RE: sun vs. ubuntu issues in dependences ?
You could track changes to krb5 here: http://hg.genunix.org/onnv-gate.hg/log?rev=krb5 but this is only for files matching krb5 phrase. In reality it is much more differences then the above link showing. On Mon, 2009-03-09 at 09:18 +0100, Fredriksson, Ulf wrote: you are absolutely right. The implementation of the kerberos which comes with ON bits tightly integrated with NFS and CIFS services, provides pam module, and actively maintained by opensolaris community. What _exactly_ are the differences? If I'm not completley wrong, the Sun kerberos is/should be based on the MIT Kerberos version. If it is, then maybe merge the change into the krb5 package(s) in a way that doesn't make the krb5 package 'unusable'.. We should have it installed and offered by default. I for one could 'care less' about the Sun kerberos and I would be 'very annoyed' if I couldn't get the MIT Kerberos V packages installed. ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: sun vs. ubuntu issues in dependences ?
Mario, you are absolutely right. The implementation of the kerberos which comes with ON bits tightly integrated with NFS and CIFS services, provides pam module, and actively maintained by opensolaris community. We should have it installed and offered by default. On Sun, 2009-03-08 at 16:46 +0100, Mario Lorenz wrote: Am 07. Mar 2009, um 13:12:12 schrieb Tim Spriggs: Tim, As has been pointed out, the sun kerberos package has several extensions that are required to get some solaris features to work. I assume that would include eg. tickets for kerberized NFS for the NFS client in kernel space would need access to the user's ticket for authentication. If this is indeed the case, User Expectation is that after kinit they are authorized to those services. On Ubuntu, this may even mean having gotten the kerberos tickets through logon. Having two sets of kerberos tools, one for this, and another for that means that Nexenta developers will rightfully be bashed for misunderstanding the single in Single-Sign-On. Please someone with detailed Solaris kerberos experience confirm or deny this, for if it is true, there is thus only two ways out: Either Ubuntu Kerberos gets extended with Solaris patches, or Solaris Kerberos gets promoted standard kerberos, and Ubuntu Kerberos goes away (ie. is replaced with a dummy package). This will likely involve changing the location of some binaries (sun krb5-config has to go to /usr/bin then), and a rebuild of all packages linking kerberos libs. (unless the packages are binary compatible which i highly doubt) Mario ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: sunwsfe package source?
Jason, just remove of conflicting add_drv PCI ID entry from myamanet-sfe/debian/postinst would do, I think. On Sat, 2009-03-07 at 07:57 -0800, Jason Upton wrote: I'd like to remove the conflict between sunwsfe and myamanet-sfe packages by renaming the myamanet-sfe version of the driver (they are currently both named sfe). But I cannot find the source for sunwsfe to change the debian/control file to remove the conflict. Is this part of nexenta-on? Thanks, Jason ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: sun vs. ubuntu issues in dependences ?
On Thu, 2009-03-05 at 20:07 +0100, Mario-Lorenz wrote: Hi, I'm working on some of the mail-related packages, and found some issues that I would like to ask which the better way for a fix would be. a) fetchmail doesnt build because it can not find krb5-config. It can not find it because the sunw kerberos one is in /usr/sun/bin, which is not in its path. It can not find the Ubuntu one because when porting that package, its name was changed to libkrb5-config. situation with libkrb5 is currently not cleared up yet. I'm voting for OpenSolaris krb5 - to be a default in Nexenta. This would preserver compatibility with OpenSolaris distros. The ubuntu packages which are using krb5 are 99% autoconf/automake aware and should be able to detect OpenSolaris diffs. To fix this, do I - change the krb5 package back so it provides krb5-config again, which might cause issues if other packages expect krb5-config to be the sun one (and have usr/sun/bin in their path) or - change the fetchmail package to include usr/sun/bin in its path and thus link against the sun kerberos instead of the nexenta one (are there any specific reasons why we need the two of them, anyway ?) or - change the fetchmail package to use libkrb5-config, which is an unnecessary deviation from ubuntu (and possibly other packages might suffer from a simmilar problem ? if we would have krb5 as I suggested above - the issue wouldn't exists. b) mutt does not build because the permissions of /var/mail differ from ubuntu. (sun var/mail has more lenient permissions, hence the configure stage thinks it can do without the mutt dotlock helper, and doesnt compile it, and the packaging stage wants to pack it in... Again, this has to be fixed in either the sun package owning /var/mail, or the mutt (and possibly other) packages... I'll try to fix that in sunwcsr. What's the permission it needs to be? c) Is there an overriding rationale on when to use Sun userspace vs. Ubuntu userspace ? sendmail, rmail etc. come to mind as examples where this happens... I would prefer to keep OpenSolaris stuff and provide an alternative for Ubuntu ones. ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: nexenta-builder and dialog package
Yes, I did some modifications to dialog, so it works better with Nexenta installer. I recommend to upload original Elatte dialog package - it is tested and it is stable. On Sat, 2009-01-17 at 12:52 -0800, Jason Upton wrote: Thanks for the feedback Tim. I've been able to build a CD image now that the new ed package doesn't conflict with base-files, but the newer version of dialog from Ubuntu Hardy doesn't have --no-escape or --no-signals options. This causes all of the dialog commands to fail and the installer doesn't work at all. Digging a bit, it looks like in Elatte that Erast had made some changes to the dialog package to add these options (presumably for nexenta-install.sh since it uses the options). I'm diffing the old Elatte packages to see what needs to change in the new Hardy version. Erast - can you explain a little bit about why these additional options are needed? Is it just for the nexenta-install.sh script or do other parts of Nexenta make use of these options? Do you think that the installer could be re-written to not need them and just use the dialog package as-is? Thanks a bunch. As soon as I get this working with the regular hardy iso I hope to enable an xfce4 iso builder with the updated/added myamanet network drivers and new memtest86+ (the memtest part of the new iso I built is working). Jason On Fri, Jan 16, 2009 at 9:08 AM, Tim Spriggs t...@tajinc.org wrote: Jason Upton wrote: OK, I have figured out most of it. In nbld-bootstrap, I changed: --component=main to: --component==main,contrib,non-free And I also had to change the hardy-unstable.bootstrap file: mirror_style main to: mirror_style release The mirror_style of main forces debootstrap into only finding packages from a hard-coded main repository regardless of what --components is set to. That looks great! This is related to the main-only zone install issues as well. The /usr/lib/nexenta-zones/hardy-unstable.bootstrap file for the zones has mirror_style set to main. If I change that to release along with a change to /usr/lib/lu/lucreatezone to change the debootstrap option to --components=main,contrib,non-free then packages like nvi can be found during the zone install. The extra contrib directory hierarchy I mentioned does not seem to affect any of this. The Packages file in the repository knows the real location and it gets passed to debootstrap properly. So while it looks strange, it still works. My next problem is related to an apparent ed and base-file conflict - the nexenta-builder and zone install are now both failing due to ed creating /bin/ed and base-files wanting to create a symlink from /bin to /usr/bin. But the dependencies are such that ed is installed before base-files (so the symlink creation errors out). I was going to change ed to install to /usr/bin/ed but I cannot find the source package in the repo Anyone know where it's at? Maybe I'll just re-upload it from upstream source. I've seen this style of conflict before, I think we need to build some kind of testing framework for packages that catch common errors like these. Perhaps lintian can be extended. I will fix the nexenta-builder and nexenta-zones packages with the changes and upload unless someone sees a flaw/problem with what I mentioned. And while I'm updating nexenta-builder, does anyone object to updating memtest86+ to the latest version (v2.11)? I need the latest to properly test one of my motherboards/chipsets. Thanks, Jason Go for it! Thanks for digging in. -Tim ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: Top links to prstat
Is top ported to OpenSolaris yet? I'd like to see zone extensions in it too... Also is anybody interested in porting GNU chmod and ls utilities to work with ZFS ACL's ? Such development, if successful, would somewhat close the gap between SUN / GNU tools and make everybody happy, I think.. On Thu, 2009-01-15 at 10:12 +0100, Jerome Warnier wrote: Running NCP 2.0 up-to-date. I noticed that top is a link to prstat, which is somewhat bad, as they both don't have the same output or feature set (for example, prstat supports zones, while top does not). I guess the best solution would be to ship GNU top in NCP instead of trying to simulate it. Regards ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: 8000 packages in the NCP2 repo: 3200 missing
BTW, 9300 packages imported now. 3700 missing... On Sun, 2008-12-28 at 10:27 -0500, Tim Spriggs wrote: I think we need to wait and see what happens when main is enabled for builds before we can judge what will happen ultimately with depends. Right. This make sense to fix prior to addressing the rest of it. I don't think having packages with unmet depends in the unstable repository is a bad thing. I think in the worst scenario we can simply migrate known good packages to testing/stable and leave unmet packages in unstable. Having the packages there helps to get people involved: if they want to use the package they can help satisfy a few depends. +1 I generally agree with this statement. I forgot that testing is not yet propagated, so we can use this... Good idea. I wonder if we could utilize apt-get's ordering simulation, i.e. to verify if package A is good do this in a script: if apt-get install -s A; then echo A: good upload hardy-testing A else echo A: bad ... fi ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: Panic on boot
you might need to verify that zpool / zfs versions of your checkpoint(s) are all the latest: zpool get version syspool = v10 zfs get -r version syspool = v3 However, keep in mind that if you upgrade to v10, you will not be able to load previous checkpoint with b85. Another possible reason is that you need to installgrub on all the slices in syspool and ensure that you updated archive. All the above can be done by using b85-based ISO, i.e. NCP2 Alpha2 for instance. On Tue, 2008-12-16 at 16:22 -0500, Tim Spriggs wrote: This is the second host with b104 I am getting this on. Has anyone else seen this: module /platform/i86pc/kernel/amd64/unix: text at [0xfb80, 0xfb91fcc3] data at 0xfbc0 module /kernel/amd64/genunix: text at [0xfb91fcd0, 0xfbb6c1c7] data at 0xfbc7c1c0 Loading kmdb... module /kernel/misc/amd64/kmdbmod: text at [0xfbce4fc0, 0xfbd8193f] data at 0xfbd81940 module /kernel/misc/amd64/ctf: text at [0xfbb6c1d0, 0xfbb76b47] data at 0xfbd9cc00 Welcome to kmdb kmdb: unable to determine terminal type: assuming `vt100' Loaded modules: [ unix krtld genunix ] [0] :c SunOS Release 5.11 Version NexentaOS_20081207 64-bit Loading Nexenta... features: 61076fffcpuid,sse3,nx,asysc,sse2,sse,pat,cx8,pae,mca,mmx,cmov,de,pge,mtrr,msr,tsc,lgpg mem = 4062756K (0xf7f89000) initialized model-specific module 'cpu_ms.AuthenticAMD.15' on chip 0 core 0 strand 0 root nexus = i86pc pseudo0 at root pseudo0 is /pseudo scsi_vhci0 at root scsi_vhci0 is /scsi_vhci isa0 at root pseudo-device: acpippm0 acpippm0 is /pseudo/acpi...@0 pseudo-device: ppm0 ppm0 is /pseudo/p...@0 ramdisk0 at root ramdisk0 is /ramdisk panic[cpu0]/thread=fbc2c520: assertion failed: nvlist_lookup_uint64(config, ZPOOL_CONFIG_POOL_TXG, txg) == 0, file: ../../common/fs/zfs/spa.c, line: 2307 fbc4d060 genunix:assfail+7e () fbc4d0b0 zfs:spa_check_rootconf+8f () fbc4d150 zfs:spa_get_rootconf+6c () fbc4d190 zfs:spa_import_rootpool+25 () fbc4d1f0 zfs:zfs_mountroot+eb () fbc4d210 genunix:fsop_mountroot+1a () fbc4d240 genunix:rootconf+12d () fbc4d280 genunix:vfs_mountroot+63 () fbc4d2b0 genunix:main+f0 () fbc4d2c0 unix:_locore_start+92 () panic: entering debugger (no dump device, continue to reboot) Loaded modules: [ scsi_vhci uppc cpu_ms.AuthenticAMD.15 zfs specfs pcplusmp cpu.generic ] kmdb: target stopped at: kmdb_enter+0xb: movq %rax,%rdi [0] ::ps SPID PPID PGIDSIDUID FLAGS ADDR NAME R 0 0 0 0 0 0x0001 fbc2b8b0 [0] ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: Panic on boot
This been fixed in b104 I believe On Tue, 2008-12-16 at 17:28 -0500, Thomas Garner wrote: I wasn't seeing them at boot, but after upgrading to b102, I saw spontaneous reboots, often several times a day. It was bad enough that I reverted to b85. Dec 2 04:00:11 filer0 ^Mpanic[cpu2]/thread=868b8dc0: Dec 2 04:00:11 filer0 genunix: [ID 335743 kern.notice] BAD TRAP: type=e (#pf Page fault) rp=868b8b24 addr=10 occurred in module zfs due to a NULL po inter dereference Dec 2 04:00:11 filer0 unix: [ID 10 kern.notice] Dec 2 04:00:11 filer0 unix: [ID 839527 kern.notice] sched: Dec 2 04:00:11 filer0 unix: [ID 753105 kern.notice] #pf Page fault Dec 2 04:00:11 filer0 unix: [ID 532287 kern.notice] Bad kernel fault at addr=0x10 Dec 2 04:00:11 filer0 unix: [ID 243837 kern.notice] pid=0, pc=0xfeb48836, sp=0x868b8b8c, eflags=0x10296 Dec 2 04:00:11 filer0 unix: [ID 211416 kern.notice] cr0: 8005003bpg,wp,ne,et,ts,mp,pe cr4: 6d8xmme,fxsr,pge,mce,pse,de Dec 2 04:00:11 filer0 unix: [ID 624947 kern.notice] cr2: 10 Dec 2 04:00:11 filer0 unix: [ID 625075 kern.notice] cr3: 72a8000 Dec 2 04:00:11 filer0 unix: [ID 10 kern.notice] Dec 2 04:00:11 filer0 unix: [ID 537610 kern.notice] gs: fe9301b0 fs: 8164 es: 82140160 ds: 160 Dec 2 04:00:11 filer0 unix: [ID 537610 kern.notice]edi: d9236db8 esi: b41c50b0 ebp: 868b8ba8 esp: 868b8b5c Dec 2 04:00:11 filer0 unix: [ID 537610 kern.notice]ebx: d9236db8 edx: 868b8dc0 ecx:0 eax: 868b8b8c Dec 2 04:00:11 filer0 unix: [ID 537610 kern.notice]trp:e err:0 eip: feb48836 cs: 158 Dec 2 04:00:11 filer0 unix: [ID 717149 kern.notice]efl:10296 usp: 868b8b8c ss: b83f9e00 Dec 2 04:00:11 filer0 unix: [ID 10 kern.notice] Dec 2 04:00:11 filer0 genunix: [ID 353471 kern.notice] 868b8a60 unix:die+93 (e, 868b8b24, 10, 2) Dec 2 04:00:11 filer0 genunix: [ID 353471 kern.notice] 868b8b10 unix:trap+1431 (868b8b24, 10, 2) Dec 2 04:00:11 filer0 genunix: [ID 353471 kern.notice] 868b8b24 unix:cmntrap+7c (fe9301b0, 8164,) Dec 2 04:00:12 filer0 genunix: [ID 353471 kern.notice] 868b8ba8 zfs:arc_read_done+1e (d9236db8, 2, 0, 85f) Dec 2 04:00:12 filer0 genunix: [ID 353471 kern.notice] 868b8be8 zfs:zio_done+239 (d9236db8, aba05a80,) Dec 2 04:00:12 filer0 genunix: [ID 353471 kern.notice] 868b8c08 zfs:zio_execute+81 (d9236db8, aba05a80,) Dec 2 04:00:12 filer0 genunix: [ID 353471 kern.notice] 868b8c48 zfs:zio_notify_parent+af (d9236db8, 85fb06e0,) Dec 2 04:00:12 filer0 genunix: [ID 353471 kern.notice] 868b8c88 zfs:zio_done+26e (85fb06e0, b717d490,) Dec 2 04:00:12 filer0 genunix: [ID 353471 kern.notice] 868b8ca8 zfs:zio_execute+81 (85fb06e0, 1, 868b8c) Dec 2 04:00:12 filer0 genunix: [ID 353471 kern.notice] 868b8ce8 zfs:zio_notify_parent+af (85fb06e0, c01a1490,) Dec 2 04:00:12 filer0 genunix: [ID 353471 kern.notice] 868b8d28 zfs:zio_done+26e (c01a1490, 97b2a860,) Dec 2 04:00:12 filer0 genunix: [ID 353471 kern.notice] 868b8d48 zfs:zio_execute+81 (c01a1490, 817e0dc0) Dec 2 04:00:12 filer0 genunix: [ID 353471 kern.notice] 868b8da8 genunix:taskq_thread+192 (97b2a840, 0) Dec 2 04:00:12 filer0 genunix: [ID 353471 kern.notice] 868b8db8 unix:thread_start+8 () Dec 2 04:00:12 filer0 unix: [ID 10 kern.notice] Dec 2 04:00:12 filer0 genunix: [ID 672855 kern.notice] syncing file systems... Dec 2 04:00:12 filer0 genunix: [ID 904073 kern.notice] done Dec 2 04:00:13 filer0 genunix: [ID 111219 kern.notice] dumping to /dev/dsk/c2d0s7, offset 213909504, content: kernel Dec 2 04:01:04 filer0 genunix: [ID 409368 kern.notice] ^M100% done: 223711 pages dumped, compression ratio 1.40, Dec 2 04:01:04 filer0 genunix: [ID 851671 kern.notice] dump succeeded Thomas On Tue, Dec 16, 2008 at 4:29 PM, Phillip Steinbachs p...@steinbachs.org wrote: On Tue, 16 Dec 2008, Tim Spriggs wrote: This is the second host with b104 I am getting this on. Has anyone else seen this: Yes, I got this the other day after doing a forced b85 - 2.0 alpha1 upgrade. It can't import the root pool, and I had to rebuild the boot archive (boot from cd and mount the appropriate syspool, update using the bootadm -R flag). -phillip ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: nexenta-on source package
That was me - I accidentally uploaded the package without source. I'm planning to push it tonight/tomorrow with more CIFS/AD fixes included. On Sun, 2008-12-14 at 12:24 +0100, C. Bergström wrote: Hi guys.. I'm trying to find the latest nexenta-on source package.. Can you push that somewhere please.. I'm also curious why it's not available already? Thanks ./C ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: b85 - b103 upgrade (needs polish)
what's the output of zfs mount command? On Mon, 2008-12-01 at 22:28 -0800, Michael Bushey wrote: I'm running SunOS random 5.11 NexentaOS_20081118 i86pc i386 i86pc Solaris - I guess that's 103b, I don't know how to tell. # apt-clone dist-upgrade This operation will upgrade your system using ZFS capabilities. Proceed ? (y/n) y Updating APT sources ... Downloading upgrades and checking if reboot will be required. This may take a few minutes. Please wait... Verifying free space... Success. Upgrade requires 149.10MB of available free space. This upgrade will require REBOOT. Proceed? (y/n) y Upgrade is in progress. Please DO NOT interrupt... Creating Upgrade Checkpoint... apt-clone.SystemCallError: mount failed: Device busy Not sure what to make of the Device busy Michael ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: nexenta-on b102 uploaded
Thanks for verification. Basically, its i.nametomajor awk script mistakenly killed dada/scsi classes for ata on upgrade in /etc/driver_classes On Tue, 2008-11-25 at 20:57 -0500, Thomas Garner wrote: Yep, the fix worked beautifully (too bad I didn't say something before I nuked my old root). No problems upgrading and rebooting. What was the fix? Thomas On Tue, Nov 25, 2008 at 3:42 PM, Erast Benson [EMAIL PROTECTED] wrote: OK, I think it should be fixed now. Could you please try to rollback this upgrade and try again? On Sun, 2008-11-23 at 21:18 -0500, Thomas Garner wrote: Got a negative experience for you. I tried upgrading from a elatte-unstable install to no avail, reinstalled to 2.0/hardy alpha2, then tried upgrading again, only to get to the same unbootable state. Here's the primary failure at boot: SunOS Release 5.11 Version NexentaOS_20081024 32-bit Loading Nexenta... NOTICE: error reading device label NOTICE: *** * This device is not bootable! * * It is either offlined or detached or faulted. * * Please try to boot from a different device.* *** NOTICE: spa_import_rootpool: error 19 Cannot mount root on /[EMAIL PROTECTED],0/[EMAIL PROTECTED],1/[EMAIL PROTECTED]/[EMAIL PROTECTED],0:a fstype zfs panic[cpu0]/thread=fec1e2a0: vfs_mountroot: cannot mount root fec37674 genunix:vfs_mountroot+324 (fec47930, fec33468,) fec37684 genunix:main+bf () Also worth noting that when I tried to dist-upgrade from elatte-unstable, the first thing it did was kill dpkg: Preparing to replace dpkg 1.13.22nexenta10 (using .../dpkg_1.14.16.6nexenta6_solaris-i386.deb) ... Unpacking replacement dpkg ... ld.so.1: dpkg: fatal: libintl.so.8: open failed: No such file or directory E: Sub-process /usr/bin/dpkg exited unexpectedly Luckily a symlink saved me from being completely dead in the water. Thomas On Sun, Nov 16, 2008 at 4:40 PM, Erast Benson [EMAIL PROTECTED] wrote: Dear Nexenta Developers, I just uploaded b102 bits (yes with 6771840 backout) into Hardy unstable and managed to upgrade couple of Alpha2 environments. So far so good and I'd like to give you a green light for upgrades of your dev. environments. As usual, if something goes wrong, use apt-clone to rollback the changes and please share your positive/negative experience with us. ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
nexenta-on b102 uploaded
Dear Nexenta Developers, I just uploaded b102 bits (yes with 6771840 backout) into Hardy unstable and managed to upgrade couple of Alpha2 environments. So far so good and I'd like to give you a green light for upgrades of your dev. environments. As usual, if something goes wrong, use apt-clone to rollback the changes and please share your positive/negative experience with us. ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: dh_installsmf parsing bug?
On Sun, 2008-11-02 at 10:52 +0530, Anil Gulecha wrote:
Hi,
I think I'm running into a parsing bug with dh_installsmf with the
package postgresql-8.3.. the manifest seems fine and defines a section
as follows..
-snip manifest-
service_bundle type='manifest' name='postgresql-8.3'
service
name='database/postgresql-8.3'
type='service'
version='1'
-
however, running:
# dh_installsmf --service postgresql-8.3 --manifest
debian/postgres.xml -ppostgresql-8.3
Wrong manifest file (postgresql-8.3 != postgresql) at
/usr/bin/dh_installsmf line 129
Somehow $name is picked up as postgresql, and not as postgresql-8.3..
so the regular expression below may be at fault.. can someone suggest
the fix.. the perl seems greek to me.
-
my ($found, $name, $group) = (0, , );
while (FD) {
if (!$found /\s*service/) {
$found = 1;
next if (!
/\s*service\s+.*name\s*=\s*[']?(\S+)[']?/);
($group, $name) = ($1, $2) if
(/[']?([\w\/]+)\/(\w+)[']?/);
change this regexp to:
/[']?([\w\-\.\/]+)\/([\w\-\.]+)[']?/
last;
}
if ($found) {
next if (/^\s*$/);
next if (! /\s*name\s*=\s*([']?\S+[']?)/);
($group, $name) = ($1, $2) if
(/[']?([\w\/]+)\/(\w+)[']?/);
and this to:
/[']?([\w\-\.\/]+)\/([\w\-\.]+)[']?/
last;
}
}
close FD;
die Couldn't parse manifest file if ($name eq || $group eq );
die Wrong manifest file ($service != $name) if ($service ne
$name);line 129
but may be we should find a better way to parse it..
Regards
Anil
___
gnusol-devel mailing list
gnusol-devel@lists.sonic.net
http://lists.sonic.net/mailman/listinfo/gnusol-devel
___
gnusol-devel mailing list
gnusol-devel@lists.sonic.net
http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: gnusol-devel Digest, Vol 29, Issue 3
Hi, this only possible for 2.0, and still nexenta-on package (you need it) has some problem yet. But good news are - we are working on it, and soon we should have easy in use nexenta-on package which would allow us to recompile nexenta ON/NWS core in debian way. On Thu, 2008-10-23 at 18:46 +0800, jklyekai wrote: Hello, can I re-compile sunwcakr_5.11.tar.gz to sunwcakr_5.11-nexenta_solaris-i386.deb? in what entrionment ? accroding to the http://www.nexenta.org/os/BuildingPackages I uncompress the sunwcakr.tar.gz and cd sunwcakr ;dpkg-buildpackage -sa then the error is like this Use of uninitialized value in -d at /usr/bin/dh_installproto line 31. Use of uninitialized value in concatenation (.) or string at /usr/bin/dh_installproto line 31. Invalid dir: at /usr/bin/dh_installproto line 31. How can I do? -- jklyekai 2008-10-23 -- Message: 1 Date: Mon, 06 Oct 2008 21:40:10 -0700 From: Erast Benson [EMAIL PROTECTED] Subject: nexenta-on package uploaded To: gnusol-devel [EMAIL PROTECTED] Message-ID: [EMAIL PROTECTED] Content-Type: text/plain Dear Nexenta Developers, nexenta-on package which includes full nexenta kernel sources uploaded into hardy repository now. Its very initial, so don't expect much... the work will continue. Those who interested should be able to re-compile nexenta kernel if needed now. Enjoy! -- Message: 2 Date: Tue, 07 Oct 2008 10:12:37 -0700 From: Tim Spriggs [EMAIL PROTECTED] Subject: Re: nexenta-on package uploaded To: Erast Benson [EMAIL PROTECTED] Cc: gnusol-devel [EMAIL PROTECTED] Message-ID: [EMAIL PROTECTED] Content-Type: text/plain; charset=ISO-8859-1; format=flowed awesome! Great work :) Erast Benson wrote: Dear Nexenta Developers, nexenta-on package which includes full nexenta kernel sources uploaded into hardy repository now. Its very initial, so don't expect much... the work will continue. Those who interested should be able to re-compile nexenta kernel if needed now. Enjoy! ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel -- ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel End of gnusol-devel Digest, Vol 29, Issue 3 *** ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: [osol-discuss] GNU libc on OpenSolaris
Right. And in addition to autotools, such port complicates further ON merges which will unavoidably lead to higher rate of errors/bugs. But because GNU/kFreeBSD exists, I do not see why GNU/kOpenSolaris can't be... On Fri, 2008-09-19 at 09:27 -0400, Michael Casadevall wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Debian's main issue is that parts of Sun's libc are not open (mostly libc_i18n; they require all bits to be open). Having seen the issues kFreeBSD has had with using glibc with their kernel, I'm not sure if its work having a ksolaris port since configure will no longer identify the platform as Solaris, so most autotools scripts will break. Michael -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (GNU/Linux) Comment: http://getfiregpg.org iEYEARECAAYFAkjTqNQACgkQpblTBJ2i2pteBACdET5A0ycn3U+G3S2R+8mCN6vq 0oAAniom7MRTL3P4TR8H1PotiT+R+qSi =8cf5 -END PGP SIGNATURE- On Fri, Sep 19, 2008 at 5:55 AM, Joerg Schilling [EMAIL PROTECTED] wrote: Erast Benson [EMAIL PROTECTED] wrote: Yes, such port makes sense and solves some of the issues (mostly GNU libc portability) but unfortunately creates new issues, which I'm sure, could be worked out and soon we should have more or less working first ISO available with support for this new exciting architecture! What do you expect from this port? In glibc, features expected on Solaris are missing. I would expect that this port would rather create portabilitly problems than solving any issue. makes sense to use glibc. This would also solve the legal problem that Debian had with linking Sun's libc with dpkg [1]. glibc is licensed under LGPL with a linking exception, so linking CDDL code against the glibc is also legal. In keeping with past glibc ports (e.g. kFreeBSD, Debian is a license troll. There are two ways to deal with this kind of trolling: 1) Ignore it comppletely 2) find evidence that the claims from Debian are nonsense. Taking actions on the Debian trolling is definitely the wrong way. BTW: Sun lawyers knows that there is no problem with linking GPLd applications against CDDL libraries. The GPL does not forbid it (in fact the GPL does not say anything about it as this is something that happens outside the GPL work). Sun would not ship GNOME and /usr/gnu/* if Sun would not be _very_ certain that Debian is trolling. Sun is happily waiting for being sued by a copyright holder of a GPLd program shipped with OpenSolaris. _this_ is one way of implementing (2) above. Jörg -- EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED](uni) [EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: [osol-discuss] GNU libc on OpenSolaris
To me, this development is just yet another Debian architecture and sure, some in Debian community will like. It also connects to Nexenta in many ways - which is good for us. We can't stop such port from happening - so I think we should embrace it as a secondary lefty architecture. On Fri, 2008-09-19 at 11:37 -0400, Michael Casadevall wrote: The kFreeBSD port has had a lot of considerable issues with porting software. Remember, we'd need to port the ON tools such as the ZFS admin tools to glibc. http://wiki.debian.org/ArchiveQualification/kfreebsd-i386 They also haven't been able to get things like the wifi tools for FreeBSD working. I'm not saying that adapting glibc is a bad thing, but we need to figure out if we really want to go down this path. Michael On Fri, Sep 19, 2008 at 10:48 AM, Erast Benson [EMAIL PROTECTED] wrote: Right. And in addition to autotools, such port complicates further ON merges which will unavoidably lead to higher rate of errors/bugs. But because GNU/kFreeBSD exists, I do not see why GNU/kOpenSolaris can't be... On Fri, 2008-09-19 at 09:27 -0400, Michael Casadevall wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Debian's main issue is that parts of Sun's libc are not open (mostly libc_i18n; they require all bits to be open). Having seen the issues kFreeBSD has had with using glibc with their kernel, I'm not sure if its work having a ksolaris port since configure will no longer identify the platform as Solaris, so most autotools scripts will break. Michael -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (GNU/Linux) Comment: http://getfiregpg.org iEYEARECAAYFAkjTqNQACgkQpblTBJ2i2pteBACdET5A0ycn3U+G3S2R+8mCN6vq 0oAAniom7MRTL3P4TR8H1PotiT+R+qSi =8cf5 -END PGP SIGNATURE- On Fri, Sep 19, 2008 at 5:55 AM, Joerg Schilling [EMAIL PROTECTED] wrote: Erast Benson [EMAIL PROTECTED] wrote: Yes, such port makes sense and solves some of the issues (mostly GNU libc portability) but unfortunately creates new issues, which I'm sure, could be worked out and soon we should have more or less working first ISO available with support for this new exciting architecture! What do you expect from this port? In glibc, features expected on Solaris are missing. I would expect that this port would rather create portabilitly problems than solving any issue. makes sense to use glibc. This would also solve the legal problem that Debian had with linking Sun's libc with dpkg [1]. glibc is licensed under LGPL with a linking exception, so linking CDDL code against the glibc is also legal. In keeping with past glibc ports (e.g. kFreeBSD, Debian is a license troll. There are two ways to deal with this kind of trolling: 1) Ignore it comppletely 2) find evidence that the claims from Debian are nonsense. Taking actions on the Debian trolling is definitely the wrong way. BTW: Sun lawyers knows that there is no problem with linking GPLd applications against CDDL libraries. The GPL does not forbid it (in fact the GPL does not say anything about it as this is something that happens outside the GPL work). Sun would not ship GNOME and /usr/gnu/* if Sun would not be _very_ certain that Debian is trolling. Sun is happily waiting for being sued by a copyright holder of a GPLd program shipped with OpenSolaris. _this_ is one way of implementing (2) above. Jörg -- EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED](uni) [EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: GNU libc on OpenSolaris
Hi David, Great work! Yes, such port makes sense and solves some of the issues (mostly GNU libc portability) but unfortunately creates new issues, which I'm sure, could be worked out and soon we should have more or less working first ISO available with support for this new exciting architecture! This is absolutely great development, and here is what I think it means for Nexenta Community: 1) We now have two architectures on-going: - NexentaCore Sun/OpenSolaris - NexentaCore GNU/kOpenSolaris 2) We should expect more developers coming on board from Debian/Ubuntu/Linux land and polish our APT repository even more 3) Nexenta OpenSolaris-oriented packaging will improve and many of Nexenta patches will be backported back to upstream - Ubuntu and Debian 4) Debian acceptance of these two architectures - might happen real soon. On Fri, 2008-09-19 at 00:18 -0400, David Bartley wrote: Hello, I've been working on porting GNU libc (glibc) to OpenSolaris (see [0] for source code and other downloads). I've also ported NPTL (the glibc thread library), using the corresponding OpenSolaris syscalls. I haven't benchmarked much, but I suspect speed should be comparable to native OpenSolaris. At this point I have a working chroot with a number of packages installed and working, and I hope to have a zone booting soon. Xorg, GNOME, and gdb all work. I also have a number of Solaris binaries working, including truss, dladm, zfs, and zpool. (I haven't tried, but it should be fairly straightforward to get dtrace working.) In order to get these working, I implemented additional OpenSolaris extensions (mostly trivial syscall wrappers). Examples are the functions defined in ucred.h, priv.h, and sys/pset.h. Since nexenta is trying to be OpenSolaris with the GNU userland, it makes sense to use glibc. This would also solve the legal problem that Debian had with linking Sun's libc with dpkg [1]. glibc is licensed under LGPL with a linking exception, so linking CDDL code against the glibc is also legal. In keeping with past glibc ports (e.g. kFreeBSD, kNetBSD), I've used the target string i486-kopensolaris-gnu (the 64-bit target would be x86_64-kopensolaris-gnu). If you want to test this out in a chroot, there's a bunch of pre-built i386 binaries available [2] that can be just untarred into a directory. You'll also need to mount /proc, /dev, and /devices in the chroot via lofs (this won't be needed if done in a zone). Many applications will require OpenSolaris headers to build, which can just be copied from an OpenSolaris/Nexenta box (just don't overwrite any headers installed by glibc). -- David [0] http://csclub.uwaterloo.ca/~dtbartle/opensolaris/ [1] http://lists.debian.org/debian-devel/2006/04/msg00069.html [2] http://csclub.uwaterloo.ca/~mspang/opensolaris/builds/ ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: A review of the base system
On Wed, 2008-09-17 at 10:11 -0400, Michael Casadevall wrote: * A LOT of source packages are missing (most of the sunw*) Will be uploading this this one in 2-3 weeks. This should not be a problem at least. But I'd like to make sure that it is build-able on NCP, even with some help of native nevada (solaris 11) zone.. ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: aspell issues
On Tue, 2008-09-16 at 13:03 +0530, Anil Gulecha wrote: On Mon, Sep 15, 2008 at 7:56 PM, Anil Gulecha [EMAIL PROTECTED] wrote: Hi, I've been building various aspell packages. Most package fine into debs, but almost all have the below issue when installing ( I've uploaded the debs anyway.) Ex: Setting up aspell-he (1.0-0-1nexenta1) ... Unsupported conversion: Invalid argument at /usr/share/perl5/Debian/DictionariesCommon.pm line 399, DICT line 1. What could be causing the above? Ok.. I looked further, and this perl module needs to be fixed for Solaris. my $converter = Text::Iconv-new ($codingsystem, utf8); to my $converter = Text::Iconv-new ($codingsystem, UTF-8); How do I find out what package installs the file /usr/share/perl5/Debian/DictionariesCommon.pm ? dpkg -S /usr/share/perl5/Debian/DictionariesCommon.pm ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: python-gobject does not install
this usually happens when you trying to access APT repo in the middle of uploads. Just wait couple minutes and try again. On Sun, 2008-09-14 at 08:28 +0530, Anil Gulecha wrote: Hi, ONe of the build dependencies for a package to be installed is python-gobject .. but [EMAIL PROTECTED]:~/work1/python-gtk2/pygtk-2.12.1# apt-get install python-gobject python-gobject-dev Reading package lists... Done Building dependency tree Reading state information... Done The following packages were automatically installed and are no longer required: bc libarchive-zip-perl perl-suid rsync tk8.4 libfile-rsyncp-perl Use 'apt-get autoremove' to remove them. The following extra packages will be installed: libffi4 libffi4-dev Suggested packages: python-gobject-dbg The following NEW packages will be installed: libffi4 libffi4-dev python-gobject python-gobject-dev 0 upgraded, 4 newly installed, 0 to remove and 4 not upgraded. Need to get 150kB/211kB of archives. After this operation, 468kB of additional disk space will be used. Do you want to continue [Y/n]? y Get:1 http://apt.nexenta.org hardy-unstable/main python-gobject 2.14.1-2nexenta3 [97.8kB] Get:2 http://apt.nexenta.org hardy-unstable/main python-gobject-dev 2.14.1-2nexenta3 [52.1kB] Fetched 52.5kB in 0s (2438kB/s) Failed to fetch http://apt.nexenta.org/dists/hardy-unstable/main/binary-solaris-i386/python/python-gobject_2.14.1-2nexenta3_solaris-i386.deb Size mismatch Failed to fetch http://apt.nexenta.org/dists/hardy-unstable/main/binary-all/python/python-gobject-dev_2.14.1-2nexenta3_all.deb Size mismatch E: Unable to fetch some archives, maybe run apt-get update or try with --fix-missing? This is after a apt-get update. Whats the issue here? Anil ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
upstream patches policy
Just talked to Ubuntu developers on #ubuntu-devel. Added section to http://www.nexenta.org/os/NexentaRepositoryPolicyMain with suggestions. ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: Incompatible libgcc?
apt-get install lib64gcc1 or similar should help.. On Fri, 2008-09-05 at 13:52 +0530, Anil Gulecha wrote: Hi, I was building the package libnspr4-dev, when make failed with an incompatible -lgcc. what does this mean and any suggested workarounds? Regards Anil [snip] dh_testdir /usr/bin/make -C mozilla/nsprpub make[1]: Entering directory `/root/work1/libnspr4-dev/nspr-4.7.1~beta2/mozilla/nsprpub' cd config; /usr/bin/make -j1 export make[2]: Entering directory `/root/work1/libnspr4-dev/nspr-4.7.1~beta2/mozilla/nsprpub/config' cc -m64 -o now.o -c -Wall -pipe -Wall -O2 -fno-strict-aliasing -g -fPIC -UDEBUG -DMOZILLA_CLIENT=1 -DNDEBUG=1 -DHAVE_VISIBILITY_HIDDEN_ATTRIBUTE=1 -DHAVE_VISIBILITY_PRAGMA=1 -DXP_UNIX=1 -DSVR4=1 -DSYSV=1 -D__svr4=1 -D__svr4__=1 -DSOLARIS=1 -DHAVE_FCNTL_FILE_LOCKING=1 -D_PR_HAVE_OFF64_T=1 -D_LARGEFILE64_SOURCE=1 -D_PR_INET6=1 -D_REENTRANT=1 -DHAVE_POINTER_LOCALTIME_R=1 -DFORCE_PR_LOG -D_PR_PTHREADS -UHAVE_CVAR_BUILT_ON_SEM -D_PR_INET6 now.c cc -m64 now.o -Wl,-Bsymbolic-functions -o now /usr/bin/ld: skipping incompatible /usr/lib/gcc/i386-pc-solaris2.11/4.2.3/libgcc.a when searching for -lgcc /usr/bin/ld: skipping incompatible /usr/lib/gcc/i386-pc-solaris2.11/4.2.3/libgcc.a when searching for -lgcc /usr/bin/ld: cannot find -lgcc collect2: ld returned 1 exit status make[2]: *** [now] Error 1 make[2]: Leaving directory `/root/work1/libnspr4-dev/nspr-4.7.1~beta2/mozilla/nsprpub/config' make[1]: *** [export] Error 2 make[1]: Leaving directory `/root/work1/libnspr4-dev/nspr-4.7.1~beta2/mozilla/nsprpub' make: *** [build-stamp] Error 2 dpkg-buildpackage: failure: debian/rules build gave error exit status 2 ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: Simple autobuild script
Good one. It uses all the right tools - which is most critical here. On Fri, 2008-09-05 at 23:25 +0530, Anil Gulecha wrote: hi, I've written a simple build script to automatically go though the building steps for a package. In the best case, where no change is required to a upstream package.. the package will be built in the current directory/$package .. Run as follows: ./attempt_build package Script attached. Anil ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Nexenta Alpha2 Hackaton minutes
Hi All, Hackaton is still in progress! Here goes short history of events so far: * NexentaCore 2.0 Alpha2 first 'testing' ISO available - ON b97/b98 based. People reported successful installations, likely we will have official Alpha2 next week, as usual, upgradable with power of Debian tools - best in the world software packaging system which we all love some much * Slowly but surely we are making progress on porting latest Ubuntu LTS packages to the new shiny Nexenta Hardy repository: [EMAIL PROTECTED]:~# apt-cache stats Total package names : 6502 (260k) Normal packages: 3907 Pure virtual packages: 41 Single virtual packages: 509 Mixed virtual packages: 49 Missing: 1996 Yes, almost 4000 packages ALREADY ported! * Work on our new powerful ZFS and Nexenta Zones integrated AutoBuilder continuing. We are hoping to quickly add 10,000 more packages in the Alpha3-Beta1 time frame by simply running AutoBuilder against the rest of Ubuntu Universe/Multiverse * There was some discussion on possibility of creation new distribution Nexenta Desktop - but so far inconclusive. This opportunity is still open for developers - DO NOT MISS IT * There was some discussion on possiblity of creation Nexenta Server distribution - that is to mirror and match Ubuntu Server - the decision so far - YES! The majority of members of Nexenta community thinks this is a great idea and willing to put efforts into it. Initial User Guide on how Nexenta Server will look like available here: http://doc.ubuntu.com/ubuntu/serverguide/C/ REMINDER: Hackaton still in progress - join and make a difference. ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: Gtk build issues
Seems like include path for GLIB is not properly set. glib.h not found
On Sun, 2008-08-31 at 16:20 +0530, Anil Gulecha wrote:
The core dump is now done away with.. i compiled glib2.16.4 removing
the -lgthread flag.
configure still fails at
configure:24210: checking for extra flags for POSIX compliance
configure:24233: gcc -c -g -O2 -Wall conftest.c 5
conftest.c: In function 'main':
conftest.c:39: warning: unused variable 'dir'
configure:24239: $? = 0
configure:24244: result: none needed
configure:24387: checking for pkg-config
configure:24417: result: /usr/bin/pkg-config
configure:24446: checking pkg-config is at least version 0.16
configure:24449: result: yes
configure:24467: checking for GLIB - version = 2.13.5
configure:24627: result: no
configure:24665: gcc -o conftest -g -O2 -Wallconftest.c -lintl 5
conftest.c:36:18: error: glib.h: No such file or directory
conftest.c: In function 'main':
conftest.c:42: error: 'glib_major_version' undeclared (first use in
this function)
conftest.c:42: error: (Each undeclared identifier is reported only once
conftest.c:42: error: for each function it appears in.)
conftest.c:42: error: 'glib_minor_version' undeclared (first use in
this function)
conftest.c:42: error: 'glib_micro_version' undeclared (first use in
this function)
configure:24671: $? = 1
configure: failed program was:
| /* confdefs.h. */
| #define PACKAGE_NAME gtk+
[snip program]
| #define ENABLE_NLS 1
| #define GTK_LOCALEDIR /usr/local/share/locale
| /* end confdefs.h. */
|
| #include glib.h
| #include stdio.h
|
| int
| main ()
| {
| return ((glib_major_version) || (glib_minor_version) ||
(glib_micro_version));
| ;
| return 0;
| }
configure:24707: error:
*** GLIB 2.13.5 or better is required. The latest version of
*** GLIB is always available from ftp://ftp.gtk.org/pub/gtk/.
glib 2.16.4 is installed on the system.
[EMAIL PROTECTED]:~/libgtk2.0-0/gtk+2.0-2.12.9# dpkg -l | grep libglib
ii libglib2.0-0 2.16.4-0nexenta3
The GLib library of C routines
ii libglib2.0-0-dbg 2.16.4-0nexenta3
The GLib libraries and debugging symbols
ii libglib2.0-data 2.16.4-0nexenta3
Common files for GLib library
ii libglib2.0-dev2.16.4-0nexenta3
Development files for the GLib library
ii libglib2.0-doc2.16.4-0nexenta3
Documentation files for the GLib library
Regards
Anil
___
gnusol-devel mailing list
gnusol-devel@lists.sonic.net
http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: Gtk build issues
We need to debug why GLIB-linked application core dumps... Do we have any other glib-linked packages uploaded and working? On Sat, 2008-08-30 at 20:59 +0530, Anil Gulecha wrote: Hi, I'm tryign to build the gtk2.0 package for hardy.. and configure fails with .. checking for pkg-config... (cached) /usr/bin/pkg-config checking pkg-config is at least version 0.16... yes checking for GLIB - version = 2.13.5... no *** Could not run GLIB test program, checking why... *** The test program compiled, but did not run. This usually means *** that the run-time linker is not finding GLIB or finding the wrong *** version of GLIB. If it is not finding GLIB, you'll need to set your *** LD_LIBRARY_PATH environment variable, or edit /etc/ld.so.conf to point *** to the installed location Also, make sure you have run ldconfig if that *** is required on your system *** *** If you have an old version installed, it is best to remove it, although *** you may also be able to get things to work by modifying LD_LIBRARY_PATH configure: error: *** GLIB 2.13.5 or better is required. The latest version of *** GLIB is always available from ftp://ftp.gtk.org/pub/gtk/. from config.log: configure:24591: gcc -o conftest -g -O2 -Wall -D_REENTRANT -D_PTHREADS -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include conftest.c -lgobject-2.0 -lgmo dule-2.0 -lgthread-2.0 -lpthread -lthread -lglib-2.0 -lintl -liconv -lintl 5 configure:24594: $? = 0 configure:24600: ./conftest ./configure: line 24601: 20986 Segmentation Fault (core dumped) ./conftest$ac_exeext configure:24603: $? = 139 configure: program exited with status 139 configure: failed program was: | /* confdefs.h. */ | #define PACKAGE_NAME gtk+ | #define PACKAGE_TARNAME gtk+ .snip program that failes. .. | } | } | return 1; | } | configure:24627: result: no configure:24665: gcc -o conftest -g -O2 -Wall -D_REENTRANT -D_PTHREADS -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include conftest.c -lintl -lgobject-2.0 -lgmodule-2.0 -lgthread-2.0 -lpthread -lthread -lglib-2.0 -lintl -liconv 5 configure:24671: $? = 0 configure:24707: error: *** GLIB 2.13.5 or better is required. The latest version of *** GLIB is always available from ftp://ftp.gtk.org/pub/gtk/. This error has widely seen on solaris packages that require glib. i've tried setting LD_LIBRARY_PATH=/usr/lib.. but that didnt help. Has anyone faced this error before? Anil ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: Bug: m4 and sunwcsu both provide /usr/bin/m4, causing a collision
On Tue, 2008-08-26 at 03:37 -0700, Michael Bushey wrote: I was thinking of porting my favourite editor XEmacs... I'm a vi guy, but this is exactly what Nexenta needs, a complete (ie comparable to Ubuntu or Debian) repository. It seems that focused push works best for Nexenta development. Read more about upcoming Hackathon events here: http://www.nexenta.org/os/Hackathon We are asking all interested hackers and developers to join our forces and push Alpha 2 release on these weekends and next following week. That is - to complete Hardy/main repository. Work coordinated on IRC: http://www.nexenta.org/chat ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: Patching a file already being patched
Usually there should be a way to disable usage of fakeroot. But I could confirm that the version of fakeroot (binary) available on SchiliX ftp worked fine for me.. On Mon, 2008-08-18 at 15:27 +0530, Anil Gulecha wrote: On Sun, Aug 17, 2008 at 8:01 PM, Anil Gulecha [EMAIL PROTECTED] wrote: Hi, i'm in a situation where i need to modify configure.in to add a flag. However, dpkg-buildpackage cribs about not being able to cleanly remove a patch.. this patch being in debian/patched, and also modifying configure.in How do i go about modifying this file to include the previous patch as well? One of the methods that I've read about is dpatch-edit-patch.. however, this complains about missing fakeroot. Has anyone worked with this command/ found any documentation that works with nexenta? Regards Anil ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: Patching a file already being patched
I do not remember, I think it was part of SchiliX ISO.. On Tue, 2008-08-19 at 10:06 +0530, Anil Gulecha wrote: On Mon, Aug 18, 2008 at 9:03 PM, Erast Benson [EMAIL PROTECTED] wrote: Usually there should be a way to disable usage of fakeroot. But I could confirm that the version of fakeroot (binary) available on SchiliX ftp worked fine for me.. I was unable to find it. Is it one among ftp://ftp.berlios.de/pub/schillix ? Anil? ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: how to tell dput to upload uninstalled packages
dput says: please install package locally, so to verify that it installs cleanly before uploading it. Have you tried this yet? On Fri, 2008-08-15 at 22:42 +0530, Anil Gulecha wrote: Hi, I'm loading 4 packages, of which 2 conflict with each other. (thus i cannot install them both at the same time). dput thus fails .. what argument do i pass dput to tell it to go ahead and upload without the inbuild check. [EMAIL PROTECTED]:~/elinks_tmp# dput *.changes $USER not set, will use login. Upload package to host nexenta Uninstalled Package. Test it before uploading it. Anil ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: how to tell dput to upload uninstalled packages
If nothing else, dput is a python script you could temporarily disable check - or better extend it with an option which will do so. On Fri, 2008-08-15 at 23:48 +0530, Anil Gulecha wrote: On Fri, Aug 15, 2008 at 11:45 PM, Erast Benson [EMAIL PROTECTED] wrote: dput says: please install package locally, so to verify that it installs cleanly before uploading it. Have you tried this yet? Yes. I've installed all individual packages (elinks, elinks-doc, elinks-data, elinks-lite) Elinks, elinks-doc and elinks-data go together. elinks-lite conflicts with the above, so i had to remove them and then tried elinks-lite which installed fine too. Anil ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: how to tell dput to upload uninstalled packages
Oh, also to force installation of conflicting packages you could do something like: dpkg --force-all -i *.deb But this is not the best way to handle this, I think, it would be better to disable check in dput. On Sat, 2008-08-16 at 00:03 +0530, Anil Gulecha wrote: On Fri, Aug 15, 2008 at 11:53 PM, Erast Benson [EMAIL PROTECTED] wrote: On Fri, 2008-08-15 at 23:48 +0530, Anil Gulecha wrote: On Fri, Aug 15, 2008 at 11:45 PM, Erast Benson [EMAIL PROTECTED] wrote: dput says: please install package locally, so to verify that it installs cleanly before uploading it. Have you tried this yet? Yes. I've installed all individual packages (elinks, elinks-doc, elinks-data, elinks-lite) Elinks, elinks-doc and elinks-data go together. elinks-lite conflicts with the above, so i had to remove them and then tried elinks-lite which installed fine too. Anil If nothing else, dput is a python script you could temporarily disable check - or better extend it with an option which will do so. I'm wondering how the current version maintainer (Tim) did it. CCed. Anil ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: Avoiding a file being added to a package
On Thu, 2008-08-14 at 20:00 +0530, Anil Gulecha wrote: On Thu, Aug 14, 2008 at 7:51 PM, Erast Benson [EMAIL PROTECTED] wrote: Another way is to use dpkg-divert - but in this case its a bug. charset.alias must be shipped with libiconv library only. I hung around at #elinks, and turns out the package has a copy of gettext in it. which geneates the file. I think you need to ask debian/rules to use system-wide libicon/gettext.. Also the package has an 'ubuntu' in it (as it's already in the our repo.. i got it via apt-get source elinks)? Is this alright? If not, how do I remove ubuntu? In debian/changelog for first line where version specified: s/ubuntu/nexenta/ apt-upstream-tool does that for you. Also, over time it will do some other policy things... so I would suggest to use it and try to improve it if you'll find bugs and time permits. Caution: you always need to check if a package exists in our repository before getting it as is from Ubuntu. Could be that some amount of work already done by someone and all we need is to fix existing package, instead of importing new one from upstream. [EMAIL PROTECTED]:~/elinks_tmp# apt-get source elinks Reading package lists... Done Building dependency tree Reading state information... Done NOTICE: 'elinks' packaging is maintained in the 'Git' version control system at: git://git.debian.org/git/collab-maint/elinks.git Need to get 3149kB of source archives. Get:1 ftp://linux.stanford.edu hardy/main elinks 0.11.3-5ubuntu2 (dsc) [1231B] Get:2 ftp://linux.stanford.edu hardy/main elinks 0.11.3-5ubuntu2 (tar) [3127kB] Get:3 ftp://linux.stanford.edu hardy/main elinks 0.11.3-5ubuntu2 (diff) [20.8kB] Fetched 3149kB in 0s (7305kB/s) dpkg-source: extracting elinks in elinks-0.11.3 dpkg-source: unpacking elinks_0.11.3.orig.tar.gz dpkg-source: applying ./elinks_0.11.3-5ubuntu2.diff.gz [EMAIL PROTECTED]:~/elinks_tmp# ls elinks-0.11.3 elinks_0.11.3-5ubuntu2.diff.gz elinks_0.11.3-5ubuntu2.dsc elinks_0.11.3.orig.tar.gz ~Anil ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: ncp2a1 build-essentials issue
I think that you could substitute some of the packages from Elatte/Dapper. (NCP 1.0). Two apt-get/dpkg options could be especially helpful here: apt-get --reinstall install ... dpkg --force-all ... But long term - we need to add missing packages to the NCP 2.0 repository.. On Tue, 2008-08-12 at 11:23 +0530, Anil Gulecha wrote: I installed ncp2 on VMware server on ubuntu guest and logged in. I cant pull in build-essential: [EMAIL PROTECTED]:~# apt-get update Get:1 http://apt.nexenta.org hardy-unstable Release.gpg [187B] Get:2 http://apt.nexenta.org hardy-unstable Release [7103B] Get:3 http://apt.nexenta.org hardy-unstable/main Packages [648kB] Get:4 http://apt.nexenta.org hardy-unstable/contrib Packages [14B] Get:5 http://apt.nexenta.org hardy-unstable/non-free Packages [3239B] Get:6 http://apt.nexenta.org hardy-unstable/main Sources [165kB] Get:7 http://apt.nexenta.org hardy-unstable/contrib Sources [14B] Get:8 http://apt.nexenta.org hardy-unstable/non-free Sources [1674B] Fetched 825kB in 12s (66.2kB/s) Reading package lists... Done [EMAIL PROTECTED]:~# apt-get install build-essentials Reading package lists... Done Building dependency tree... Done E: Couldn't find package build-essentials [EMAIL PROTECTED]:~# apt-get install build-essential Reading package lists... Done Building dependency tree... Done You might want to run `apt-get -f install' to correct these: The following packages have unmet dependencies: build-essential: Depends: libc6-dev but it is not going to be installed or libc-dev but it is not going to be installed Depends: type-handling but it is not going to be installed tasksel: Depends: aptitude (= 0.2.15-1) but it is not installable E: Unmet dependencies. Try 'apt-get -f install' with no packages (or specify a solution). How is the gnusolaris.org machine currently setup.. with ncp1 + devzone. From what Tim metioned on IRC, I gathered a similar setup could be done locally (using ncp1 or 2a). regards Anil ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Nexenta Hackathon in Progress - Status
Hi, A lot of fun these days and a lot of achievements! June's Hackathon-in-progress status: * Debian tools such as: devscripts, dpkg, apt, debarchiver, etc all ported * compiled bunch of NCP 2.0 packages (gcc prerequisites) * newisys-nas.stanford.edu machine prepared. It is 4-core 2.6GHz Opteron with 8GB RAM * gcc-4.2 and Hardy's binutils tool-chain is available NOW * rebuild of Hardy with new tool chain - pass #1 complete * started work on very first NCP 2.0 Alpha1 ISO * preparation to second Hardy rebuild in progress Outstanding issues: --- * Not all uploaded packages in Nexenta Hardy repository could be built out of source + diff * Many packages still missing - but we are making good progress on this one * Xorg is not yet uploaded * Firefox is not yet uploaded * Auto Builder work not yet started - with ZFS-powered Auto Builder in place we are planning to quickly port up to 10,000+ Ubuntu/Hardy packages in fully automatic way Since we have one more build machine - SSH access for developers much easier to grant - apply now at #nexenta on freenode IRC More information here: http://www.nexenta.org/os/Hackathon --- The Nexenta Team ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
1.0.1-test4 available for testing
This is to announce 1.0.1-test4 (elatte NCP 1.x branch) release. Main fixes and changes: * 100+ new packages in elatte-testing repository * LSI mega_sas driver for LSI SAS/SATA HBAs * IBM ServeRAID FreeBSD ported ips driver for 4x,5x,6x series HBAs * ZFS write-throttle fixes from b87 and b91 Give it a try! As usual, you can upgrade from elatte-testing APT repository. Enjoy! ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: unison
In this case, we probably need to split unison package. Following NCP policy here to accept console-only/daemon-only/library-only packages... On Thu, 2008-05-29 at 04:14 +0300, Marcus Sundman wrote: Leonardo Rodrigues de Mello [EMAIL PROTECTED] wrote: I am working at porting unison debian package and it dependency to nexenta. Any progress? If you're having problems with the gui then forget about the gui for now, since the gui is only used on the client side anyway, if it's used at all. - Marcus ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Announcing Hackathon event this Saturday (05/31/08)
Dear NexentaOS Developers and Users! Starting this Saturday, we are inviting all Debian/Nexenta and OpenSolaris developers and users to participate in our first Hackathon event. Hackathon starts on this Saturday 05/31/2008 and continues during first week of June. Development coordinated at #nexenta on freenode IRC. We are migrating from Ubuntu Dapper to Ubuntu Hardy user land and will be hacking on very initial version of NCP 2.0. You don't need to install Nexenta, (ZFS-powered) build environment and SSH access to the dedicated Zone will be automatically (see devzone package) provided to interested developers by request on #nexenta freenode IRC. The work coordinated at #nexenta freenode IRC and this static web page: http://www.nexenta.org/Hackathon ZFS-powered Nexenta Zones managed on specially prepared NexentaOS build servers connected to the public Internet. Servers and Internet collocation kindly provided by Stanford University, CA. Join now and make a difference! It is going to be fun! --- The Nexenta Team ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: [nexenta-changes] Accepted Package: write-throttle-zfs
http://hg.genunix.org/onnv-gate.hg/rev/1a2a7cfb9f26 On Thu, 2008-05-29 at 00:09 +0200, Nico Sabbi wrote: Il Wednesday 28 May 2008 19:43:19 Erast Benson ha scritto: Hi Guys, just uploaded ZFS write-throttle fix into unstable. Give it a try! can you explain what this fix does, please? ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: [Fwd: [nexenta-changes] Incoming Package: suphp_0.6.1-1nexenta1]
Daniel, yes, this is the problem. It should return 0 on success. Could you put 'echo' in the code and see which line causing this? We would then could come up with the fix. On Thu, 2008-05-15 at 14:55 +0100, Daniel Llewellyn wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Erast Benson wrote: I think there is a bug in apache1 postinst. Try to run it like this: # /var/lib/dpkg/info/apache.postinst configure Then see why it complains in postinst. This might need some fixing... hmm, I ran the above command, but it just returns with no output. running `echo $?` reports that it's return code was 1. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.8 (Darwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkgsQOEACgkQVR0QzBAS6zYkSQCdGPU+blCrV/lgPOx0o7dqInBC AjEAn33S/SED1qgvpSlUG4owkvcz6ZeS =JUF/ -END PGP SIGNATURE- ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
easySMF
For those who interested in creating SMF services for Debian packages, here is nice web GUI, which should generate initial SMF .xml file: http://es.opensolaris.org/easySMF/ Examples of SMFed services in Nexenta are rsync, apache2, etc. We are using dh_installsmf deb helper and the process pretty straight forward. ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: 1.0.1-test3 available for testing
Great. Thanks Tim. Could you elaborate on what does it mean (from practical perspective) to use preferred settings? Should user change anything from the default installation? If so, we should make this change permanent.. I also tested it quite a bit on multiple machines - looks good. I'm planning to push -test4 this week, and 1.0.1 next week. On Wed, 2008-04-23 at 11:35 -0700, Tim Spriggs wrote: I have found a work-around for the x4100 if an admin is willing to match the other NICs settings instead of using preferred settings. I don't have any other critical bugs to note. Thanks, -Tim Erast Benson wrote: Thanks for the information! By any chance, do you have bug number for this? Could you look it up if fix is already in? On Wed, 2008-04-09 at 13:52 -0700, Tim Spriggs wrote: Hi Erast, There is a new bug in the e1000g driver that does not allow flow control to be advertised correctly. This means that link aggregates on the x4100 don't work where there are two nge interfaces and two e1000g interfaces. This seems to affect the latest Sol10U4 patches and the latest Nexenta testing base. -Tim Erast Benson wrote: Guys, I just uploaded 1.0.1-test3 for your testing pleasure. Please give it a try - if we will not find much of show stoppers in this one, we will move it to stable - probably by the end of this week. Few important fixes in this release: 1) libdlpi now works more reliably with recent name vanity putback 2) significantly improved speed of boot_archive creation - up to 5 times faster now... on my test box it takes ~ 7 seconds to recreate archives (comparing to 60 seconds before the change) 3) disabled broken UFS/manual partitioning related code in installer Link: http://www.nexenta.org/releases ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: 1.0.1-test3 available for testing
Thanks for the information! By any chance, do you have bug number for this? Could you look it up if fix is already in? On Wed, 2008-04-09 at 13:52 -0700, Tim Spriggs wrote: Hi Erast, There is a new bug in the e1000g driver that does not allow flow control to be advertised correctly. This means that link aggregates on the x4100 don't work where there are two nge interfaces and two e1000g interfaces. This seems to affect the latest Sol10U4 patches and the latest Nexenta testing base. -Tim Erast Benson wrote: Guys, I just uploaded 1.0.1-test3 for your testing pleasure. Please give it a try - if we will not find much of show stoppers in this one, we will move it to stable - probably by the end of this week. Few important fixes in this release: 1) libdlpi now works more reliably with recent name vanity putback 2) significantly improved speed of boot_archive creation - up to 5 times faster now... on my test box it takes ~ 7 seconds to recreate archives (comparing to 60 seconds before the change) 3) disabled broken UFS/manual partitioning related code in installer Link: http://www.nexenta.org/releases ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: NCP 1.0.1-test1 available for testing
I'm sorry you have such an impression - but, hey, this is open source stuff - don't complain and fix it yourself if you can. Remember, most of Nexenta developers doing this work for free at their free time. On Wed, 2008-03-19 at 14:51 +0200, Marcus Sundman wrote: Erast Benson [EMAIL PROTECTED] wrote: Test! Test! Test! This may sound overly negative, but what's the point of testing when nexenta devs obviously don't give a frack (as shown by the utter lack of response to my error report here 3 weeks ago)? - Marcus ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: Custom Nexenta Build
apt-get install nexenta-builder On Thu, 2008-02-28 at 19:39 +0530, Harshavardhana wrote: Hi Nexeanta Group, I wanted to know is there a way that i can build a custom distribution like nexenta with minimal applications, in my case 1. Bash prompt. 2. OpenSolaris kernel. 3. Some GNU Utilities. or 1. OpenSolaris kernel. 2. Busybox. So with this i will boot into opensolaris and have only prompt available. Does nexenta provides its build scripts downloadable?. I have tried Schillix, Nevada build scripts but nothing give me reasonable documentation regarding proper procedural builds. As we have in Linux it's quite easy to bring up a Initrd type image and work around. Is this something possible with Solaris too? Can anyone give me some pointers here regarding how to write our own build scripts or atleast to look upon for referencing other scripts. Thanks in advance ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: how to track the status of my uploaded package
I assume you already subscribed to nexenta-changes@ mailing list? You should see your upload there, and some of FTP-masters will examine it and accept it - you should see accept or reject. On Thu, 2008-02-14 at 23:39 +0800, Aubrey Li wrote: I uploaded a new package, and wonder how to track the status of it? For example, if/when is it accepted? if not, why? ..., etc. Thanks, -Aubrey ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: xsvc driver is missing
apt-get install sunwxsvc On Tue, 2008-01-29 at 16:30 +0100, Nico Sabbi wrote: On Tuesday 29 January 2008 15:26:04 Aubrey Li wrote: On nexenta, xsvc driver is missing. So some solaris special applications failed due to no /dev/xsvc. Is it intended? Why? -Aubrey I asked the same question a couple of times but I never had a convincing answer: now xsvc is CDDL like the rest, so why not including it in Nexenta? ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Nexenta Core Platform RC3 available
Nexenta Core Platform (NexentaCP) RC3 is available at: http://www.nexenta.org Release highlights: * OpenSolaris build 80+ (non-debug) + critical fixes from b81-b83 [1]: - critical fixes for native CIFS server - critical fixes for ZFS - critical bug fixes for SATA and networking stacks * CIFS client now included into default installation * Xen Dom0 (experimental support) [2] * GRUB-integrated memory test for x86 included * Installer fixes, now installs on machines with 256MB just fine * APT repository - status COMPLETE! ~2500 tested packages included * Minor bug fixes all over.. * Many new documents and articles now available [3] Developers and Users - join official Nexenta IRC #nexenta now! [1] http://www.nexenta.org/os/PatchSet_b80 [2] http://www.nexenta.org/os/NexentaXenDom0 [3] http://www.nexenta.org/os/Documentation Enjoy! The Nexenta Team. ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: SUNWspro stuff
On Fri, 2008-01-25 at 13:37 +0800, Aubrey Li wrote: I really want to make nexenta to be a development platform. So, besides Sun Studio, I'd like to the following package are added. 1) ON source tools: Mercurial done (in unstable now) 2) ON Specific Build Tools: SUNWonbld.i386.tar.bz2 doesn't make sense without actual fixes for ON tree to be build-able on NCP... 3) cscope: browse ON source code done (in unstable now) 4) I know multi-console is removed from solaris, if the default terminal is console, not X, I want the multi-console to be back. will wait till actual VC integration... 5) memory requirement. I noticed nexenta installation needs at least 384M memory. Otherwise the installation failed. How is this value determined? why so big? why not 256M? fixed. Thanks, -Aubrey ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
Re: The future of Nexenta revealed
On Tue, 2007-07-17 at 15:07 +0200, Martin Man wrote: Nico Sabbi wrote: what I'd really like to have in some more documentation, especially guides on how to package ON stuff (not all of it is available in Nexenta, including pieces that are CDDL such as the xsvc driver, BTW why?) so that we poor users can contribute more yes, the scripts to compile package nevada should be available publicly in the long run, I personally would like them to become part of the repository itself, much in the same way as kernel-package does for linux kernel. Just FYI: I have already tried to compile Nevada under Nexenta to see what are the external dependencies and gotchas in order to have Nexenta self-contained. Hopefully Erast will give us more clue regarding the ON build scripts. I'd say, it would make sense to make ON compilable on Nexenta first, workout issues and come up with diff first. And after that I will try to find some time to put everything together as a buildable sunw source package. Currently those scripts are very much SXCR specific and needs to be fixed to work properly in native env too. -- Erast ___ gnusol-devel mailing list gnusol-devel@lists.sonic.net http://lists.sonic.net/mailman/listinfo/gnusol-devel
