New version available, please upgrade...
From the samba lists it seems that there are quite a few serious bugs introduced in this release so although I would like to test it I think it would be better to wait till the upstream fixes these issues with 3.0.25a.
samba 3.0.25a is out from samba-announce-list on 2007-05-25 Release Announcements ===================== This is the second production release of the Samba 3.0.25 code base and is the version that servers should be run for for all current bug fixes. Major bug fixes included in Samba 3.0.25a are: o Missing supplementary Unix group membership when using "force group". o Premature expiration of domain user passwords when using a Samba domain controller. o Failure to open the Windows object picker against a server configured to use "security = domain". * Authentication failures when using security = server. Changes to MS-DFS Root Share Behavior ===================================== Please be aware that the initial value for the "msdfs root" share parameter was changed in the 3.0.25 release series and that this option is now disabled by default. Windows clients frequently require a reboot in order to clear any cached information about MS-DFS root shares on a server and you may experience failures accessing file services on Samba 3.0.25 servers until the client reboot is performed. Alternately, you may explicitly re-enable the parameter in smb.conf. Please refer to the smb.conf(5) man page for more details.
Created attachment 120696 [details] samba-3.0.25a.ebuild I needed a fix in this release so I took the time to bump it myself. The patches in "files" were no longer needed but the additional patches had to be fixed up. Another file to follow.
Created attachment 120697 [details] samba-3-gentoo-0.3.17.tar.bz2 Here is the patchset.
Created attachment 121451 [details] Error log from build with USE="oav" I'm getting an error when trying to compile samba-3.0.25a with the "oav" USE flag enabled. It compiled fine with USE="-oav", however. Don't know if this is due to a bug in the actual samba vscan code or a result of the samba-vscan-0.3.6b patch. Also, /etc/init.d/samba disappeared after the merge. -------------- emerge --info: -------------- Portage 2.1.2.9 (default-linux/amd64/2006.1/server, gcc-4.1.2, glibc-2.5-r3, 2.6.20-hardened-r2 x86_64) ================================================================= System uname: 2.6.20-hardened-r2 x86_64 AMD Athlon(tm) 64 Processor 3000+ Gentoo Base System release 1.12.9 Timestamp of tree: Thu, 07 Jun 2007 19:00:01 +0000 distcc 2.18.3 x86_64-pc-linux-gnu (protocols 1 and 2) (default port 3632) [disabled] ccache version 2.4 [enabled] dev-java/java-config: 1.3.7, 2.0.32 dev-lang/python: 2.4.4-r4 dev-python/pycrypto: 2.0.1-r5 dev-util/ccache: 2.4-r7 sys-apps/sandbox: 1.2.17 sys-devel/autoconf: 2.13, 2.61 sys-devel/automake: 1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2, 1.10 sys-devel/binutils: 2.16.1-r3 sys-devel/gcc-config: 1.3.16 sys-devel/libtool: 1.5.22 virtual/os-headers: 2.6.17-r2 ACCEPT_KEYWORDS="amd64" AUTOCLEAN="yes" CBUILD="x86_64-pc-linux-gnu" CFLAGS="-march=athlon64 -msse3 -O2 -pipe" CHOST="x86_64-pc-linux-gnu" CONFIG_PROTECT="/etc" CONFIG_PROTECT_MASK="/etc/env.d /etc/env.d/java/ /etc/gconf /etc/php/apache2-php5/ext-active/ /etc/php/cgi-php5/ext-active/ /etc/php/cli-php5/ext-active/ /etc/revdep-rebuild /etc/terminfo" CXXFLAGS="-march=athlon64 -msse3 -O2 -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="ccache distlocks metadata-transfer nodoc noinfo parallel-fetch sandbox sfperms strict userfetch" GENTOO_MIRRORS="http://gentoo.osuosl.org http://gentoo.netnitco.net" MAKEOPTS="-j2" PKGDIR="/usr/portage/packages" PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --delete-after --stats --timeout=180 --exclude=/distfiles --exclude=/local --exclude=/packages --filter=H_**/files/digest-*" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/usr/local/portage" SYNC="rsync://rsync.gentoo.org/gentoo-portage" USE="amd64 apache2 bash-completion bitmap-fonts bzip2 cli cracklib crypt dri fortran iconv isdnlog java5 jpeg jpeg2k libg++ midi mudflap ncurses nls nptl nptlonly openmp pam pcre perl ppds pppd python readline reflection session spl ssl tcpd truetype truetype-fonts type1-fonts unicode xml xorg zlib" ALSA_CARDS="ali5451 als4000 atiixp atiixp-modem bt87x ca0106 cmipci emu10k1x ens1370 ens1371 es1938 es1968 fm801 hda-intel intel8x0 intel8x0m maestro3 trident usb-audio via82xx via82xx-modem ymfpci" ALSA_PCM_PLUGINS="adpcm alaw asym copy dmix dshare dsnoop empty extplug file hooks iec958 ioplug ladspa lfloat linear meter mulaw multi null plug rate route share shm softvol" ELIBC="glibc" INPUT_DEVICES="keyboard mouse evdev" KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" USERLAND="GNU" VIDEO_CARDS="apm ark chips cirrus cyrix dummy fbdev glint i128 i810 mach64 mga neomagic nv r128 radeon rendition s3 s3virge savage siliconmotion sis sisusb tdfx tga trident tseng v4l vesa vga via vmware voodoo" Unset: CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LC_ALL, LDFLAGS, LINGUAS, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS
@Akshay Shah: samba-vscan (enabled with the oav USE-flag) is in general a seperate package (a vfs-module) which needs the samba source-tree to build and has therefore been integrated into Samba. The problem is that the VFS-interface has been changed for samba-3.0.25 and this breaks samba-vscan completely. Currently, I'm therefore working on two things: a) Patch samba-vscan to work with 3.0.25 b) Find out what exactly is needed from samba to build samba-vscan and other extensions. This is mostly done now and it seems that we can split-out samba-vscan from the samba-ebuild if I install some additional header-file. Furthermore, I have to check which patches aren't needed anymore (since the arm-cross-compile bug seems to be fixed in upstreams bug-tracker, but it wasn't in samba-3.0.25-rc3 and the responsible developer didn't answer). This might also involve some ebuild-rewriting since I rather prefer manual installation in the ebuild instead of package patching. If anyone wants to help with one of the above issues, mail me.
Hi folks, i dunno anything about the changelog, but version 3.0.25b is out (since June, 26.). Is there an ebuild update planned soon? Are there still any serious known regressions/bugs? Thanks in advance for an info update! Greetings ;)
Created attachment 123823 [details] samba-3-gentoo-0.3.18.tar.bz2 Needed for samba-3.0.25b
Created attachment 123824 [details] samba-3.0.25b.ebuild samba-3.0.25b.ebuild compiles fine on amd64. I have not tested it though.
Ok, after talking with the samba-guys, I'm going to drop samba-vscan. The problem is that upstream (for samba-vscan) doesn't release new versions and the current one is broken for samba-3.0.25 due to API-changes. Upstream recommends not to install the headers needed to build samba-vscan without the complete samba-source-tree, so I'll just going to drop it and hopefully be able to release samba-3.0.25b at the end of next week. So, the only task left is to find out which patches are still needed from the package and which can be replaced by some workarounds in the ebuild.
I just compiled samba-3.0.25b, but unfortunately the initscript was no longer in the '/etc/init.d/' directory (indeed, it even deleted the initscript from samba-3.0.24-r3. Was it left out intentionally? If so, how do I control the samba daemons?
Created attachment 124367 [details] /etc/init.d/samba
Sorry. it did that for me when I used the ebuild from James for version 3.0.25a. When i bumped the ebuild to 3.0.25b it did not happen possibly because I did not run dispatch-conf. Here is what I have for /etc/init.d/samba. I would normally check into this but the dev who handles samba has promised a release soon.
Thanks John. I grabbed one from a samba install on another computer, but I'll use yours anyway. I just wasn't sure if the lack of the initscript was intentional. Thanks for the clarification.
For anyone else who installs this before it hits portage: I checked to see why the init scripts were being clobbered and the solution was you need to copy /usr/portage/net-fs/samba/files/samba-init and /usr/portage/net-fs/samba/files/samba-conf to the files dir of your overlay. Before creating the digest and then proceed as normal with the install. http://gentoo-wiki.com/HOWTO_Installing_3rd_Party_Ebuilds
I dunno if this has been addressed already, but winbindd has a bug with the permissions on the socket file/directory (/var/cache/samba/winbindd_priviledged). It's created as 0755 (req. by winbindd) but I get a starup error that the permission level is too high (max 0750). I dunno if there is a patchfile or ot.
(In reply to comment #10) > Ok, after talking with the samba-guys, I'm going to drop samba-vscan. The > problem is that upstream (for samba-vscan) doesn't release new versions and the > current one is broken for samba-3.0.25 due to API-changes. Upstream recommends > not to install the headers needed to build samba-vscan without the complete > samba-source-tree, so I'll just going to drop it and hopefully be able to > release samba-3.0.25b at the end of next week. > > So, the only task left is to find out which patches are still needed from the > package and which can be replaced by some workarounds in the ebuild. > 4 weeks passed.
And now 3.0.25c is out. This is a very important package and needs to be maintained on a timely basis.
+1. 3.0.25 has some fixes for Samba talking to the RTM version of Windows Vista which I really need. If it wasn't for that I'd not be so anxious to see the new version show up.
I tried to bump the ebuild the normal way (just by renaming the 3.0.25b ebuild) but the compile fails on some python stuff: building 'smb' extension x86_64-pc-linux-gnu-gcc -pthread -fno-strict-aliasing -DNDEBUG -march=k8 -O3 -pipe -fomit-frame-pointer -msse3 -D_SAMBA_BUILD_=3 -fPIC -I/usr/include/python2.4 -c python/py_smb.c -o build/temp.linux-x86_64-2.4/python/py_smb.o -march=k8 -O3 -pipe -fomit-frame-pointer -msse3 -D_SAMBA_BUILD_=3 -DHAVE_CONFIG_H -DLDAP_DEPRECATED -march=k8 -O3 -pipe -fomit-frame-pointer -msse3 -D_SAMBA_BUILD_=3 -I/var/tmp/portage/net-fs/samba-3.0.25c/work/samba-3.0.25c/source/iniparser/src -Iinclude -I./include -I. -I. -I./lib/replace -I./lib/talloc -I./tdb/include -I./libaddns -I./librpc -DHAVE_CONFIG_H -DLDAP_DEPRECATED -I/var/tmp/portage/net-fs/samba-3.0.25c/work/samba-3.0.25c/source/lib -D_SAMBA_BUILD_=3 python/py_smb.c: In function 'py_smb_connect': python/py_smb.c:51: error: wrong type argument to unary exclamation mark python/py_smb.c: In function 'py_smb_session_request': python/py_smb.c:71: warning: assignment discards qualifiers from pointer target type error: command 'x86_64-pc-linux-gnu-gcc' failed with exit status 1 make: *** [python_ext] Error 1
I did the same, but as the python bindings are experimental and not normally necessary at all, -python as a use flag allows for success. Chris
(In reply to comment #20) > I tried to bump the ebuild the normal way (just by renaming the 3.0.25b ebuild) > but the compile fails on some python stuff: I'm guessing (really, just guessing) that this is an upstream bug. Mainly because the actual code in source/python appears unchanged between 3.0.25b and 3.0.25c and 3.0.25b still builds with the +python flag. Chris
(In reply to comment #22) > (In reply to comment #20) > > I tried to bump the ebuild the normal way (just by renaming the 3.0.25b ebuild) > > but the compile fails on some python stuff: > > I'm guessing (really, just guessing) that this is an upstream bug. Mainly > because the actual code in source/python appears unchanged between 3.0.25b and > 3.0.25c and 3.0.25b still builds with the +python flag. > > Chris > Yes, this is an upstream bug. I have posted a patch upstream. I'll attach some patches for Samba-3.0.25c, and an ebuild (including AFS support).
Created attachment 129222 [details, diff] Patch to fix compile problems with afsacl vfs module.
Created attachment 129224 [details, diff] Patch to make log2pcap compile with --with-afs.
Created attachment 129226 [details, diff] Patch to make py_smb.c build in 3.0.25c.
Created attachment 129230 [details] Ebuild for samba-3.0.25c, including AFS support.
*** Bug 190370 has been marked as a duplicate of this bug. ***
(In reply to comment #27) > Created an attachment (id=129230) [edit] > Ebuild for samba-3.0.25c, including AFS support. you didnt provide the AFS patches
(In reply to comment #29) > (In reply to comment #27) > > Created an attachment (id=129230) [edit] > > Ebuild for samba-3.0.25c, including AFS support. > > you didnt provide the AFS patches > Yes I did, see https://bugs.gentoo.org/attachment.cgi?id=129222 and https://bugs.gentoo.org/attachment.cgi?id=129224
You need to rename the three patches from Johan as: 3.0.25c-afsacl.patch 3.0.25c-afsincludes.patch 3.0.25c-py_smb.patch In that order. And the ebuild as samba-3.0.25c.ebuild
(In reply to comment #31) > You need to rename the three patches from Johan as: > 3.0.25c-afsacl.patch > 3.0.25c-afsincludes.patch > 3.0.25c-py_smb.patch > > In that order. > > And the ebuild as > samba-3.0.25c.ebuild > It looks like Bugzilla mix up "Description" with "Filename". If you look at "Details" for the attachments, they have the correct filename. Normally I write something useful into the "Description" field, not just the filename, as the filename is saved to...
Ok, the latest version is in the CVS and should hit the mirrors soon. Thanks to Johan Andersson for the py_smb-patch. Could you please open a new bug for the AFS-support and add your patches there? Since I wasn't able to build AFS on my machine I couldn't test them and they haven't been added. Please note that OpenAntiVirus support has been removed. Sorry that it took so long to get this into the tree, but as you might know, we could need more devs here...
Hi, thanks for bringing in the new Samba Version. But I think the new Version should not be marked "Stable" now? Thanks, Frank
no, it shouldn't be stable indeed. Sorry, I'm under a lot of pressure lately.
