ADMIN$ for srvsvc NetShareGetInfo test via 81c9b98 s4-join: use the command line loadparm in provision during a join via 1d7d09a s4-loadparm: fixed dumping of non-default parms with testparm via ed8ea4e s4-server: move the creation of the IPC$ share into ntvfs from 0afb299 Fix bug #7791 - gvfsd-smb (Gnome vfs) fails to copy files from a SMB share using SMB signing. http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master - Log - commit ee50bdd42f01bfcebb2e75a923a01fa1d18e5694 Author: Andrew Tridgell tri...@samba.org Date: Thu Nov 11 13:56:44 2010 +1100 s4-loadparm: fix the FLAG_DEFAULT settings on specially handled parameters parameters with special handlers also need to clear the FLAG_DEFAULT flag when set Pair-Programmed-With: Andrew Bartlett abart...@samba.org Autobuild-User: Andrew Tridgell tri...@samba.org Autobuild-Date: Thu Nov 11 03:39:23 UTC 2010 on sn-devel-104 commit 21fe96e8f9e1f3f3d047ed893633228c2aef881c Author: Andrew Tridgell tri...@samba.org Date: Thu Nov 11 13:09:29 2010 +1100 s4-provision: include command line provision options in the generated smb.conf this saves the smb.conf using lp.dump_globals() to ensure that any command line options (for example directory overrides) are saved in the generated smb.conf Pair-Programmed-With: Andrew Bartlett abart...@samba.org commit 60cb372bd6c874af1050ff3d723bacee8c9cbf8e Author: Andrew Tridgell tri...@samba.org Date: Thu Nov 11 11:18:48 2010 +1100 s4-devel: useful script for testing join with empty smb.conf this sets up all the needed --options for a join with a prefix other than the one compiled in commit e908c263916c1e18468054d8a0416e5a1425c25d Author: Andrew Tridgell tri...@samba.org Date: Thu Nov 11 11:16:23 2010 +1100 s4-test: use IPC$ instead of ADMIN$ for srvsvc NetShareGetInfo test IPC$ is guaranteed to exist on servers that do RPC, whereas ADMIN$ isn't Pair-Programmed-With: Andrew Bartlett abart...@samba.org commit 81c9b98ecd6b0b4ec412ea7a0963002dc17f4f66 Author: Andrew Tridgell tri...@samba.org Date: Thu Nov 11 10:45:13 2010 +1100 s4-join: use the command line loadparm in provision during a join this allows a join with an empty smb.conf to override locations of files correctly with --option Pair-Programmed-With: Andrew Bartlett abart...@samba.org commit 1d7d09a8f72a9f7f888538b7f7c63969407e9c2d Author: Andrew Tridgell tri...@samba.org Date: Thu Nov 11 10:39:19 2010 +1100 s4-loadparm: fixed dumping of non-default parms with testparm when using testparm without -v we should only dump non-default parameters. This patch fixes up the handling of the FLAG_DEFAULT flag in loadparm to correctly mark parameters as default or not, including parametric options Pair-Programmed-With: Andrew Bartlett abart...@samba.org commit ed8ea4ed18f08bfc93190513f2a2e8e058655daf Author: Andrew Tridgell tri...@samba.org Date: Thu Nov 11 10:35:38 2010 +1100 s4-server: move the creation of the IPC$ share into ntvfs the IPC$ share is only used by the ntvfs backends, and doesn't need to be created on every load of smb.conf. This fixes a problem with testparm showing the ipc$ share when it isn't defined in smb.conf. This also removes the admin$ share, which really shouldn't be on by default. The admin$ share is used for remote software installation, and normally exposes the c:\windows directory on a windows server. That makes no sense on Samba. If for some reason a admin$ share is needed, then the admin can create one as usual. Exposing /tmp via admin$ by default seems like a bad idea. Pair-Programmed-With: Andrew Bartlett abart...@samba.org --- Summary of changes: source4/ntvfs/ntvfs_base.c | 26 source4/param/loadparm.c| 93 --- source4/scripting/devel/config_base | 43 source4/scripting/python/samba/join.py |2 +- source4/scripting/python/samba/provision.py | 43 source4/torture/rpc/srvsvc.c|6 +- 6 files changed, 143 insertions(+), 70 deletions(-) create mode 100755 source4/scripting/devel/config_base Changeset truncated at 500 lines: diff --git a/source4/ntvfs/ntvfs_base.c b/source4/ntvfs/ntvfs_base.c index d8d73e9..1b1a533 100644 ---
[SCM] Samba Shared Repository - branch v3-6-test updated
The branch, v3-6-test has been updated via 254d3da Fix bug #7791 - gvfsd-smb (Gnome vfs) fails to copy files from a SMB share using SMB signing. from 2791ecb Fix memleak I accidently introduced when reading from tdb. http://gitweb.samba.org/?p=samba.git;a=shortlog;h=v3-6-test - Log - commit 254d3da934086e0bbfb74f23167eefbcc0759e1a Author: Jeremy Allison j...@samba.org Date: Wed Nov 10 17:59:05 2010 -0800 Fix bug #7791 - gvfsd-smb (Gnome vfs) fails to copy files from a SMB share using SMB signing. The underlying problem is that the old code invoked by cli_write() increments cli-mid directly when issuing outstanding writes. This should now be done only in libsmb/clientgen.c to make metze's new signing engine works correctly. Just deleting this code fixes the problem. Jeremy. Autobuild-User: Jeremy Allison j...@samba.org Autobuild-Date: Thu Nov 11 02:50:08 UTC 2010 on sn-devel-104 (cherry picked from commit 0afb2995a2177afa2eb7b8f99887a39cdaf23a15) --- Summary of changes: source3/libsmb/clireadwrite.c |7 ++- 1 files changed, 2 insertions(+), 5 deletions(-) Changeset truncated at 500 lines: diff --git a/source3/libsmb/clireadwrite.c b/source3/libsmb/clireadwrite.c index 24c9d9f..81fb88c 100644 --- a/source3/libsmb/clireadwrite.c +++ b/source3/libsmb/clireadwrite.c @@ -702,8 +702,7 @@ static bool cli_issue_write(struct cli_state *cli, off_t offset, uint16 mode, const char *buf, - size_t size, - int i) + size_t size) { char *p; bool large_writex = false; @@ -777,8 +776,6 @@ static bool cli_issue_write(struct cli_state *cli, cli_setup_bcc(cli, p+size); } - SSVAL(cli-outbuf,smb_mid,cli-mid + i); - show_msg(cli-outbuf); if (direct_writes) { /* For direct writes we now need to write the data @@ -827,7 +824,7 @@ ssize_t cli_write(struct cli_state *cli, if (!cli_issue_write(cli, fnum, offset + bsent, write_mode, buf + bsent, - size1, issued)) + size1)) return -1; issued++; } -- Samba Shared Repository
[SCM] Samba Shared Repository - branch v3-6-test updated
The branch, v3-6-test has been updated via cc2d13f Add test that detects problems in the SMB signing code when run through cli_write() (which doesn't use the new async methods). from 254d3da Fix bug #7791 - gvfsd-smb (Gnome vfs) fails to copy files from a SMB share using SMB signing. http://gitweb.samba.org/?p=samba.git;a=shortlog;h=v3-6-test - Log - commit cc2d13f7f3cfdcc1ed710612ebaee985e81d9a25 Author: Jeremy Allison j...@samba.org Date: Wed Nov 10 17:54:57 2010 -0800 Add test that detects problems in the SMB signing code when run through cli_write() (which doesn't use the new async methods). Unfortunately, and I think due to the socket wrapper code, this doesn't detect the failure on the build farm, but running the RW-SIGNING test separately against a Samba or Windows server using signing does. Jeremy. (cherry picked from commit ac0ffba004160693ea5829775d732146ef261c47) --- Summary of changes: selftest/target/Samba3.pm |2 + source3/script/tests/test_smbtorture_s3.sh |2 +- source3/torture/torture.c | 46 +--- 3 files changed, 38 insertions(+), 12 deletions(-) Changeset truncated at 500 lines: diff --git a/selftest/target/Samba3.pm b/selftest/target/Samba3.pm index 9eb3ccd..9620ed8 100644 --- a/selftest/target/Samba3.pm +++ b/selftest/target/Samba3.pm @@ -528,6 +528,8 @@ sub provision($$) # min receivefile size = 4000 read only = no + server signing = auto + smbd:sharedelay = 10 # smbd:writetimeupdatedelay = 50 map hidden = no diff --git a/source3/script/tests/test_smbtorture_s3.sh b/source3/script/tests/test_smbtorture_s3.sh index 1abf7b6..4680f64 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 DIR-CREATETIME TCON TCONDEV RW1 RW2 RW3 +tests=$tests DIR DIR1 DIR-CREATETIME TCON TCONDEV RW1 RW2 RW3 RW-SIGNING tests=$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 6213457..340d6eb 100644 --- a/source3/torture/torture.c +++ b/source3/torture/torture.c @@ -49,6 +49,7 @@ static fstring multishare_conn_fname; static bool use_multishare_conn = False; static bool do_encrypt; static const char *local_path = NULL; +static int signing_state = Undefined; bool torture_showall = False; @@ -175,7 +176,7 @@ static struct cli_state *open_nbt_connection(void) zero_sockaddr(ss); - if (!(c = cli_initialise())) { + if (!(c = cli_initialise_ex(signing_state))) { printf(Failed initialize cli_struct to connect with %s\n, host); return NULL; } @@ -308,7 +309,7 @@ static struct cli_state *open_bad_nbt_connection(void) zero_sockaddr(ss); - if (!(c = cli_initialise())) { + if (!(c = cli_initialise_ex(signing_state))) { printf(Failed initialize cli_struct to connect with %s\n, host); return NULL; } @@ -409,7 +410,7 @@ static bool torture_open_connection_share(struct cli_state **c, hostname, NULL, port_to_use, sharename, ?, username, workgroup, -password, flags, Undefined, retry); +password, flags, signing_state, retry); if (!NT_STATUS_IS_OK(status)) { printf(failed to open share connection: //%s/%s port:%d - %s\n, hostname, sharename, port_to_use, nt_errstr(status)); @@ -884,7 +885,7 @@ static bool run_readwritemulti(int dummy) return test; } -static bool run_readwritelarge(int dummy) +static bool run_readwritelarge_internal(int max_xmit_k) { static struct cli_state *cli1; uint16_t fnum1; @@ -899,9 +900,17 @@ static bool run_readwritelarge(int dummy) cli_sockopt(cli1, sockops); memset(buf,'\0',sizeof(buf)); - cli1-max_xmit = 128*1024; + cli1-max_xmit = max_xmit_k*1024; + + if (signing_state == Required) { + /* Horrible cheat to force + multiple signed outstanding + packets against a Samba server. + */ + cli1-is_samba = false; + } - printf(starting readwritelarge\n); +
[SCM] Samba Shared Repository - branch v3-6-test updated
The branch, v3-6-test has been updated via bb61529 s3/vfs_time_audit: use monotonic clock for time deltas from cc2d13f Add test that detects problems in the SMB signing code when run through cli_write() (which doesn't use the new async methods). http://gitweb.samba.org/?p=samba.git;a=shortlog;h=v3-6-test - Log - commit bb61529e58e44d62da44fec35cdb16c1feeb696d Author: Björn Jacke b...@sernet.de Date: Wed Nov 10 00:40:29 2010 +0100 s3/vfs_time_audit: use monotonic clock for time deltas --- Summary of changes: source3/modules/vfs_time_audit.c | 777 ++ 1 files changed, 444 insertions(+), 333 deletions(-) Changeset truncated at 500 lines: diff --git a/source3/modules/vfs_time_audit.c b/source3/modules/vfs_time_audit.c index b510edb..84e4175 100644 --- a/source3/modules/vfs_time_audit.c +++ b/source3/modules/vfs_time_audit.c @@ -44,16 +44,17 @@ static int smb_time_audit_connect(vfs_handle_struct *handle, const char *svc, const char *user) { int result; - struct timeval tv; + struct timespec ts1,ts2; double timediff; if (!handle) { return -1; } - GetTimeOfDay(tv); + clock_gettime_mono(ts1); result = SMB_VFS_NEXT_CONNECT(handle, svc, user); - timediff = timeval_elapsed(tv); + clock_gettime_mono(ts2); + timediff = nsec_time_diff(ts2,ts1)*1.0e-9; if (timediff audit_timeout) { smb_time_audit_log(connect, timediff); } @@ -62,12 +63,13 @@ static int smb_time_audit_connect(vfs_handle_struct *handle, static void smb_time_audit_disconnect(vfs_handle_struct *handle) { - struct timeval tv; + struct timespec ts1,ts2; double timediff; - GetTimeOfDay(tv); + clock_gettime_mono(ts1); SMB_VFS_NEXT_DISCONNECT(handle); - timediff = timeval_elapsed(tv); + clock_gettime_mono(ts2); + timediff = nsec_time_diff(ts2,ts1)*1.0e-9; if (timediff audit_timeout) { smb_time_audit_log(disconnect, timediff); @@ -82,13 +84,14 @@ static uint64_t smb_time_audit_disk_free(vfs_handle_struct *handle, uint64_t *dfree, uint64_t *dsize) { uint64_t result; - struct timeval tv; + struct timespec ts1,ts2; double timediff; - GetTimeOfDay(tv); + clock_gettime_mono(ts1); result = SMB_VFS_NEXT_DISK_FREE(handle, path, small_query, bsize, dfree, dsize); - timediff = timeval_elapsed(tv); + clock_gettime_mono(ts2); + timediff = nsec_time_diff(ts2,ts1)*1.0e-9; /* Don't have a reasonable notion of failure here */ if (timediff audit_timeout) { @@ -103,12 +106,13 @@ static int smb_time_audit_get_quota(struct vfs_handle_struct *handle, SMB_DISK_QUOTA *qt) { int result; - struct timeval tv; + struct timespec ts1,ts2; double timediff; - GetTimeOfDay(tv); + clock_gettime_mono(ts1); result = SMB_VFS_NEXT_GET_QUOTA(handle, qtype, id, qt); - timediff = timeval_elapsed(tv); + clock_gettime_mono(ts2); + timediff = nsec_time_diff(ts2,ts1)*1.0e-9; if (timediff audit_timeout) { smb_time_audit_log(get_quota, timediff); @@ -121,12 +125,13 @@ static int smb_time_audit_set_quota(struct vfs_handle_struct *handle, SMB_DISK_QUOTA *qt) { int result; - struct timeval tv; + struct timespec ts1,ts2; double timediff; - GetTimeOfDay(tv); + clock_gettime_mono(ts1); result = SMB_VFS_NEXT_SET_QUOTA(handle, qtype, id, qt); - timediff = timeval_elapsed(tv); + clock_gettime_mono(ts2); + timediff = nsec_time_diff(ts2,ts1)*1.0e-9; if (timediff audit_timeout) { smb_time_audit_log(set_quota, timediff); @@ -141,13 +146,14 @@ static int smb_time_audit_get_shadow_copy_data(struct vfs_handle_struct *handle, bool labels) { int result; - struct timeval tv; + struct timespec ts1,ts2; double timediff; - GetTimeOfDay(tv); + clock_gettime_mono(ts1); result = SMB_VFS_NEXT_GET_SHADOW_COPY_DATA(handle, fsp, shadow_copy_data, labels); - timediff = timeval_elapsed(tv); + clock_gettime_mono(ts2); + timediff = nsec_time_diff(ts2,ts1)*1.0e-9; if (timediff audit_timeout) { smb_time_audit_log(get_shadow_copy_data, timediff); @@ -161,12 +167,13 @@ static int smb_time_audit_statvfs(struct vfs_handle_struct *handle,
Build status as of Thu Nov 11 07:00:02 2010
URL: http://build.samba.org/ --- /home/build/master/cache/broken_results.txt.old 2010-11-10 00:00:06.0 -0700 +++ /home/build/master/cache/broken_results.txt 2010-11-11 00:00:03.0 -0700 @@ -1,9 +1,9 @@ -Build status as of Wed Nov 10 07:00:02 2010 +Build status as of Thu Nov 11 07:00:02 2010 Build counts: Tree Total Broken Panic build_farm 0 0 0 -ccache 37 6 0 +ccache 37 5 0 ccache-maint 36 4 0 ldb 37 26 0 libreplace 37 37 0 @@ -14,9 +14,9 @@ samba-docs 0 0 0 samba-web0 0 0 samba_3_current 37 33 1 -samba_3_master 37 19 0 +samba_3_master 37 20 0 samba_3_next 37 35 0 -samba_4_0_test 42 32 1 +samba_4_0_test 42 33 2 talloc 37 16 0 tdb 35 13 0