Re: [OpenIndiana-discuss] Cannot run texlive 2024
On Tue, Aug 27, 2024 at 1:29 PM Udo Grabowski (IMK) wrote: > > > > On 8/27/24 14:20, Stephan Althaus wrote: > > On 8/27/24 12:53, Udo Grabowski (IMK) wrote: > >> On 8/27/24 12:45, Apostolos Syropoulos via openindiana-discuss wrote: > >>> Each year I compile TeXLive so I have binaries that have been > >>> compiled on a fairly recent version of OI/hipster.If you like, I can > >>> upload somewhere the binaries that I have compiled and I am using > >>> every day. > >> > >> Oh, that would be great, would save a lot of work ! > >> > >> > >>> > >>> On Tuesday, August 27, 2024 at 01:26:12 PM GMT+3, Udo Grabowski > >>> (IMK) wrote: > >>> On 8/27/24 12:06, Udo Grabowski (IMK) wrote: > >>>> Hi, > >>>> > >>>> downloaded TeXLive 2024, but unfortunately this seems to be > >>>> > >>>> incompatible with illumos libc.so.1 : > At some point recently (it's not in the release notes, so undocumented) TeX Live started being built with Solaris 11, which isn't compatible with illumos. You'll have to go back to an older release from the historic archive, eg https://ftp.tu-chemnitz.de/pub/tug/historic/systems/texlive/2022/tlnet-final -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Fonts in some websites become pixelated
On Wed, Jul 31, 2024 at 4:42 AM scavenger via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > Please review the way you configure fonts in your OS. In > /etc/fonts/conf.d, you are putting conflicting configuration files. For > example: 70-yes-bitmaps.conf and 70-no-bitmaps.conf are both there. Can you > guess what is the effect of this? I can't. > Oh dear, that's horribly broken. The way that fontconfig works is to install a whole list of template configuration files into .../conf.avail, and then symlinks are placed into conf.d to pick up the configuration you actually want. The OpenIndiana build appears to trample all over this. There's a minor bug, in that you can use --with-templatedir='/etc/fonts/conf.avail' rather than the current mechanism of leaving it at the default /usr/share/fontconfig/conf.avail and moving it after the install. But the actual problem is caused by creating symbolic links in conf.d for *all* the files in conf.avail, rather than (a) either leaving the links that 'make install' creates alone, or (b) removing and recreating only the symlinks that are present in conf.d (the way that solaris-userland does it, where it notes that 'Make links in conf.d be relative links to ../conf.avail instead of absolute'). > Jul 31, 2024, 08:07 by openindiana-discuss@openindiana.org: > > > For example, github.com. It seems the bitmap fonts are picked. Should > you configure it out of the box to disable bitmap fonts? > > > > ___ > > openindiana-discuss mailing list > > openindiana-discuss@openindiana.org > > https://openindiana.org/mailman/listinfo/openindiana-discuss > > > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] CMake can't find LLVM
On Wed, Jul 31, 2024 at 1:02 AM scavenger via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > This is the error message: > > > By not providing "FindLLVM.cmake" in CMAKE_MODULE_PATH this project has > > asked CMake to find a package configuration file provided by "LLVM", > but > > CMake did not find one. > > > > Could not find a package configuration file provided by "LLVM" with > any of > > the following names: > > > > LLVMConfig.cmake > > llvm-config.cmake > > > > Add the installation prefix of "LLVM" to CMAKE_PREFIX_PATH or set > > "LLVM_DIR" to a directory containing one of the above files. If "LLVM" > > provides a separate development package or SDK, be sure it has been > > installed. > > I have installed build-essential and I have llvm-config. > Follow the instructions; something like (depending on which version of llvm is installed and/or you wish to use) cmake ... -DCMAKE_PREFIX_PATH=/usr/clang/18.1/lib/cmake In general handling multiple versions of a package is a tricky problem for cmake, not helped by cmake's inconsistencies in searching for the cmake configuration files. I guess it would be possible to link the cmake files into the default locations using a mediator to select the "default" version, which would at least give an answer most of the time. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] wallpaper and mwm on oi
On Fri, Jun 28, 2024 at 7:51 PM Matthew Yapit via openindiana-discuss <
openindiana-discuss@openindiana.org> wrote:
> Hello,
> I'm using oi SunOS iron 5.11 illumos-cd9ea253fe i86pc i386 i86pc.
> I've recently built mwm from lesstif-0.95.2 sources. I added a startup
> file motif.desktop to /usr/share/xsessions. I'd like to have a
> wallpaper. I found I can do this from .profile with
>
> display -window root -backdrop
> /export/home/matt/Documents/Pictures/wallpaper/motif.png &
>
That's the correct command (or, at least, a correct command). The tricky
bit is to
get it to run in the right place.
> The problem is, it is a bit clunky. The image shows up briefly whenever
> I login to MATE and other window managers.
>
Yes, it will draw the desktop background, but MATE and other newish window
managers then put a real window in front of it to draw their wallpaper on.
> I'm aware of ~/Mwm ~/.mwmrc and ~/.Xdefaults. So far it has been futile
> to edit these files in reguards to starting a wallpaper in mwm at login.
>
> Does anyone know how to start a wallpaper from imagemagick or xsetroot
> in mwm at startup via some script or dot file?
>
The way I would do this is to create a wrapper. Rather than call mwm
directly
from the motif.desktop file, create a /usr/bin/mwm-session file (as a shell
script), and
Exec that from the desktop file. It might look like
#!/bin/sh
if [ -f ${HOME}/.mwmstartup ]; then
. ${HOME}/.mwmstartup
fi
exec mwm
Then you, or any user, can create a $HOME/.mwmstartup file that contains
any customization commands you wish. Or pre-launch your favourite
applications.
(Historically, we would have done essentially the same thing in our
~/.xinitrc
files. This was well before things like sessions and desktop files had even
been thought of, so you have to replicate some of the functionality.)
--
-Peter Tribble
http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/
___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] ask for a small contribution related Pulseaudio,
On Sat, Feb 10, 2024 at 8:17 AM Carsten Grzemba via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > For do some preliminary testing on update Pulseaudio I am interested in > a small contribution. > > If anyone has more then one audio device, I am interested in the output of: > > $ ls -l /dev/audio* > > $ ls -l /dev/sound* > > > $ ls -l /dev/dsp* > My experience is that with multiple sound devices, the wrong one is picked as the default more often than not. On my machine, I've had to replace the unversioned links to device 1 rather than the default 0. And set AUDIODEV to /dev/dsp1 to get GStreamer applications to work. Some things talk to /dev/mixer as well... > ls -l /dev/audio* lrwxrwxrwx 1 root root 7 Apr 10 2019 /dev/audio -> sound/1 lrwxrwxrwx 1 root root 10 Apr 10 2019 /dev/audioctl -> sound/1ctl > ls -l /dev/sound* total 12 lrwxrwxrwx 1 root root 9 Mar 20 2019 0 -> audiohd:0 lrwxrwxrwx 1 root root 12 Mar 20 2019 0ctl -> audiohd:0ctl lrwxrwxrwx 1 root root 9 Mar 20 2019 1 -> audiohd:1 lrwxrwxrwx 1 root root 12 Mar 20 2019 1ctl -> audiohd:1ctl lrwxrwxrwx 1 root root 48 Mar 20 2019 audiohd:0 -> ../../devices/pci@0,0/pci1028,5a4@3:sound,audio0 lrwxrwxrwx 1 root root 51 Mar 20 2019 audiohd:0ctl -> ../../devices/pci@0,0/pci1028,5a4@3:sound,audioctl0 lrwxrwxrwx 1 root root 46 Mar 20 2019 audiohd:0dsp -> ../../devices/pci@0,0/pci1028,5a4@3:sound,dsp0 lrwxrwxrwx 1 root root 48 Mar 20 2019 audiohd:0mixer -> ../../devices/pci@0,0/pci1028,5a4@3:sound,mixer0 lrwxrwxrwx 1 root root 49 Mar 20 2019 audiohd:1 -> ../../devices/pci@0,0/pci1028,5a4@1b:sound,audio1 lrwxrwxrwx 1 root root 52 Mar 20 2019 audiohd:1ctl -> ../../devices/pci@0,0/pci1028,5a4@1b:sound,audioctl1 lrwxrwxrwx 1 root root 47 Mar 20 2019 audiohd:1dsp -> ../../devices/pci@0,0/pci1028,5a4@1b:sound,dsp1 lrwxrwxrwx 1 root root 49 Mar 20 2019 audiohd:1mixer -> ../../devices/pci@0,0/pci1028,5a4@1b:sound,mixer1 > ls -l /dev/dsp* lrwxrwxrwx 1 root root 4 Feb 8 2020 /dev/dsp -> dsp1 lrwxrwxrwx 1 root root 18 Mar 20 2019 /dev/dsp0 -> sound/audiohd:0dsp lrwxrwxrwx 1 root root 18 Mar 20 2019 /dev/dsp1 -> sound/audiohd:1dsp > ls -l /dev/mixer* lrwxrwxrwx 1 root root 7 Dec 9 18:36 /dev/mixer -> sndstat lrwxrwxrwx 1 root root 20 Mar 20 2019 /dev/mixer0 -> sound/audiohd:0mixer lrwxrwxrwx 1 root root 20 Mar 20 2019 /dev/mixer1 -> sound/audiohd:1mixer > cat /dev/mixer SunOS Audio Framework Audio Devices: 0: audiohd#0 onboard1, a (DUPLEX) 1: audiohd#1 onboard1, a (DUPLEX) Mixers: 0: audiohd#0 onboard1, a Unknown HD codec: 0x80862807 1: audiohd#1 onboard1, a Unknown HD codec: 0x10ec0280 -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] DNS server and domain from DHCP?
Hi, On Tue, Dec 19, 2023 at 6:33 PM Philip Kime wrote: > I have used ipadm to configure an interface via DHCP, which works find. > However, the DHCP server also provide the DNS server and default domain > which was not being set on my server. I found that I had entries in > /etc/resolv.conf which seem to be set by SMF and are repopulated after > reboot. Two questions: > > * Where can I find the dns server and domain settings in SMF? I tried > looking in dns/client but I can't see config/nameserver property in there. > I wonder if this is nwam. It can also save and restore network settings, apparently. Some details are embedded in /lib/svc/method/net-loc and /lib/svc/method/net-nwam, but I'm not sure if there's much in the way of accurate documentation. (But I would start with: /usr/sbin/nwamcfg list /usr/sbin/nwamcfg "select loc Automatic ; get -V dns-nameservice-configsrc" and investigate from there.) * If delete the SMF settings for server and domain, will the OI dhcp client > set them as it does on all other OSes on my network? > > PK > -- > Dr Phil. Philip Kime > BA (Hons) Msc (Dist) PhD > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] container with same ip
On Thu, Nov 16, 2023 at 10:45 PM Goetz T. Fischer wrote: > the zone seems fine now but nat is not working at all. i tried a bunch of > examples from whatever docs i > could find but no luck. Have you enabled ipv4 forwarding in the global zone? Something like routeadm -e ipv4-forwarding (It turns out that there are so many ways to configure ipv4 forwarding - routeadm as above, ipadm set-prop, or ipadm set-ifprop on a per-interface basis.) > i did the following on the host (global zone): > > dladm create-etherstub etherstub0 > dladm create-vnic -l etherstub0 gvnic1 > ipadm create-ip gvnic1 > ipadm create-addr -T static -a 10.0.0.1/24 gvnic1/v4 > dladm create-vnic -l etherstub0 zvnic1 > > then i installed the zone using zvnic1 in its config. i assigned 10.0.0.2 > and set the gateway to > 10.0.0.1. for testing i changed the zone's sshd port to 23456 and i can > reach that from the global zone. > so far so good. > > now it would be time for nat and the redirections. e1000g0 is my real nic > and just for this example > let's say its ip is 1.2.3.4. > i added the following to ipnat.conf: > > map e1000g0 from 10.0.0.0/24 to any -> 1.2.3.4 > rdr e1000g0 from any to any port = 23456 -> 10.0.0.2 port 23456 > > when i try to connect to 1.2.3.4 on port 23456 from the outisde > (foreign/outside ip 1.2.3.5) i get a > connection but nothing happens. ipnat -l shows: > > List of active MAP/Redirect filters: > map e1000g0 from 10.0.0.0/24 to any -> 1.2.3.4/32 > rdr e1000g0 from any to 1.2.3.4/32 port = 23456 -> 10.0.0.2 port 23456 tcp > > List of active sessions: > RDR 10.0.0.223456 <- -> 1.2.3.4 23456 [1.2.3.5 50249] > > > when i try to reach the internet from inside the zone i get nothing > either. so it seems the redirect > works but nat does not. > once more, any help would be very welcome ... > > > On Wed, 15 Nov 2023 17:50:52 +, Peter Tribble wrote: > > The way I do this (on OmniOS and Tribblix, but conceptually it's the same > > on all Solaris/illumos > > variants): > > > > 1. Create an etherstub > > > > 2. In the global zone, create a vnic over that etherstub, and then give > it > > an address eg 10.0.0.1 > > > > 3. For each zone, create a vnic over that etherstub, and assign the vnic > to > > that zone > > as an exclusive-ip device (so the addresses will be 10.0.0.2, 10.0.0.3 > etc) > > > > 4. Run haproxy or nginx (or something similar, whatever you're familiar > > with) in the global zone as > > a reverse proxy so it's listening on the system's main IP address, and > > proxies the traffic to the zone(s). > > This can be name-based websites (either from the host header for http or > > SNI for https), or port-based > > for things that can't handle routing based on names (eg ssh). > > > > 5. If you want to get out of the zone, set up simple nat in the global > zone > > and have each zone point > > it's default router at the global zone's address on the etherstub. But if > > you don't want the zones to > > be able to escape, don't handle nat. (In that case you might want a local > > dns server or forward proxy > > if you need to handle outbound requests.) > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] container with same ip
On Tue, Nov 14, 2023 at 5:24 PM Goetz T. Fischer wrote: > thanks but i don't think this is the right one for me. i want to run > services inside the container which > are available from the outside through the same ip as the host os. > or maybe i got the crossbow concept wrong? > The way I do this (on OmniOS and Tribblix, but conceptually it's the same on all Solaris/illumos variants): 1. Create an etherstub 2. In the global zone, create a vnic over that etherstub, and then give it an address eg 10.0.0.1 3. For each zone, create a vnic over that etherstub, and assign the vnic to that zone as an exclusive-ip device (so the addresses will be 10.0.0.2, 10.0.0.3 etc) 4. Run haproxy or nginx (or something similar, whatever you're familiar with) in the global zone as a reverse proxy so it's listening on the system's main IP address, and proxies the traffic to the zone(s). This can be name-based websites (either from the host header for http or SNI for https), or port-based for things that can't handle routing based on names (eg ssh). 5. If you want to get out of the zone, set up simple nat in the global zone and have each zone point it's default router at the global zone's address on the etherstub. But if you don't want the zones to be able to escape, don't handle nat. (In that case you might want a local dns server or forward proxy if you need to handle outbound requests.) > On Tue, 14 Nov 2023 12:12:51 -0500, John D Groenveld wrote: > > In message <20231114174007794762.87d3d...@r-a-c.de>, "Goetz T. Fischer" > writes: > >> first choice are zones. however as far as i've seen they have one > catch: they > >> need their own ip. > >> what i'm looking for though is more like a jail, just to not let > certain progr > >> ams reach outside of their > >> jail in case they get hacked. > > > > You might assign your zones private IP addresses with Crossbow NICs > > and virtual switches: > > https://www.usenix.org/legacy/event/lisa09/tech/full_papers/tripathi.pdf> > > > > John > > groenv...@acm.org > > > > ___ > > openindiana-discuss mailing list > > openindiana-discuss@openindiana.org > > https://openindiana.org/mailman/listinfo/openindiana-discuss > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] OI 2023.05 USB image doesn't work in QEMU
On Thu, Sep 14, 2023 at 7:36 PM James Madgwick wrote: > Hi, > > It seems that the "OI-hipster-gui-20230502.usb" image doesn't boot in > QEMU. With the command "qemu-system-x86_64 -m 3G --enable-kvm -hda > OI-hipster-gui-20230502.usb" I only get as far as below: > That's what you would expect. If you set up a usb device in qemu and attached the usb image to that (and used that as the boot device) it would probably work. Presenting the usb image as IDE disk 0 only confuses the installer. What's going on here is that the live boot archive only pulls the root filesystem from media, and it then has to mount /usr from the media, and that involves finding where the media is. The usb image is expecting to find itself on a usb device, so it will enumerate all the usb devices and try each in turn to see if it can find /usr (and it's own /usr, as well) on it. If that fails, then you fall into system maintenance mode as below. (And booting from the iso works, as you see, because *that* is looking for /usr on the cdrom ,which is where you've presented it.) > - > Loading unix... > Loading /platform/i86pc/amd64/boot_archive... > Loading /platform/i86pc/amd64/boot_archive.hash... > Booting... > NOTICE: MPO disabled because memory is interleaved > > WARNING: cpu1: CPUID.0xD returns invalid value: hw_low = 0, hw_high = > 0, xsave_size = 0, ymm_size = 0, ymm_offset = 0 > > Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms > 470.182.03 Fri Feb 24 03:20:38 UTC 2023 Hostname: openindiana > Remounting root read/write > Probing for device nodes ... > Preparing live image for use > Requesting System Maintenance Mode > (See /lib/svc/share/README for more information.) > Console login service(s) cannot run > > Enter user name for system maintenance (control-d to bypass): > - > Trying to use the newer virtual hardware (with -M q35) results in boot > loop. If I use OVMF UEFI firmware, then it instead hangs after the > NVIDIA message. This is using QEMU version 6.2.0 from Debian. > > If I instead try the CD/DVD ISO with "qemu-system-x86_64 -m 3G > --enable-kvm -cdrom OI-hipster-gui-20230502.iso", then I get to the > language selection prompt as expected. > > Is this some kind of problem with the USB image? I recall having this > same issue a couple of years ago and gave up on the USB image - what is > the difference between these two anyway? > > I've not tried the USB image with a physical PC yet - as I've always > just used DVD, though I don't see why it should behave any differently. > > Regards, > James > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] No sound output - no rights ?
On Thu, Aug 24, 2023 at 9:21 PM Stephan Althaus < stephan.alth...@duedinghausen.eu> wrote: > Hello! > > I have no audio output - anyone else, too ? > Are you logged in on the graphical console? The way this normally works is that the login display manager will change ownership of the audio devices to whoever is logged in on the console, using logindevperm(5) via di_devperm_login() and di_devperm_logout() - there's a patch 11-login-device-permissions.patch for lightdm that actually does this for you. > $ prtconf -d | grep -i audio > pci1028,17b1 (pciex10de,fbc) [NVIDIA Corporation GM107 High > Definition Audio Controller [GeForce 940MX]], instance #6 > pci1028,7b1 (pciex8086,a171) [Intel Corporation CM238 HD Audio > Controller], instance #5 > > $ sudo audioctl list-devices > Password: > audiohd#5 > audiohd#6 > > $ audiotest > Sound subsystem and version: SunOS Audio 4.0 (0x00040003) > Platform: SunOS 5.11 illumos-a98a962ff1 i86pc > /dev/sound/audiohd:5dsp: Permission denied > /dev/sound/audiohd:6dsp: Permission denied > > *** Errors were detected *** > > $ ls -l /dev/sound/audiohd:5dsp > lrwxrwxrwx 1 root root 49 Apr 11 2019 > /dev/sound/audiohd:5dsp -> ../../devices/pci@0,0/pci1028,7b1@1f > ,3:sound,dsp5 > > $ ls -l /devices/pci@0,0/pci1028,7b1@1f,3:sound,dsp5 > crw--- 1 root root 30, 81 Apr 11 2019 > /devices/pci@0,0/pci1028,7b1@1f,3:sound,dsp5 > > > There's no group having rights to access the device, nor "others" - is > this normal ? > > > Regards, > > Stephan > > > > > _______ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] what do pkg install phases mean?
On Sun, Mar 5, 2023 at 4:19 PM Till Wegmüller wrote: > Hi > > IPS works on images of the OS. And it does so in an Atomic way. Speed is > not the main goal. Stability is. > As a historical interlude, speed was very much a key focus of creating IPS to replace SVR4 packaging. One of the key criticisms of the old packaging system was that it was considered inordinately slow (one manifestation of this was that the development workflow for Solaris involved installing updated bits, SVR4 was too slow for impatient engineers, hence bfu as a hack, but that rendered the system unsupportable, but is what led to onu). The emphasis on performance can be seen in several areas of the design - downloading and updating just the files you need rather than whole packages; eliminating the overhead of maintaining the shared contents file. It's unfortunate that the original aim of improving performance got lost along the way. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Net booting OI Hipster using iPXE
Oh, network support (net-fs-root) got taken out at some point then. :-( Apparently here: https://github.com/OpenIndiana/slim_source/commit/d6a47f43379aae363a27ab88378daffcdf1901f6 "Drop support for generating auto-install images" That's an unfortunate casualty, as network booting was useful outside the narrow context of AI. On Thu, Sep 22, 2022 at 6:46 PM Till Wegmueller wrote: > Unfortunately not quite :) > > The Shell scripts in slim_source[0] at this time only support mounting > usb or cdrom but downloading the .zlib files into a tmpfs could be > added. PR's Welcome :) Let me know if you need help how to modify the > script. OmniOSCE has it's own handling of installation which supports -B > install_media > > -Till > > [0] > > https://github.com/OpenIndiana/slim_source/blob/oi/hipster/usr/src/cmd/slim-install/svc/media-fs-root#L70-L107 > > On 22/09/2022 14.03, Peter Tribble wrote: > > I *think* adding the following to the kernel line ought to do the trick > > > > -B install_media=http://1.2.3.4/media/directory/ > > > > where the url there is the directory containing the solaris.zlib file > (and > > for some releases, any other zlib files). > > > > You'll probably need to use the IP address. > > > > On Thu, Sep 22, 2022 at 4:59 PM Fox Martin > wrote: > > > >> Apologies if this has been answered elsewhere, I searched long and hard > >> and could not find what I really needed. > >> I have a working iPXE server with various Linux and Windows images > served > >> up over HTTP. > >> > >> For OI I tried the usual tricks of specifying the kernel and initrd and > >> then issuing a boot command but the installer always stops just after > >> "Probing for device nodes ... > >> Preparing live image for use" > >> and then > >> "Requesting System Maintenance Mode" appears. > >> > >> Are there kernel parameters or something else I need to issue to tell > the > >> OI system more about the location of boot folder, solaris.zlib etc? > >> > >> If there are links to documentation which solve this issue I would be > >> grateful to get them. > >> > >> thanks > >> Martin > >> > >> > >> ___________ > >> openindiana-discuss mailing list > >> openindiana-discuss@openindiana.org > >> https://openindiana.org/mailman/listinfo/openindiana-discuss > >> > > > > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Net booting OI Hipster using iPXE
I *think* adding the following to the kernel line ought to do the trick -B install_media=http://1.2.3.4/media/directory/ where the url there is the directory containing the solaris.zlib file (and for some releases, any other zlib files). You'll probably need to use the IP address. On Thu, Sep 22, 2022 at 4:59 PM Fox Martin wrote: > Apologies if this has been answered elsewhere, I searched long and hard > and could not find what I really needed. > I have a working iPXE server with various Linux and Windows images served > up over HTTP. > > For OI I tried the usual tricks of specifying the kernel and initrd and > then issuing a boot command but the installer always stops just after > "Probing for device nodes ... > Preparing live image for use" > and then > "Requesting System Maintenance Mode" appears. > > Are there kernel parameters or something else I need to issue to tell the > OI system more about the location of boot folder, solaris.zlib etc? > > If there are links to documentation which solve this issue I would be > grateful to get them. > > thanks > Martin > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] runtime/python-39 - missing file
On Tue, Jul 5, 2022 at 8:39 AM Predrag Zecevic via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > Hi all, > > It seems that runtime/python-39 does not deliver file: > ---8<-- > :; ls -al /usr/lib/amd64/pkgconfig/python3.pc > ls: cannot access '/usr/lib/amd64/pkgconfig/python3.pc': No such file or > directory > That file is mediated (because it can be delivered by all the different versions of python3). What does 'pkg mediator' say? I would expect you might need something like pfexec pkg set-mediator -V 3.9 python (assuming you want python3.9 to be the default python). > :; pfexec pkg fix runtime/python-39 > No updates necessary for this image. > > :; pkg contents runtime/python-39 | grep pkgconfig > usr/lib/amd64/pkgconfig/python-3.9-embed.pc > usr/lib/amd64/pkgconfig/python-3.9.pc > usr/lib/amd64/pkgconfig/python3-embed.pc > usr/lib/amd64/pkgconfig/python3.pc > > :; ls -al /usr/lib/amd64/pkgconfig/python*3* > -r--r--r-- 1 root bin 292 Jan 2 2020 > /usr/lib/amd64/pkgconfig/python-3.5.pc > -r--r--r-- 1 root bin 292 Nov 12 2020 > /usr/lib/amd64/pkgconfig/python-3.7.pc > lrwxrwxrwx 1 root root 13 Nov 12 2020 > /usr/lib/amd64/pkgconfig/python-3.7m.pc -> python-3.7.pc > -r--r--r-- 1 root bin 311 Jan 28 2021 > /usr/lib/amd64/pkgconfig/python-3.9-embed.pc > -r--r--r-- 1 root bin 285 Jan 28 2021 > /usr/lib/amd64/pkgconfig/python-3.9.pc > > How to fix that? > > Regards. > > -- > Predrag Zečević > predrag.zecevic.1...@googlemail.com > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Take me to your leader...
Hi! On Mon, Jun 6, 2022 at 9:12 PM Andrew Robinson wrote: > I'm new to this. I have a computer that meets all the requirements for > Hipster > and I am trying to boot from a DVD. I select the default option but then it > responds with: > > Console login service(s) cannot run > This means that OpenIndiana doesn't have a driver for the DVD drive. (The early boot uses BIOS routines, but as soon as the OS starts properly you need a real driver.) You don't say what hardware you have got, but one thing to check is that the DVD (and disk drive) isn't handled by some form of raid, such as Intel RST; if there's a raid mode, it needs to be disabled. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] ssh -X mate-terminal error
On Mon, Apr 25, 2022 at 6:52 PM Stephan Althaus < stephan.alth...@duedinghausen.eu> wrote: > Hello! > > This does not work anymore: > > ssh -X mate-terminal > I've often found it necessary to use -Y instead of -X, and that's probably necessary in this case. > .. from an OI box to another OI box. > > It works between a Linux vm (ubuntu impish) and the OI boxes in both > directions. > > Error Message is: > > $ ssh 192.168.2.52 -X mate-terminal > > (mate-terminal:4778): Gdk-ERROR **: 19:47:32.619: The program > 'mate-terminal' received an X Window System error. > This probably reflects a bug in the program. > The error was 'BadLength (poly request too large or internal Xlib length > erro'. >(Details: serial 178 error_code 16 request_code 72 (core protocol) > minor_code 0) >(Note to programmers: normally, X errors are reported asynchronously; > that is, you will receive the error a while after causing it. > To debug your program, run it with the GDK_SYNCHRONIZE environment > variable to change this behavior. You can then get a meaningful > backtrace from your debugger if you break on the gdk_x_error() > function.) > > It does not work on localhost either. > > I did not manage to place the variable GDK_SYNCHRONIZE=1 at the right > place it seems :-/ > > Any hints ? > > Stephan > > > > > ___________ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] geany editor?
On Mon, Mar 28, 2022 at 2:06 PM Andreas Wacknitz wrote: > Am 28.03.22 um 02:02 schrieb Jason Martin: > > > > agrellum@openindiana:~$ geany > > ld.so.1: geany: fatal: relocation error: file /usr/lib/libgeany.so.0: > > symbol g_get_os_info: referenced symbol not found > > Killed > > > It looks like geany needs a rebuild after our necessary downgrade of > glib to 2.62.6. Yes, it picked up g_get_os_info which came in with 2.64 > Alas, geany doesn't compile with this old glib version. > Hm. Builds and runs for me (on Tribblix) against glib 2.56.4 and geany itself only claims to want glib 2.32.0, so it ought to be OK. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] send/recv multi part zfs filesystem
Recursive zfs send and receive is a little non-obvious. It's a bit old but some notes I wrote: https://ptribble.blogspot.com/2012/09/recursive-zfs-send-and-receive.html On Sat, Jan 15, 2022 at 2:41 PM hput via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > setup HOST ubuntu-21.10 > Vbox VM OI/hipster > Installed from 2021.04.31 > updated 220114; uname -a >SunOS oi 5.11 illumos-c12f119108 i86pc i386 i86pc > > --- --- ---=--- --- --- > > Having trouble getting send/recv working to expectations. Maybe > expectations are wrong. > > I studied up a bit in oracle docs on send/recv But skimming thru > again just now I don't see a zfs multipart fs being send/recv'd. > Examples I find are all like: zfs send p0/one@snp | zfs recv p1/one > > Maybe using the -R operator but not sure if there are examples of > mutipart zfs fs being send/recv 'ed I hav'nt studied beyond the basic > send/recv. > > It looked much more complicated using the -R operator. > > So rsynced a bunch of data from a defunct windows machine using > SystemRescue on usb stick to get access to the data. > > I rsync'ed some 283 GB of data to ( an OI running from Vbox vm) to > p0/rhosts/2x2/F-win most of F: drive on the windows mach > > rsync -vv -rlptgoD --stats /sdd1/ OI:/rhosts/2x2/F-win > > Eventually produced 283 GB of data at OI:/rhosts/2x2/F-win. All in > the last segment F-win. > > --- --- ---=--- --- --- > > This may be a good place to explain that both file systems I worked on > are setting on separate raidz2 pools. > > pool p0 has 9 180GB discs > pool p1 has 6 190GB discs > > --- --- ---=--- --- --- > > I then decided to try my hand at send/recv but on the one machine zfs > fs to zfs fs. > > I created a similarly named p1/rhosts/2x2/F-win > > unmounted both fs and: > > zfs send p0/rhost/2x2/F-win@snap | zfs recv -F p1/rhosts/2x2/F-win > > And that started working but I had so much trouble from floundering > like a large trout for hours that I gave up on it waiting to learn > more. > > I did get the data moved from the win-mach to OI at least. > > --- --- ---=--- --- --- > > OK, so I have a question or two but will start with this: > > Is it even possible to send/recv a multi segmented zfs FS to another > multi segmented zfs fs. > > Or is it necessary to break it down so you are sending only 1 segment > at one time. > > so like: p0/rhosts@snap to p1/rhosts > > p0/rhosts/2x2@snap to p1/rhosts/2x2 > > And so on till all segments are sent. > > I'd be happy enough to learn the way just above (send 1 segment at a > time) is best but would really like to > > send/recv pool0/seg1/seg2/seg3 all at once to pool1/seg1/seg2/seg3 > > One thing that seems dubious is that I'd be faced with which snapshot > to use and I think that might rule the whole thing out. > > In fact I think I just talked myself out of the whole idea of sending all > 'segs' at once. > > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] SSDs for OI hipster ZFS root mirror?
On Sat, Aug 21, 2021 at 10:32 AM James wrote: > On 19/08/2021 23:58, Carl Brewer wrote: > > Further to this - is it worth disabling atime on the ZFS root pool > > that's on the SSDs? I don't imagine it's a lot of data, but it would add > > up over the years. > > Do you ever look at the access times? > On some occasions, this is very useful - you can work out what software installed is actually in use, and it's a way of detecting unexpected things running. We turn it off at $WORK because of two things - we run pkg verify every night, which means the access times are meaningless because we read every file daily anyway, and we hold a lot of regular snapshots and the atime writes end up consuming a significant amount of snapshot space. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Installing on Intel B460M chipset intel i5-10400F
On Fri, Aug 13, 2021 at 9:36 AM Toomas Soome via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > > > > On 13. Aug 2021, at 11:13, Carl Brewer wrote: > > > > The furthest I got was to use the minimal install 20210430 USB image, it > booted, albeit slowly. > > > > I ran the boot in verbose mode, it got as far as prompting for the type > of keyboard, which I confirmed as US English (47). > > > > > > This does sound like problems reading out the usb stick, can you try > different stick and slot, especiall if the system has usb 2 and usb 3. > As I recall, the keyboard prompt comes fairly late in the process - certainly after accessing the USB media to pull in /usr from it, so that would imply it's found the USB stick and is happy with it. The question then would be which smf service is getting stuck. By verbose, was that 'boot -v'? If so, you can track progress through smf with 'boot -m verbose' or 'boot -m debug' (I think - it's been a while). -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Update boot environment - VirtualBox now causes reboot
On Mon, Jul 5, 2021 at 10:14 AM russell wrote: > Good Morning, > > Created a new BE this morning 2021-07-05 08:43 with latest Hipster > download, system booted normally and was able to use all my normal > desktop applications. My last update was on the 2021-06-24 20:51 > > Started VirtualBox, the list of VMs appeared, I attempted to start a VM > this immediately reported an error, locked the system which eventually > rebooted. > Has VirtualBox on OpenIndiana been rebuilt to account for the following? If not, it presumably needs to be. https://illumos.topicbox.com/groups/discuss/T38b7ddf6f14b64a6/flag-day-for-kernel-modules-that-call-installctx > The VM that failed to start generated the following log entry > > > 00:00:02.755860 Installed Extension Packs: > 00:00:02.755870 Oracle VM VirtualBox Extension Pack (Version: 6.1.22 > r144080; VRDE Module: VBoxVRDP) > 00:00:02.756476 Console: Machine state changed to 'Starting' > 00:00:02.779587 Qt version: 5.12.10 > > This is completely different from a normal start found in VBoz.log.1 > > 00:00:00.712952 VirtualBox VM 6.1.22 r144080 solaris.amd64 (Jun 5 2021 > 12:16:14) release log > 00:00:00.712954 Log opened 2021-06-19T13:40:04.770473000Z > 00:00:00.712954 Build Type: release > 00:00:00.712958 OS Product: SunOS > 00:00:00.712958 OS Release: 5.11 > 00:00:00.712959 OS Version: illumos-76e6cd87e3 > 00:00:00.713027 DMI Product Name: MS-7C56 > 00:00:00.713053 DMI Product Version: 1.0 > 00:00:00.713056 Firmware type: failed - VERR_NOT_SUPPORTED > 00:00:00.713552 Host RAM: 32648MB (31.8GB) total, 22793MB (22.2GB) > available > 00:00:00.713555 Executable: /opt/VirtualBox/amd64/VirtualBoxVM > 00:00:00.713555 Process ID: 1752 > 00:00:00.713556 Package type: SOLARIS_64BITS_GENERIC (OSE) > 00:00:00.732246 Installed Extension Packs: > 00:00:00.732261 Oracle VM VirtualBox Extension Pack (Version: 6.1.22 > r144080; VRDE Module: VBoxVRDP) > 00:00:00.732887 Console: Machine state changed to 'Starting' > 00:00:00.752406 Qt version: 5.12.10 > 00:00:00.752418 X11 Window Manager code: 4 > 00:00:00.756364 GUI: UIMediumEnumerator: Medium-enumeration finished! > > I checked to make sure that the entries in /etc/system.d were present in > the new BE and it includes the /etc/system.d/smap with the entry "set > disable_smap = 1" and /etc/system.d/hma with "set hma_disable = 1". > > Is any one else experiencing the same problem or knows how to correct? > > TIA > > Russell > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] We need install images which work!
On Tue, May 4, 2021 at 3:53 PM Reginald Beardsley via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > I'm going to try one more time. > > > --- > The 2021.04.30 GUI Live Image has major problems. > > gparted(1m) dumps core > > the documentation mentioned on the first GUI installer screen is missing > > the release notes are missing > > the GUI installer fails to install the boot loader > > the text installer incorrectly states EFI installs are limited to 2 TB > > - > > All of these issues are also present in 2020.10.31 and possibly older > releases. This is a huge deterrent to new users. > Assuming new users hit them. Most users simply take the obvious direct route. If there's a button that says "Install" they'll hit that and completely ignore anything and everything else that might be present. Based on that simple workflow, the current OI gui installer (which I haven't used for ages, but still looks like the original OpenSolaris installer) is good enough. It's pretty much on a par with other distro installers (and I've run dozens of them this year already). It could be a lot quicker, it could be more responsive, but it's not noticeably worse in terms of useability or functionality or reliability than many of the current crop of Linux installers - worse than some, better than some. That's not necessarily a ringing endorsement, but the point is that for most of the target audience the sky isn't falling and I think it echos what someone (Judah?) said - stick to the straight and narrow and life will be easier. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] How to mount usb devices?
On Mon, Mar 29, 2021 at 2:06 PM Gary Mills wrote: > On Sun, Mar 28, 2021 at 04:24:58PM -0500, Gary Mills wrote: > > On Sun, Mar 28, 2021 at 08:38:33PM -0700, irixuzer wrote: > > > I'm new to the list. Ive been trying to mount usb devices. After the > first > > > time, which was successful, I've been getting the following: > > > > If you are using the dbus service, the usual case with OI, USB devices > > get mounted automatically. You don't have to do anything. The FMRI > > of that service is svc:/system/dbus:default . > > I see I made a mistake in this message. It's the hal service that > mounts USB devices automatically. The FMRI is: > > svc:/system/hal:default > > This service is present on all my OI systems, but then I installed > them initially with the GUI installer. I don't know if the hal > service is installed by the text installer. > Not quite, hal handles device insertion/removal and wires up the device tree. It's rmvolmgr that handles the mounting piece. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] How to mount usb devices?
On Mon, Mar 29, 2021 at 12:42 PM Udo Grabowski (IMK) wrote: > > On 29.03.21 13:19, Stephan Althaus wrote: > > On 03/29/21 12:53 PM, Udo Grabowski (IMK) wrote: > >>> > >>> Do you have the /etc/logindevperm file with the entries listed above ? > >>> /dev/vt/7 is owned by you, and /dev/vt/console_user is linking to that, > >>> so those entries should work. These are used by the login process, or > >>> gdm. If you use lightdm, I'm not shure if that supports logindevperm > >>> (somewhere the "light" has its origin ...). > >>> > >> > >> It could also be that this depends on this gdm setting: > >> > >> # cat /etc/gdm/custom.conf > >> > >> [security] > >> CheckDirOwner=false > >> > >> You have to logout and restart gdm (svcadm restart gdm) to have this > >> activated > > > > Yes, i am using lightdm as this is the default. > > > > There seemed to be an attempt to integrate something like that into > lightdm: > > < > https://openindiana.org/pipermail/openindiana-discuss/2014-October/016590.html > > > > Unfortunately, the link therein is gone. Gabriele ? > I would assume that it's this patch that's already applied to the OI lightdm: https://github.com/OpenIndiana/oi-userland/blob/oi/hipster/components/desktop/lightdm/patches/11-login-device-permissions.patch I have a similar patch in Tribblix, although I'm using SLiM. But generally that fix to call di_devperm_login and di_devperm_logout should be all that's needed for the desktop login to get the permissions correct. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Hipster 2020.10 text installer ISO Wow!!!!
On Tue, Mar 23, 2021 at 3:19 PM Reginald Beardsley via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > I thought I'd make another stab at this. I updated to the 3.61 BIOS from > HP's web site. > > With milestone single user or multi-user using the "txt install" ISO > "prtconf -v" runs fine if IDE is selected, however, in RAID+AHCI it won't > run as shown in the photo. > The photo indicates that the CDROM can't be found - it's hidden behind the RAID and is inaccessible to OI. Given that most of the OS is still on the media, which you can't get to, essentially nothing will work. If you want to poke around further in this mode, then you'll need something that can either find its media via another route (such as booting a USB image or using PXE or iPXE), or try one of the distributions that don't need access to the media to finish the boot (which means current OmniOS or Tribblix). > scanpci is not available nor is find which leaves me a bit mystified as to > what we are running in the image above. > > Despite statements on HP's web site, it does appear that there is more > than one version of the mainboard. > > Reg > > > > > On Wednesday, March 3, 2021, 09:08:02 PM CST, Gary Mills < > gary_mi...@fastmail.fm> wrote: > > On Wed, Mar 03, 2021 at 07:57:39AM -0600, Gary Mills wrote: > > > > I looked around for my notes on the first install of OI on my Z400, > > but couldn't find anything. Maybe it was that easy that I didn't > > bother making notes. I do recall that only needed to follow the first > > part of the instructions. The main thing is the add_drv and rebuilding > > the boot archive. There was only one BE, of course. > > I found my notes. They were from Febuary 2018. Here they are, in > point form: > > o Changed controller to AHCI+RAID in BIOS > o Booted OI hipster-20171031 USB stick > o Started a shell window > o Got a root shell with `sudo ksh' > o Found `RAID controller' in `prtconf -v' output > o Ran: update_drv -a -i 'pci8086,2822.103c.1309.0' ahci > o Added line to driver_aliases: > ahci "pci8086,2822.103c.1309.0" > o Ran the installer > o Checked new BE: > # mkdir /tmp/a > # beadm mount openindiana /tmp/a > # wc -l /tmp/a/etc/driver_aliases /etc/driver_aliases > 1772 /tmp/a/etc/driver_aliases > 1772 /etc/driver_aliases > o Files were the same > o Rebuild the boot archive: > # bootadm update-archive -v -R /tmp/a > o No output > o Umount: > # beadm unmount openindiana > o Rebooted from the installer GUI > o Normal boot of new BE > > > -- > -Gary Mills--refurb--Winnipeg, Manitoba, Canada- > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Hipster 2020.10 text installer ISO Wow!!!!
On Thu, Mar 4, 2021 at 4:31 PM cretin1997 via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > > Unrelated but it's an interesting piece of information. Illumos still > depends on proprietary closed source tools from the Sun's era to build > itself. Check the documents about how to build illumos-gate, it's still > needs onbld. > While illumos still has some old bits of closed binaries, onbld isn't one of them - its source is included in illumos-gate itself. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Hipster 2020.10 text installer ISO Wow!!!!
On Thu, Mar 4, 2021 at 3:45 PM Reginald Beardsley via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > OK I just made several attempts: > > First attempt was with 2020.10 GUI install ISO. That was forced to > maintenance mode and was unable to locate libpcidb.so.1 when I ran prtconf > -v. > > It completely ignored the 2019.04 USB image even with it selected via F9. > The 20202.10 text ISO also fails for lack of libpcidb.so.1 > > I'll patch the libpcidb.so.1 failure when I have some time. I think that's > more productive than making a 2017.10 USB image to test. > > Unfortunately, this is another example of poor QC of the release ISO > image. How hard is it to verify that all the libraries needed in single > user mode are actually installed? A shell script to do that should be an > integral part of the process for building an ISO image. > Only you're not even in single user mode at that point. The dropping to maintenance mode failure indicates that the ISO has failed to boot; in particular it hasn't mounted /usr, which is loaded from the ISO later. This usually indicates that the OS can't find the CD at all. (There's a very minimal subset of files from /usr on the root filesystem that are actually needed to boot.) I've seen this failure to find the CD drive on laptops fairly often, they have a habit of wiring up the CD drive rather oddly. Also my Dell desktop couldn't see the CD at all while it was in the fake raid mode, i had to disable that to get it to work at all. But that failure to be able to locate the CD is the likely root cause of missing libraries, it's nothing to do with release engineering. Both OmniOS and Tribblix don't have this failure mode as they don't have the split boot. We just get the bootloader to slurp the entire OS into memory. It's the sort of trick that isn't really practical for the GUI boot, though. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Does OI compress the zpool by default?
On Tue, Mar 2, 2021 at 4:16 PM Chris wrote: > On 2021-02-28 07:38, cretin1997 via openindiana-discuss wrote: > > zfs get compression and zfs get compressratio results seemed to tell > that > > it's not > > on by default. But, why? > The reason is or should be, because of its potential cost. That is; > compression can consume a great deal more CPU cycles. > ZFS compression is a good thing. You win by having to move less data through the entire I/O stack. Originally, compression was off by default because 15 years ago there were still incredibly slow systems still being used. Worse, zfs compression was single-threaded, which was a bit of a nightmare on something like the old coolthreads boxes. Neither are true any more. Over the years, CPU capacity has increased much faster than I/O performance, and we have better algorithms - and better optimized code - to play with. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Tribblix and Pale Moon
On Tue, Mar 2, 2021 at 11:31 AM cretin1997 via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > I don't know the email address of Peter Tribble so I post here hopefully > he will see it. > > Just tried your OS on VirtualBox. It's amazing, a full desktop that's only > 2GB in total. The OI's VM should be 4x than that. > > I installed the Pale Moon browser and I noticed you did build it with > official branding. The terms under which this is permitted are covered by the redistribution license http://www.palemoon.org/redist.shtml If you can point to something that violates this (specifically Point 8) then let me know and I'll fix it to bring it into compliance. > If you were granted by the Pale Moon developers to use their official > branding, please dismiss the following part of this mail. > > If you weren't, you are in trouble. Please build Pale Moon with official > branding off. Please have a look at this: > > https://forum.palemoon.org/viewtopic.php?f=65&t=26065 > > In short, remove these lines: > > # Please see https://www.palemoon.org/redist.shtml for restrictions when > using the official branding. > ac_add_options --enable-official-branding > export MOZILLA_OFFICIAL=1 > > from the .mozconfig file describe here: > > https://developer.palemoon.org/build/sunos/ > > and you will save yourself a lot of trouble. > > Sincere. > > > Sent with ProtonMail Secure Email. > > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Resizing?
On Mon, Mar 1, 2021 at 8:19 PM L. F. Elia via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > I don't need to do this, so this can be an academic discussion: I can > resize a drive in a hypervisor, but can I resize a zpool afterwards or are > we talking exporting, creating a new pool and importing on it? > Thank you for all the excellent info you provide > If you enlarge the underlying drive, then you can tell zfs to make use of the available space with a command like: zpool online -e rpool c2t0d0 where you have to give the pool name and the name of the device. You can only enlarge a pool this way; shrinking a device isn't possible. And you may need a reboot for the guest to see the changes (eg on AWS, I have to reboot instances to pick up the size changes of EBS volumes). -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Anyone running OI on VPS?
On Thu, Feb 25, 2021 at 3:47 PM cretin1997 via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > Do you run OI on VPS or dedicated server? > I run other illumos distributions (OmniOS, Tribblix) on dedicated hardware, VPS providers, and the cloud. > Does OI suitable to run on a VPS at all? > Certainly. Like anything, provided you meet the requirements. > It seems normal cheap VPS plan of many host providers will be not enough. > OI takes much more disk space than Linux because of Boot Environments. That varies. You can trim boot environments, so they don't take that much space. And I find Linux distributions generally to chew through a lot of space, especially as you get all sorts of odd packages installed unnecessarily because they're thought to be required (turning off recommends can help). > So it seems I will need to upgrade the storage plan. OI also needs more > memory to work because of pkg so I will need more memory, too. This end up > with a not cheap at all VPS whose resources (cores, memory, storage space) > will be same same as my home PC. > > If it's so, does a dedicate server should be more preferred? > Not really, it's true that you might struggle with the smaller VPS sizing, but even if you have to avoid those you can still go a lot smaller than a dedicated server. > But if it's the case, why someone consider running OI in the first place > instead of Linux since Linux is a much cheaper option? > As always, choose whichever option best serves your needs. > I'm talking about normal KVM host providers here. > The things that can cause problems are (a) if they force virtio-scsi, which we don;t yet have, or (b) if they don't offer some sort of install from ISO option, which can be worked around in some cases. > I have no idea about the cloud (AWS, GCE, Azure,...) at all. > I'm not aware of AMIs for OI, but they're not hard to makes, so AWS wouldn't be a problem. I've run OmniOS on Azure; I've not run any illumos on GCE myself, but I think it's possible. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] after Installation, no X11?
g mate_install is on of those? > >> > >> I will see for a faster desktop so, X11 with twm is what I am > >> running on FreeBSD-Boxes, CDE on AIX and so. I saw, there is > >> a cde available so I will go to installing that and give it a > >> try. > >> > >> Tanks a lot, > >> Rolf > >> > >> > >> ___________ > >> openindiana-discuss mailing list > >> openindiana-discuss@openindiana.org > >> https://openindiana.org/mailman/listinfo/openindiana-discuss > > > > > > > > > > ___ > > openindiana-discuss mailing list > > openindiana-discuss@openindiana.org > > https://openindiana.org/mailman/listinfo/openindiana-discuss > > > > ___ > > openindiana-discuss mailing list > > openindiana-discuss@openindiana.org > > https://openindiana.org/mailman/listinfo/openindiana-discuss > > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] The name of a new (?) OS
On Fri, Feb 5, 2021 at 5:20 PM Apostolos Syropoulos via openindiana-discuss wrote: > Some days ago someone posted a message about an OS (not Q4OS) > does anyone remember the name of the OS? If I recall correctly > its name started with "R". I looked into the list archive but > I could not find the name. > ReactOS? -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] A dumb idea
On Fri, Jan 29, 2021 at 3:51 AM Hung Nguyen Gia via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > What about just import the FreeBSD Ports system and like DragonflyBSD uses > transformations scripts (https://github.com/DragonFlyBSD/DeltaPorts) to > transform it into illumos-ports something like this ( > https://github.com/DragonFlyBSD/DPorts) and enjoying the large amount of > software we will never able to port by our own with little effort? > That doesn't address the fundamental problem: you still have to do the port to illumos. In general, those of us porting stuff tend to look at what's in other systems to see if and how they've solved similar problems. But importing the system as a whole means you're now having to deal with not only the idiosyncracies of the software you're interested in but also having to deal with problems from the alien system itself. To be successful, you require buy-in from that system - so something like pkgsrc is fine, for example. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] When will AMD XIII CPUs be supported
On Sat, Jan 23, 2021 at 2:22 AM Gary Mills wrote: > On Fri, Jan 22, 2021 at 11:39:20AM -0800, Chris wrote: > > I don't have a specific setting for ahci. As a rule, when available, > > I choose that. But it's simply not available. > > > > Here's the output for mine > > > > ~ > > 11:35am Fri, 22 > > OIDEV# prtconf -D > [...] > > pci1462,7309, instance #0 (driver name: nge) > > pci1462,7309, instance #0 (driver name: nv_sata) > > disk, instance #2 (driver name: sd) > > pci1462,7309, instance #1 (driver name: nv_sata) > > That's peculiar: there are two drivers attached to the same hardware > device. I've never seen that before. > That's normal. pci1462,7309 is the MCP61 PCI bridge that multiple devices are connected to. Consider this on my Intel system: # prtconf -D ... pci1028,5a4, instance #0 (driver name: e1000g) pci1028,5a4, instance #0 (driver name: ehci) hub, instance #1 (driver name: hubd) pci1028,5a4, instance #1 (driver name: audiohd) pci1028,5a4, instance #1 (driver name: ehci) hub, instance #2 (driver name: hubd) You get a clearer picture, expanding the individual devices, with -d: # prtconf -dD ... pci1028,5a4 (pciex8086,153a) [Intel Corporation Ethernet Connection I217 -LM], instance #0 (driver name: e1000g) pci1028,5a4 (pciex8086,8c2d) [Intel Corporation 8 Series/C220 Series Chi pset Family USB EHCI #2], instance #0 (driver name: ehci) hub, instance #1 (driver name: hubd) pci1028,5a4 (pciex8086,8c20) [Intel Corporation 8 Series/C220 Series Chi pset High Definition Audio Controller], instance #1 (driver name: audiohd) pci1028,5a4 (pciex8086,8c26) [Intel Corporation 8 Series/C220 Series Chi pset Family USB EHCI #1], instance #1 (driver name: ehci) hub, instance #2 (driver name: hubd) -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] A mailing list for Tribblix
On Sat, Jan 23, 2021 at 5:43 AM Hung Nguyen Gia wrote: > Sorry for posting here. I know it's the wrong place but it's the place > that I most likely could meet P. Tribble. > > Mr. Tribble, as you already has a Google account, and even has two blogs > running, may I ask you to create a google group for Tribblix? > > It's doesn't need much effort to create and maintain it. But Tribblix > users will have a place for themselves, don't have to post here anymore. > While I'm in some ways flattered that it might be thought worthwhile for Tribblix to have its own mailing list, it's not at all clear to me that adding more isolated islands of communication to the community is a good thing. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] OmniTribblix, vpnc and audio mixer
On Thu, Jan 21, 2021 at 9:13 AM Tony Brian Albers wrote: > Hi Guys, > > Pardon me for posting here, but I haven't been able to find a mailing > list for Tribblix. > There isn't one. Traffic would likely be light... > Since sound is not working properly in OI, I've decided to try out > OmniTribblix. I'm running m24 > Sound is likely to be the same - all the drivers and underlying plumbing come from illumos. OI may have more utilities and libraries to talk to it (it might have pulseaudio to layer on top, I think) and some desktop integration. Tribblix used to have the Xfce mixer applet, but that seems to have been dropped by the Xfce project. > Has anyone found out how to get vpnc into tribblix? > Never tried. I had this working on Solaris 10, but that shows how long ago I last had need of it. > And: I cannot find an audio-mixer of any kind -am I doing something wrong? > We used to have sdtaudioctl and mixerctl. For whatever limited manipulations I need (and to be honest it's usually to mute stupid autoplaying things on web pages) then audioctl can be used. The underlying sound system is quite sophisticated; we don't have a great deal that exposes that to the end user. > TIA > > /tony > -- > Tony Albers - Systems Architect - IT Development Royal Danish Library, > Victor Albecks Vej 1, 8000 Aarhus C, Denmark > Tel: +45 2566 2383 - CVR/SE: 2898 8842 - EAN: 5798000792142 > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Shell to use?
> As I have said on this list: pkgsrc bootstrap on OI is 4x slower than on > FreeBSD. Building packages also that slower. > > I think the problem is of the shell. Because I see it checking for > something very slowly. > Process creation (fork/exec) has been known to be slow on Solaris (and thus illumos) for years. Unfortunately things like configure and an awful lot of build tools spawn a huge number of processes - this is horrendously inefficient on all platforms, but the slowdown on the platforms where most of this stuff originates isn't so obviously bad that there's motivation to fix it. (Generally, Solaris and illumos are optimized for well-behaved software, Linux is optimised for badly behaved software. Unfortunately, there's a lot more badly behaved software out there than good stuff.) > From your own experience, which shell is the fastest? > I tried replacing sh with dash as the system shell, and it gave a decent improvement in things like boot times. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Could you update your sysroot, jclulow?
On Sun, Jan 17, 2021 at 11:15 AM Hung Nguyen Gia via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > I wanted to use your sysroot here to build a cross compiler for OI from > Linux: > > https://github.com/illumos/sysroot > > But it seemed you didn't include the C++ part. I couldn't find them. There > is no C++ headers, e.g: cmath, iostream,... nor the C++ standard library > itself. > Those are supplied by the toolchain, in our case usually gcc. And you would normally need to ensure that the versions in the sysroot are compatible (for whatever definition of compatible you're interested in) with the toolchain you're using. > This sysroot is also a bit old, too. > A sysroot can be used for a variety of uses. This one is targetted at building software on a current host that will still run correctly on an older system. (This is also the way that Oracle built Java for Solaris, I believe, using a Solaris 10 sysroot on a Solaris 11 build system so that the resulting binaries will still run on Solaris 10.) Generally I would expect anyone using a sysroot to have a pretty specific target in mind. It's not that hard to create, either simply copying an existing system that has the bits you want, or using packaging - pkg has the image-create functionality designed for exactly this purpose. Please correct me if I'm wrong. > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Will OpenIndiana switch to SquashFS?
On Sat, Jan 9, 2021 at 6:37 AM Hung Nguyen Gia via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > Most Linux distributions now employing SquashFS for their live system so > they are blazing fast even though being run from a slow USB 2.0 stick. I'm > posting this mail on one of such live system. I found OpenIndiana is using > another technology: > https://ptribble.blogspot.com/2012/10/those-strange-zlib-files.html > > Does this technology comparable to SquashFS? And if SquashFS is better, is > there any plan to switch to SquashFS? > The illumos distributions have alway used a compressed filesystem for the live image. There's no difference to squashfs in that respect - we're just normally compressing the whole image. There's no benefit to changing to a different scheme that's essentially identical. We do actually have a direct equivalent to squashfs already - dcfs. I've not seen it used much, although it was used on SPARC to compress the files in the boot archive. (On x86, the bootloader can read a compressed boot archive, so you don't need it there.) -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Merging OI + OmniOS? (And OpenZFS vs ZFS)
On Wed, Dec 30, 2020 at 11:39 AM Kalle Anka via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > A question. I am sure this has been discussed before, but maybe it is time > to update the discussion? :) > As I have understood it, OmniOS is a robust server OS with no good desktop > environment. It is possible to install X11, though: > https://geekblood.wordpress.com/2017/10/26/installing-x11-and-a-desktop-environment-on-omnios/ > Would it be possible to merge OI and OmniOS? Say, try to transform OI into > a desktop environment ontop OmniOS? Both OSes share lot of common ground, > so technically it would not be impossible to for instance port MATE to > OmniOS? > OpenSolarish is a bit fragmented as of now. There are several developers > working on their own stuff, with different IPS repos, etc. If we could > unite some of the work, OpenSolarish would benefit as a whole. > Not necessarily. And that's the problem. Fundamentally, it's very easy if all you have to do is maintain a single-purpose distribution and have total control of everything that goes in and out. Once you try and build a general-purpose distribution, you need some way to coordinate between different groups working with different cadences and trying to produce different products for different types of user. That generates friction and conflict, and requires additional manpower (aka waste, in lean terminology). The reality is that there isn't a huge amount of duplicate work - distros can pick and choose whatever they want/need from others *as they see fit* and *to their own timescales*. > OmniOS would get new developers, a good desktop environment, and also lot > of users coming from OI. > OI would benefit from all the work the OmniOS developers are doing in > getting a stable and robust server OS, so it would free up OI developers to > more quickly advance desktop environment. Lot of synergy effects and > win-win. OmniOS does the server backend, and OI do the desktop frontend. > This lessens the burden for the developers. The more developers, the better. > The situation is similar to two competing camps developing ReactOS - the > open sourced WindowsXP clone. Why not unite the two teams instead? It would > be much less work for the OI developers if we could build on the excellent > work from OmniOS team, and vice versa. Remember, we all love OpenSolarish > and it would be better to have one single strong distro, than several > competing distros. FreeBSD has only one distro and that is better than the > Linux fragmentation mess today. There are only so many OpenSolarish > developers, and our community would benefit from one single strong distro. > Or, have I missed something? Like, for instance, do the illumos community > also suffer from "Not Invented Here" syndrome like some other OSes? The > only truly innovative OS is Solaris, and lets keep it that way by > consolidating and uniting our teams so we can make faster progress? I am > interested in trying to help the community too. Because Solaris is the best > OS! :) > > > BTW. Earlier I reported that OpenZFS renders ZFS disks unusable: If you > import a zpool v28 and ZFS v5, into Linux using OpenZFS, then Solaris 11.3 > cannot import the zpool again. Someone suggested that Linux OpenZFS renames > the disks as "/dev/sda" instead of "/dev/c0t0d0" in the zpool. So this > problem could maybe be fixed by importing and exporting the zpool by a > illumos distro which would automatically rename the "/dev/sda" entries to > "/dev/c0t0d0" back which means Solaris v11.3 could import the zpool. I have > not tested this yet, but it sounds plausible? > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] OI new user questions
On Sat, Dec 5, 2020 at 4:02 PM Kalle Anka via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > I am a long time Solaris user that tried out Linux for a couple of years, > namely Ubuntu LTS. Alas, the Ubuntu updates caused numerous problems, in > some cases causing a reinstall. At the end I switched to Ubuntu 2020.10 > hoping it would be more stable. It was not. > > Another problem is that OpenZFS v0.8.4 renders Solaris 11.3 disks > unusable. Try this: > -Create a ZFS disk in Solaris 11.3 (using zpool version 28, and zfs > version 5) > -Import the disk into Ubuntu 2020.10 using OpenZFS v0.8.4 > -Copy data to the zpool using Linux zfs send recv > -Import the zpool back into Solaris 11.3 - This will fail. Solaris says > the disk is UNAVAIL and I need to use a backup to restore data. > So if you use OpenZFS to import a zpool, chances are you cannot import the > disk back into Solaris. > As I recall, this is because ZFS embeds the labels of the devices used into the pool, so you have to allow for them being in an alternate format. I can't recall offhand whether we fixed that in illumos. > 2) Does OI support Sunray? (Solaris 11.4 does not support Sunray) > Several people have got this working successfully - it was even displayed at the illumos booth at FOSDEM this year. > 5) I hope there is a mpt sas driver, because there was one in Solaris > 11.3. Can anyone confirm? I have this LSI2008 SAS card for my JBOD raidz3 I > would like to use. > Yes. > 6) Virtualbox, does it work fine on OI? Is it problematic? > You have to use the VirtualBox that comes with OI. The Solaris version you can download from virtualbox.org is no longer compatible and simply won't work. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Using distribution constructor problems
to > >>>>pkg:// > >>>> openindiana.org/install/distribution-constructor@0.5.11-2018.0.0.1043 > >>>> Reason: This version is excluded by installed incorporation > >>>> consolidation/install/install-incorporation@0.5.11-2018.0.0.1040 > >>>> Reason: This version is excluded by installed incorporation > >>>> consolidation/userland/userland-incorporation@0.5.11-2018.0.0.11763 > >>>> > >>> > -- > >>>> > >>>> I followed: > >>>> > >>>> mkdir distro-const > >>>> cd distro-const > >>>> pfexec pkg install install/distribution-constructor > >>>> cp /usr/share/distro_const/slim_cd/slim_cd_x86.xml . > >>>> pfexec distro_const build slim_cd_x86.xml > >>>> > >>>> but cannot get past the 3rd step. > >>>> > >>>> Did I miss something? > >>>> Cheers, > >>>> Lonnie > >>>> ___ > >>>> openindiana-discuss mailing list > >>>> openindiana-discuss@openindiana.org > >>>> https://openindiana.org/mailman/listinfo/openindiana-discuss > >>> > >>> > >>> > >>> ___ > >>> openindiana-discuss mailing list > >>> openindiana-discuss@openindiana.org > >>> https://openindiana.org/mailman/listinfo/openindiana-discuss > >>> > >> > > ___ > > openindiana-discuss mailing list > > openindiana-discuss@openindiana.org > > https://openindiana.org/mailman/listinfo/openindiana-discuss > > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] PATH=/usr/xpg4/bin /bin/ksh -c "rm -f"
On Mon, Jun 17, 2019 at 11:15 AM James wrote:
> Hello, please can someone help me understand this xpg4, ksh, rm oddity?
> Normally: rm with no args complains and adding -f makes it accept no args.
>
> $ rm
> usage: rm [-fiRr] file ...
> $ rm -f
> $
>
> On OpenSolaris [but not 10, 11.3] with /usr/xpg4/bin in the path it
> behaves differently but only with /bin/ksh or /bin/sh (ksh93)
>
> $ PATH=/usr/xpg4/bin /bin/ksh -c "rm -f"
> Usage: rm [-cFdfirRuv] file ...
> ...
> $ PATH=/usr/xpg4/bin /bin/truss -f /bin/ksh -c "rm -f" |& grep rm
> 4436: lstat("/usr/xpg4/bin/rm", 0xFC7FFFDFDCE0) = 0
> Usage: rm [-cFdfirRuv] file ...
> ...
>
I can't see a built-in rm or alias in ksh and if it were I would expect
> -f to have the same effect. The path must be making ksh act
> differently. What am I not understanding here?
>
In ksh, /usr/xpg4/bin/rm is a builtin.
% ksh builtin | grep /rm
/bin/rmdir
/usr/ast/bin/rm
/usr/ast/bin/rmdir
/usr/xpg4/bin/rm
You can see this from your ouput above too: for one thing, that's the
ksh builtin usage message; for another truss doesn't show an execve.
You appear to be ending up in this code:
http://src.illumos.org/source/xref/illumos-gate/usr/src/lib/libcmd/common/rm.c#395
(And if you add '-u libcmd::' to your truss invocation you can see it
calling the b_rm
function there.)
--
-Peter Tribble
http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/
___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] installing pthreads for compiling
On Thu, Mar 28, 2019 at 4:46 PM Lonnie Cumberland wrote: > Hi Peter, > > Just wanted to let you know that I was able to get your "tslim" to compile > for which it produced the slim executable. > > Any suggestion on testing and installation? > It comes with SMF integration, so make install svcadm restart manifest-import svcadm enable slim ought to work. The configuration is slightly tailored to Tribblix, but you could get slim.conf to run any script of your choosing. > Cheers, > Lonnie > > On Tue, Mar 26, 2019 at 2:20 PM Peter Tribble > wrote: > > > On Tue, Mar 26, 2019 at 5:51 PM Lonnie Cumberland > > wrote: > > > > > Hello Andreas. > > > > > > Thanks for your response. > > > > > > Unfortunately, although I just installed the "system/header" and > > > "build-essential" packages, when I tried a clean "cmake" on the code, > it > > > still says that it cannot find "pthread.h" for some reason. > > > > > > > The file CMakeFiles/CMakeOutput.log ought to have more verbose details of > > why it failed. > > > > > > > I was just trying to build a project that I found on Github "slim" ( > > > https://github.com/iwamatsu/slim) which is a very lightweight login > > > greeter that I wanted to try as my understanding is that the > > > "LightDM+GTK-Greeter" needs a lot of support libraries and has a large > > > footprint although I am not completely sure about this part and may > have > > to > > > try it. > > > > > > > You may find https://github.com/ptribble/tslim (which is a Solarish port > > of > > SLiM) > > to work better. (Although if you can't find pthread.h for regular slim, > my > > fork > > shouldn't be any different. But mine is designed for use in illumos, and > is > > used > > as the login manager in Tribblix.) > > > > > > > Cheers, > > > Lonnie > > > > > > > > > > > > On Tue, Mar 26, 2019 at 1:07 PM Andreas Wacknitz > > > wrote: > > > > > > > Am 26.03.19 um 16:57 schrieb Lonnie Cumberland: > > > > > Hello, > > > > > > > > > > I have setup Openindiana from the live DVD just now and have > > installed > > > > > gcc-8, cmake, and some libraries as I want to compile a small > > > > application, > > > > > but am getting a message that "pthreads,h - not found" > > > > Are you sure that it's pthreads.h? I only see pthread.h in my build > > env. > > > > You could check whether the > > > > package "system/header" is installed. We also have the package > > > > "build-essential" which should install > > > > a foundation of what is needed to compile software. > > > > > > > > Regards, > > > > Andreas > > > > > > > > > > I looked on the OpenIndiana package library online and found > > > > > > > > > > library/pth@2.0.7,5.11-2018.0.0.0:20180211T123900Z > > > > > < > > > > > > > > > > http://pkg.openindiana.org/hipster/info/0/library%2Fpth%402.0.7%2C5.11-2018.0.0.0%3A20180211T123900Z > > > > > > > > > > > > > > > and > > > > > x11/library/libpthread-stubs@0.3,5.11-2018.0.0.0:20180212T053005Z > > > > > < > > > > > > > > > > http://pkg.openindiana.org/hipster/info/0/x11%2Flibrary%2Flibpthread-stubs%400.3%2C5.11-2018.0.0.0%3A20180212T053005Z > > > > > > > > > > > > > > > for which I tried to install them but it said that no update was > > > > necessary > > > > > for these images. > > > > > > > > > > Can someone please tell me how I might be able to fix this? > > > > > Thanks, > > > > > Lonnie > > > > > ___ > > > > > openindiana-discuss mailing list > > > > > openindiana-discuss@openindiana.org > > > > > https://openindiana.org/mailman/listinfo/openindiana-discuss > > > > > > > > > > > > ___ > > > > openindiana-discuss mailing list > > > > openindiana-discuss@openindiana.org > > > > https://openindiana.org/mailman/listinfo/openindiana-discuss > > > > > > > > > > > > > -- > > > Best Regards, > > > > > > Lonnie T. Cumberland, PhD. > > > Email: lonnie.cumberl...@outstep.com > > > ___ > > > openindiana-discuss mailing list > > > openindiana-discuss@openindiana.org > > > https://openindiana.org/mailman/listinfo/openindiana-discuss > > > > > > > > > -- > > -Peter Tribble > > http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ > > ___ > > openindiana-discuss mailing list > > openindiana-discuss@openindiana.org > > https://openindiana.org/mailman/listinfo/openindiana-discuss > > > > > -- > Best Regards, > > Lonnie T. Cumberland, PhD. > Email: lonnie.cumberl...@outstep.com > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] installing pthreads for compiling
On Tue, Mar 26, 2019 at 5:51 PM Lonnie Cumberland wrote: > Hello Andreas. > > Thanks for your response. > > Unfortunately, although I just installed the "system/header" and > "build-essential" packages, when I tried a clean "cmake" on the code, it > still says that it cannot find "pthread.h" for some reason. > The file CMakeFiles/CMakeOutput.log ought to have more verbose details of why it failed. > I was just trying to build a project that I found on Github "slim" ( > https://github.com/iwamatsu/slim) which is a very lightweight login > greeter that I wanted to try as my understanding is that the > "LightDM+GTK-Greeter" needs a lot of support libraries and has a large > footprint although I am not completely sure about this part and may have to > try it. > You may find https://github.com/ptribble/tslim (which is a Solarish port of SLiM) to work better. (Although if you can't find pthread.h for regular slim, my fork shouldn't be any different. But mine is designed for use in illumos, and is used as the login manager in Tribblix.) > Cheers, > Lonnie > > > > On Tue, Mar 26, 2019 at 1:07 PM Andreas Wacknitz > wrote: > > > Am 26.03.19 um 16:57 schrieb Lonnie Cumberland: > > > Hello, > > > > > > I have setup Openindiana from the live DVD just now and have installed > > > gcc-8, cmake, and some libraries as I want to compile a small > > application, > > > but am getting a message that "pthreads,h - not found" > > Are you sure that it's pthreads.h? I only see pthread.h in my build env. > > You could check whether the > > package "system/header" is installed. We also have the package > > "build-essential" which should install > > a foundation of what is needed to compile software. > > > > Regards, > > Andreas > > > > > > I looked on the OpenIndiana package library online and found > > > > > > library/pth@2.0.7,5.11-2018.0.0.0:20180211T123900Z > > > < > > > http://pkg.openindiana.org/hipster/info/0/library%2Fpth%402.0.7%2C5.11-2018.0.0.0%3A20180211T123900Z > > > > > > > > > and > > > x11/library/libpthread-stubs@0.3,5.11-2018.0.0.0:20180212T053005Z > > > < > > > http://pkg.openindiana.org/hipster/info/0/x11%2Flibrary%2Flibpthread-stubs%400.3%2C5.11-2018.0.0.0%3A20180212T053005Z > > > > > > > > > for which I tried to install them but it said that no update was > > necessary > > > for these images. > > > > > > Can someone please tell me how I might be able to fix this? > > > Thanks, > > > Lonnie > > > ___ > > > openindiana-discuss mailing list > > > openindiana-discuss@openindiana.org > > > https://openindiana.org/mailman/listinfo/openindiana-discuss > > > > > > ___ > > openindiana-discuss mailing list > > openindiana-discuss@openindiana.org > > https://openindiana.org/mailman/listinfo/openindiana-discuss > > > > > -- > Best Regards, > > Lonnie T. Cumberland, PhD. > Email: lonnie.cumberl...@outstep.com > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Native speakers' help with translation wanted
On Tue, Dec 4, 2018 at 8:42 AM Michal Nowak wrote: > Hi, > > we are slightly updating the English "Welcome to OpenIndiana Hipster" > home page one can see when starting Firefox on MATE installation medium. > If you have release/os-welcome package installed you can see the > (localized) page at > >file:///usr/share/doc/openindiana-welcome/html/index.html > > Can a native English speaker read the changed text at > https://github.com/OpenIndiana/openindiana-welcome/pull/10/files and > provide feedback (either in the PR or here), please? > Here you go: [Quite a lot of the native englishness is getting commas correct.] "OpenIndiana is an illumos based Unix distribution" Is it OK to use Unix, which is a trademark? "OpenIndiana obtains its name from Project Indiana, an open source effort by Sun Microsystems (now Oracle Corporation) to produce OpenSolaris, a community developed Unix distribution based on Sun Solaris. Project Indiana was led by Ian Murdock, founder of the Debian Linux distribution." The OpenSolaris *distribution* wasn't anything to do with the community, it was purely an internal Sun effort. "Oracle's Sun Studio has been replaced ..." The Sun Studio compiler has been replaced ... In terms of changes, replacing GNOME with MATE would seem to merit a mention. "When operating system is updated ..." When the operating system is updated ... "ZFS features like Boot Environment and snapshots are leveraged in every respect of the process." Boot Enviromnents (plural) respect should be aspect "When Global Zone is updated, IPS makes sure, that packages in Zones are updated as well, or, at minimum, are kept in a coherent state." When the global zone is updated IPS ensures that packages in zones are consistent with the global zone, updating packages as necessary. "The Crossbow network virtualization framework and Role-Based Access Control (RBAC) model frameworks are also very well integrated with the rest of the system." Role-Based Access Control (RBAC) and Crossbow network virtualization are well integrated with the rest of the system. "Locality Groups (NUMA affinity) have been provided by the core operating system for many years, contrasting with other systems' reliance on "third-party" libraries." I'm not sure that's actuallly true, our NUMA support is pretty old now. "Considering the cross-pollination of technologies with FreeBSD and Linux, our code-base had contributed in a notable extent to operating system's innovation. Hence we still strive for relevance, observability, and consistency!" Considering the cross-pollination of technologies with FreeBSD and Linux, our code-base has contributed to a notable extent in operating system innovation. Hence we still strive for relevance, observability, and consistency! "Currently build servers, hosting and bandwidth is donated by EveryCity hosting in the UK." Currently, build servers, hosting, and bandwidth, is donated by EveryCity hosting in the UK. "Because our success as a operating system distribution depends upon the success of our community, please, feel welcome in our mailing lists, IRC channels, and GitHub projects." Because our success as a operating system distribution depends upon the success of our community, please feel welcome to participate in our mailing lists, IRC channels, and GitHub projects. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] OpenOffice on Hipster
On Tue, Jul 24, 2018 at 7:16 PM, Reginald Beardsley via openindiana-discuss wrote: > > Can anyone explain what's going on here: > > # pkg search openoffice > INDEX ACTION VALUE PACKAGE > pkg.fmri setopenindiana.org/desktop/office/openoffice > pkg:/desktop/office/openoffice@4.1.2-2016.0.1.0 > # pkg install openoffice > No updates necessary for this image. > # pkg info openoffice > pkg: info: no packages matching the following patterns you specified are > installed on the system. Try querying remotely instead: > > openoffice > > > pkg goes through the usual gyrations of making a plan, etc and then says > no updates are needed. Running "find / -name '*office*' " does not find an > executable. > If you look at the package metadata for hipster, it contains the following: set name=pkg.fmri value=pkg://openindiana.org/desktop/office/openoffice@4.1.2,5.11-2016.0.1.0:20180705T215200Z set name=pkg.obsolete value=true set name=org.opensolaris.consolidation value=userland That's it. It's an empty, obsolete package. If you're after LibreOffice, then a look at http://sfe.opencsw.org/libreoffice52-quick-install may be of assistance. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Where in the disk the parity stores, RAID-Z, RAIDZ2
Hi, I want to write driver to use in ZFS environment. > Could anybody told me The parity bits (RAIDZ,RAIDZ2) places. > Does the FS'es has own its region or continuously? > It's distributed, and not constant width. There's a helpful picture to represent the layout of data and parity in this article: https://www.delphix.com/blog/delphix-engineering/zfs-raidz-stripe-width-or-how-i-learned-stop-worrying-and-love-raidz > And when I get sources and build environment in my sandbox, > I will start try from /dev/null, Someone told me it locates? > One of the simplest recent driver additions was /dev/full, which might be worth taking a look at (and it shows that most of the work wasn't actually in driver code): https://github.com/illumos/illumos-gate/commit/b1c760bd3306ff30af69800801f30f3dd4b6394e -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Amazon AMI
On Thu, May 24, 2018 at 2:14 PM, Jacques Hugo wrote: > Hi guys, > > I’m assuming the question was asked before, but did anyone successfully > create an AMI image of OI Hipster? > The OpenZFS project has one, but it's only available in us-east-1 as far as I know (ami-c5c0a7d3 is the current one). So you can run it directly there, or copy it if you want or need to run it in a different region. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] find cycle's detected
On Sat, May 12, 2018 at 3:59 PM, Jacques Hugo wrote: > Hi all, > > Has anyone seen this error when compiling/installing from source: > > find: cycle detected for /lib/crypto/32/ > Something is picking up /usr/sbin/install (which behaves differently to the BSD and GNU versions of install). Many packages now are smart enough to autodetect, but not all. You'll need to help it find the gnu install. A couple of ways are: - 'make install INSTALL=ginstall' - ensuring /usr/gnu/bin is at the front of the path -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Connecting Serial Modem to Illumos (ACM9
On Sun, Apr 29, 2018 at 12:56 PM, Till Wegmüller wrote: > Hello Dear OI Community > > I am trying to connect my Ultimaker to my OpenIndiana Server. It is a > serial device accessed through a python software. > > When connecting the USB I can see it showing up in dmesg: > Apr 29 11:34:36 node02 usba: [ID 349649 kern.info] usba:no > driver found for interface 1 (nodename: 'data') of Ultimaker BV > (www.ultimaker.com) Ultimaker 2.0 > Apr 29 11:34:36 node02 usba: [ID 912658 kern.info] USB 1.10 interface > (usbif2341,10.config1.0) operating at full speed (USB 1.x) on USB 3.0 > root hub: modem@0, usbsacm1 at bus address 2 > Apr 29 11:34:36 node02 usba: [ID 349649 kern.info] Ultimaker BV > (www.ultimaker.com) Ultimaker 2.0 > Apr 29 11:34:36 node02 genunix: [ID 936769 kern.info] usbsacm1 is > /pci@0,0/pci1043,84ca@14/communications@1/modem@0 > Apr 29 11:34:36 node02 genunix: [ID 408114 kern.info] > /pci@0,0/pci1043,84ca@14/communications@1/modem@0 (usbsacm1) online > > But I do not know where that file would be under /dev. On linux it's > somewhere under /dev/ttyUSB or /dev/ttyACM. > > I have a device under /dev/term that links to > ../../devices/pci@0,0/pci1043,84ca@14/communications@1/modem@0:0 like > usbsacm1 should. But that device seems not to be correct. > > Does anybody know where that device usbsacm1 would be under /dev? or do > I need anything aditional to get access to that device? > Is there anything in /dev/cua ? (There should be an equivalent in there to the device in /dev/term.) -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] surprising behaviour of mount
On Thu, Feb 8, 2018 at 3:30 PM, Marc Lobelle wrote: > Hello All, > > Today I had a very curious message when trying to mount a file system on a > non-empy directory > > cannot mount '/export/home/ml/mydirname': directory is not empty > > Afaik, mounting on a non empty directory has always been allowed: the > original contents is just temporarily hidden. > zfs has always refused to mount over a non-empty directory by default > mounting was done automatically as a byproduct of > > zpool import -d /dev/lofi szpool > > szpool had been created by > > zpool create -m /export/home/ml/mydirname szpool /dev/lofi/1 > > Is there a good reason to prohibit mounting on a non-empty directory in > this case ? It's to stop you making mistakes. > Is there a way to avoid this constraint ? > 'zfs mount -O' will override the non-empty check. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] recompiling a program for openindiana
On Mon, Nov 20, 2017 at 9:02 AM, Marc Lobelle wrote: > Hello, > > I am trying to recompile a program called srm (available on sourceforge ) > for openindiana. It works as rm but makes sure that there is no trace of > the destroyed file in the blocks of the free list. > This program uses #if defined (__linux__) and #if defined (__OpenBSD__) > and I should replace this code with something appropriate for openindiana. > __linux__ etc are predifines preprocessor macros, presented in > > https://sourceforge.net/p/predef/wiki/OperatingSystems/ > > However, I do not see openindiana in there, so what should I use ? > Note that if you're using ZFS (which is the default file system on OpenIndiana) then the overwriting which srm does will have no effect - the copy-on-write mechanism that ZFS uses for data integrity ensures that the "overwrite" will go to a different, unused, part of the device. Therefore, srm won't do any good. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] switching from Ubuntu
On Tue, Nov 7, 2017 at 11:53 PM, Lonnie Cumberland wrote: > Hi Till, > > I have been searching over the links that you provided for Qualcomm Atheros > AR8161 Gigabit Ethernet" and "Broadcom Limited BCM43228 802.11 a/b/g/n" > wireless network cards but it seems that they are not supported yet. > The atge driver has been included in illumos (and thus OpenIndiana) for many years. It doesn't list the AR8161 as supported (so it won't recognize it) but it's worth trying. (This is from memory, so I hope I got it right.) As root: update_drv -a -i '"pciex1969,1091"' atge devfsadm -i atge dladm show-phys -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] [discuss] OpenIndiana iso boot problem using GRUB2
On Sat, Nov 4, 2017 at 9:29 PM, Ganapathy S A wrote:
> Hello All,
>
> Booted the four illumos distributions (iso): SmartOS, OmniOSce, Tribblix,
> and OpenIndiana by following this SmartOS wiki
> <https://wiki.smartos.org/display/DOC/Booting+SmartOS+from+GRUB2>. All
> booted properly but OpenIndiana! I know this isn't the recommended way to
> boot but this seems very convenient, i.e., download the iso image, then
> boot either by burning it to CD/DVD or by using GRUB2 like most Linux
> distributions do! Please have a look at the boot message and the grub.cfg
> file below.
>
The fundamental difference here is that the ISO for
SmartOS/OmniOSce/Tribblix has a boot_archive
that is a complete instance of the OS, whereas OpenIndiana has a split-root
structure where it
subsequently mounts /usr from a file on the media.
If you want to boot OI this way then you need some way of fooling it into
finding the solaris.zlib file
where it's expecting it. It looks for physical USB or optical media, which
I'm guessing you're trying to
avoid. The other option might be to fool it into thinking it's doing a
network boot by adding something like
-B install_media=http://some.site/location/
where that URL contains the two .zlib files off the iso. But generally the
old Caiman boot/install
didn't have this kind of boot mechanism in mind when it was developed.
(Earlier versions of Tribblix and OmniOS also had the same split-off /usr,
but both have changed to a
single archive to make implementation easier and more flexible.)
>
> *OpenIndiana boot message: *
>
> SunOS Release 5.11 Version illumos-2727bb055f 64-bit
> Copyright (c) 1983, 2010, Oracle and/or its affiliates. All rights
> reserved.
> ERROR: failed to attach AMD IOMMU
> Hostname: openindiana
> Remounting root read/write
> Probing for device nodes ...
> Preparing live image for use
> Requesting System Maintenance Mode
> (See /lib/svc/share/README for more information.)
> Console login service(s) cannot run
>
> Enter user name for system maintenance (control-d to bypass): jack
> Enter jack password (control-d to bypass):
> Login incorrect or user jack not authorized
>
> Enter user name for system maintenance (control-d to bypass): root
> Enter root password (control-d to bypass):
> single-user privilege assigned to root on /dev/console.
> Entering System Maintenance Mode
>
> Nov 4 20:12:07 su: 'su root' succeeded for root on /dev/console
> ld.so.1: bash: fatal: libncurses.so.5: open failed: No such file or
> directory
> ld.so.1: bash: fatal: relocation error: file /usr/bin/bash: symbol PC:
> referenced symbol not found
> Requesting System Maintenance Mode
> (See /lib/svc/share/README for more information.)
> Console login service(s) cannot run
>
> Enter user name for system maintenance (control-d to bypass):
>
> */boot/grub/grub.cfg*
>
> set timeout=5
> set default=0
> set color_normal="white/black"
> set menu_color_normal="white/black"
> set menu_color_highlight="black/white"
>
> menuentry " SmartOS 201708 " {
> loopback loop (hd0,msdos1)/smartos-20170831T155808Z.iso
> multiboot (loop)/platform/i86pc/kernel/amd64/unix
> /platform/i86pc/kernel/amd64/unix -B smartos=true
> module (loop)/platform/i86pc/amd64/boot_archive
> /platform/i86pc/amd64/boot_archive type=rootfs name=ramdisk
> }
>
> menuentry " OmniOS 201709 " {
> loopback loop (hd0,msdos1)/omniosce-r151022s.iso
> multiboot (loop)/platform/i86pc/kernel/amd64/unix
> /platform/i86pc/kernel/amd64/unix
> module (loop)/platform/i86pc/amd64/boot_archive
> /platform/i86pc/amd64/boot_archive
> }
>
> menuentry " Tribblix 201709 " {
> loopback loop (hd0,msdos1)/tribblix-0m20.1.iso
> multiboot (loop)/platform/i86pc/kernel/amd64/unix
> /platform/i86pc/kernel/amd64/unix
> module (loop)/platform/i86pc/boot_archive
> /platform/i86pc/boot_archive
> }
>
> menuentry " OpenIndiana 201710 " {
> loopback loop (hd0,msdos1)/OI-hipster-gui-20171031.iso
> multiboot (loop)/platform/i86pc/kernel/amd64/unix
> /platform/i86pc/kernel/amd64/unix
> module (loop)/platform/i86pc/amd64/boot_archive
> /platform/i86pc/amd64/boot_archive
> }
>
>
> Any help is appreciated. Thanx.
>
>
> Regards,
> Ganapathy
>
> *illumos-discuss* | Archives
> <https://illumos.topicbox.com/groups/discuss/discussions/T8d8addbc5041f73f-M1c79045d5b28fc2dacd920c7>
> | Powered by Topicbox <https://topicbox.com>
>
--
-Peter Tribble
http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/
___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] [openindiana-discuss]OI KVM status on host AMD CPUs
On Fri, Oct 20, 2017 at 7:55 AM, Rares Aioanei wrote: > Hey, > The OI virtualization wiki states that at the moment, KVM host support > is limited to modern Intel CPUs. However, the page seems outdated > (2014, if memory serves), so my question is : do you guys know of any > recent progress on AMD support? Nothing much recent. There have been a couple of forks working in that direction: https://github.com/jclulow/illumos-kvm https://github.com/chneukirchen/illumos-kvm One problem, I imagine, is that those people with the skills to add AMD support don't have any AMD hardware, as it's almost completely vanished from the market. > And also, I couldn't find the code > that enables OI to serve as host, if anyone can point me in the right > direction, I'd appreciate it. > Thanks. > > -- > Rares Aioanei > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Trying to run Java 1.7.0.11 on OpenIndiana Hipster 2017.04
On Wed, Sep 27, 2017 at 8:10 AM, Stephan Budach wrote: > Hi, > > > I am trying to replace omniOS on a RSF-1 node with OI and I need to run > some old Java agent on it which requires, Java 1.7.0.11, so I copied over > the Java folder from the previous installation and tweaked the symlinks > accordingly, but when running java -version, I am getting this error: > > > > root@zfsha02gh79:/usr/bin# java -version > Error: dl failure on line 864 > Error: failed /usr/java_1.7.0_11/jre/lib/i386/server/libjvm.so, because > ld.so.1: java: fatal: libCrun.so.1: open failed: No such file or directory > > > Can I get this version of Java running on OI? > Thanks, > Stephan > Should work, but you'll need the old C++ runtime installed. pkg install system/library/c++/sunpro (as found by, eg, 'pkg search libCrun.so.1') -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] So is Solaris officially supported till 2034?
On Tue, Sep 5, 2017 at 11:45 AM, Gabriele Bulfon wrote: > The real concern is...in the short or long period, how much support for > Solaris kernels can we expect from third parties such VMware? > They may completely remove the "Oracle Solaris 11 64bit" from the list of > supported VMs. > This may happen elsewhere, as happened recently at MongoDB. > This may hurt the illumos ecosystem in the short/long term. > The solution here is quite simple. Make sure that the suppliers of software that you need are aware that there is demand from Solaris/illumos users. For commercial products, this involves using the product and paying for it. You can't expect vendors to spend hard cash supporting a platform that doesn't generate revenue. For open source projects, get involved. Engage with the project, submit bug reports, feedback, fixes, and generally be visible. Generally, many projects want to be as widely adopted as possible, and welcome engagement from as wide a user base as possible. For most projects out there, any offer of help will be welcomed with open arms. If we all just picked one or two of our favourite projects and dived in, that would help greatly. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] possible to modify the log file location for CRON?
On Thu, Aug 3, 2017 at 6:34 AM, Jerry Kemp wrote: > Wondering if it is possible to change/configure the directory location for > cron from the stock /var/cron/log location to somewhere else. > Not that I know of. It's hard-coded in the source. Why do you want to change it? > Looking at the man page, All I really see as configurable is that logging > can be turned off in the /etc/default/cron file. The PATH environmental > variable can also be set here. > > I also ran strings across the 'cron' binary, in hopes of turning up > something. If it is/was there, I missed it. > > Is it possible to configure the cron log file location using the 'as > supplied' cron daemon that ships with OpenIndiana? > > Thanks, > > Jerry > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] New request for info `Are we able to install recent vbox versions'
On Thu, Jul 27, 2017 at 3:50 PM, Harry Putnam wrote: > Peter Tribble writes: > > > Sure. > > > > Over time, functions get added to libc. We don't change the version > number > > of libc itself; the versioning is handled internally. > > Wow.. thanks for such a thorough explanation. > > So, it must be some hefty work to get that straightened out or it > would be. > > I see the bug was reported quite some time ago. Do you suppose > that correction will happen any time this year? Or is there just not > much interest in it? > Oh there's been a lot of interest in it. Alexander's been doing a lot of work validating and testing. I'm cautiously optimistic this will get fixed soon. The fix at a technical level is very simple. Getting people happy with it takes a bit longer, and making fixes to align with Oracle Solaris is a politically charged subject for some. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] New request for info `Are we able to install recent vbox versions'
On Wed, Jul 26, 2017 at 10:33 PM, Harry Putnam wrote: > Peter Tribble writes: > > > On Sun, Jul 16, 2017 at 5:34 PM, Harry Putnam > wrote: > > > >> I posted this query July 9th but got no takers far as a response. > >> Is this question in appropriate here? Is it a foolish question? > >> > >> If both answers are no, I'd really like to hear from someone who > >> either knows first hand or has a solid opinion about it. > >> > >> I don't want to start messing up my OS if installing qt5 and trying to > >> run recent version of Vbox are likely to cause me grief, I might not > >> be able to contend with. > >> But I do need to get a newer version of Vbox installed and working. > >> > > > > VirtualBox 5.0.x works fine (may need the vboxconfig.sh patch). > > > > VirtualBox 5.1.x doesn't work, most likely due to illumos bug 5709. > > Peter, is this the bug with the title: > `Add binary compatibility with Solaris 10 update' > > Just making sure I'm reading the right one. That's the one. > I'm too dense to > understand how that impacts installing recent versions of virtualbox. > > Can you explain a bit? > Sure. Over time, functions get added to libc. We don't change the version number of libc itself; the versioning is handled internally. You can see this by using tools such as elfdump and pvs. Eg. # elfdump -v /lib/libc.so | more Version Definition Section: .SUNW_version index version dependency [1] libc.so.1[ BASE ] [2] ILLUMOS_0.22ILLUMOS_0.21 [3] ILLUMOS_0.21ILLUMOS_0.20 [4] ILLUMOS_0.20ILLUMOS_0.19 ... there's a lot more versions. If you run 'pvs -vs /lib/libc.so' then you get to see which functions were added to libc in that version. The ILLUMOS labels correspond to functions added by the illumos project since the fork; the SUNW symbols correspond to Solaris releases. (And there's a few other things in there too.) The source for that information is here: http://src.illumos.org/source/xref/illumos-gate/usr/src/lib/libc/port/mapfile-vers which has the advantage of having useful comments about release names and other things of archaeological interest. You can query a binary or shared library to see what it needs, the same elfdump -v as before. So if you look at /bin/true you will find that its requirements are rather minimal. The interesting file in the case of VirtualBox is /opt/VirtualBox/amd64/libQt5XcbQpaVBox.so.5 # elfdump -v /opt/VirtualBox/amd64/libQt5XcbQpaVBox.so.5 Version Needed Section: .SUNW_version index fileversion [2] libpthread.so.1 SUNW_1.2 [3] libm.so.2 SUNW_1.1 [4] libc.so.1 SUNW_1.22.7 [5] SUNW_1.22[ INFO ] [6] SUNW_1.1 [ INFO ] [7] SUNW_0.7 [ INFO ] [8] SUNWprivate_1.1 [9] libgcc_s.so.1 GCC_3.0 OK, so this tells you that it needs the SUNW_1.22.7 version of libc. And if you look with pvs -vs you can see exactly which function it needs: # pvs -vs /opt/VirtualBox/amd64/libQt5XcbQpaVBox.so.5 ... libc.so.1 (SUNW_1.22.7): asprintf; ... Now, go back to our libc (or the mapfile source). We jump straight from SUNW_1.22.6 (which is Solaris 10 update 9) to SUNW_1.23 (which is Solaris 11, or OpenSolaris as it was). We're missing SUNW_1.22.7 entirely. That version corresponds to either S10 update 10 or update 11 (something after update 9). We do have the functions present in our libc, but have them labelled as SUNW_1.23. Oracle added that intermediate version when they added (backported) a handful of the Solaris 11 functions to Solaris 10. So the fix we would like for bug 5709 is to add that intermediate version with the functions that are affected being moved from SUNW_1.23. Then VirtualBox will find SUNW_1.22.7 just as it expects. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] roughly normal rsyncing data point to point on localhost
On Mon, Jul 17, 2017 at 8:30 PM, Harry Putnam wrote: > Setup: OS: Openindian/hipster (A branch of Solaris 11) > Hardware: HP xw8600 workstation: >2x Intel(R) Xeon(R) CPU X5470 @ 3.33GHz - 32 GB Ram > --- --- ---=--- --- --- > > I curious to find out if what I'm seeing while rsyncing data from > point to point on localhost is about what I should expect. > > I do realize the many factors might impact this kind of testing. > > But just want a non-scientific in general response from a few of you > old hands that will if this is way off or what. > > Sorry if this is way tmi . . . > > The hardware in use is listed at the top of this message. > > $rsync is rsync-3.1.2 > > I'm transferring a recently deleted linux running in Vbox vm back into > place. > > $DST = /vb/vm/ where /vb/vm is >pool/vb and pool/vb/vm > > SRC = /vb/vm/.zfs/snapshot/2weeksago/vm1 > > rsync -vvrptgd0l --stats $SRC $DST/ > > So, rsyncing vm1 back into position from a snapshot > > --- --- ---=--- --- --- > > I'm seeing, over time (30min), very close to 1GB transferred every 60 > seconds > > Can anyone say if this par for the course or what? > Assuming that's on a traditional spinning-rust hard drive, seems within expectations. That equates to about 16meg a second. The writes will tend to be written sequentially, but the reads may be random (especially for a VM image that has put its own data layout atop the zfs one). Even with a 128k block size, that would be 128 IOPs which is about what you would expect for a hard drive. My first port of call here is to run 'iostat -xnz 1' - it will show how much data is being transferred (the kr/s and kw/s columns) as well as the number of requests (r/s and w/s), as long as how busy (%b) the device is. Current versions also show the statistics for the pool and the underlying disk(s) so you can see how the reads and writes get broken up between the layers. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] New request for info `Are we able to install recent vbox versions'
On Sun, Jul 16, 2017 at 5:34 PM, Harry Putnam wrote: > I posted this query July 9th but got no takers far as a response. > Is this question in appropriate here? Is it a foolish question? > > If both answers are no, I'd really like to hear from someone who > either knows first hand or has a solid opinion about it. > > I don't want to start messing up my OS if installing qt5 and trying to > run recent version of Vbox are likely to cause me grief, I might not > be able to contend with. > But I do need to get a newer version of Vbox installed and working. > VirtualBox 5.0.x works fine (may need the vboxconfig.sh patch). VirtualBox 5.1.x doesn't work, most likely due to illumos bug 5709. You don't need qt5, as VirtualBox comes with its own. (Talk about getting qt5 is more so that VirtualBox can be built from source to avoid compatibility issues with the downloaded package.) > From July 9th: > > From: Harry Putnam > Subject: Should we be able to install and run recent vbox versions > Newsgroups: gmane.os.openindiana.general > To: openindiana-discuss@openindiana.org > Date: Sun, 09 Jul 2017 17:56:53 -0400 (6 days, 18 hours, 29 minutes ago) > Message-ID: <867ezhl1ei@local.lan> > > Setup: OS: Openindian/hipster (A branch of Solaris 11) >Hardware: HP xw8600 workstation: > 2x Intel(R) Xeon(R) CPU X5470 @ 3.33GHz - 32 GB Ram > --- --- ---=--- --- --- > I've heard that recent versions of Vbox (above 5.0.X) are based on qt5 > > Should we be able to install recent versions of Vbox from the VBox web > pages, provided we have qt5 installed? > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] To the OmniOS, OI and SmartOS community
On Thu, May 18, 2017 at 10:53 AM, Guenther Alka wrote: > > There are many users who want exact these OmniOS features not available in > OI. > The discussion now is if such an effort should be based on the current > OmniOS bloody > or can be based on a common development effort together with the OI > community. > > I want the discussion going into a common future as I fear that a new > separate OmniOS community effort will only weaken Illumos/OI and cannot > survive in the long term. My hope is that a stronger OI/OmniOS is the > outcome instead where OmniOS is a stable add-on feature for OI snapshots. > I don't think that's viable, at least in the short term. OmniOS has a completely different ethos to OI, and there's a reason people chose one distro or the other. Attempting to force both requirements out of a single distro will likely mean that you end up satisfying neither set of users, while making it much harder for maintainers who have to support both sets of users. Longer term is a slightly different picture. It's unlikely that OmniOS will continue to diverge from illumos, and I would like to see the features backported into illumos proper. Once that's done it makes more sense to think about sharing, because at least you then have a common illumos codebase. > Am 18.05.2017 um 11:24 schrieb Alexander Pyhalov: > >> On 05/18/17 12:15 PM, Guenther Alka wrote: >> >>> News about OmniOS and Discuss about free Illumos distributions and >>> future OmniOS/OI options >>> >>> I have started a thread in a popular forum with a strong ZFS focus to >>> discuss this >>> also with regular users outside the maillists. >>> >>> https://forums.servethehome.com/index.php?threads/omnios-151 >>> 022-long-term-stable.14367/page-2 >>> >> >> Hi. >> So far, as I understand, OmniOS developers choose to support existing >> infrastructure and build systems and there's no development on merging OI >> and OmniOS code base. Without OmniOS developers interested in this, this >> talk is useless. >> > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Where is tcpd in OI
On Sat, Apr 29, 2017 at 3:49 PM, Stephan Budach wrote: > > I am trying to install check_mk_agent in OI hpister, but it's missing the > tcpd program, which I cannot even find using pkg search. > Can anyone tell me, if tcpd is available in OI and if not, if there's a > supplement for it? > The tcpd program comes with tcp-wrappers which is part of illumos. On IPS distros the following ought to work: pkg install library/security/tcp-wrapper -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Wiki https certificate is expired
On Wed, Apr 12, 2017 at 7:39 PM, Adam Števko wrote: > > I am also using letsencrypt, but I don’t have an official client running. > I am using getssl, which is working correctly on other parts of OI infra, > so this is a new thing for me. I will have to find out what’s going on. > If somebody was interested, it would be nice if we were able to provide > some letsencrypt client in oi-userland. > I use dehydrated (on OmniOS, but that shouldn't matter) and am pretty happy with it. https://github.com/lukas2511/dehydrated -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] nonblocking i/o problem
On Wed, Apr 5, 2017 at 9:16 AM, Jeff Woolsey wrote: > Based on an assurance > [https://wiki.illumos.org/display/illumos/illumos+FAQs, Is illumos > compatible with Solaris/OpenSolaris?] that OI-hipster was upward > binary-compatible with S10 We probably need to fix the FAQ clarify the conditions under which that is actually true. As an aside, there are limitations. The conditions were always pretty strict, so that compatibility only applies to (a) Solaris itself, and (b) only those interfaces explictly marked as being compatible. This explicitly excludes any 3rd-party applications and libraries that are shipped with Solaris, and as of Solaris 10 there were a *lot*. In particular, link with the system openssl and you're in trouble. What I've found in general is that older (pre-S10) applications tend to be fine, S10-native applications can be problematic. There's also the fact that we don't claim compatibility with any version of Solaris after the fork. So S10U10 is OK, S10U11 is problematic, S11 is essentially a no-go area. > and that any discrepancies were to be > reported, and some S10 exploits becoming more widely known, I upgraded > from Solaris S10XU8 to OI-hipster-2016.10. There was some teething, > and one thing I still haven't resolved is an S10 binary that works right > in S10 and wrong in OI regarding non-blocking reads of stdin. It > happens to be my main project, of course... > > truss, in short, says: > > Solaris X86 S10U8 > > fcntl(0, F_GETFL) = 2 > fcntl(0, F_SETFL, FWRITE|FNONBLOCK) = 0 > read(0, 0xFD7FFF0C0028, 4096) Err#11 EAGAIN > fcntl(0, F_SETFL, FWRITE|FOFFMAX) = 0 > > Openindiana-hipster 2016.10 > > fcntl(0, F_GETFL) = 8194 > fcntl(0, F_SETFL, FWRITE|FOFFMAX|FNONBLOCK) = 0 > read(0, 0xFD7FFF0C0028, 4096) = 0 > fcntl(0, F_SETFL, FWRITE|FOFFMAX) = 0 > > > Does anybody know why the behavior is different, and how to get the > former under OI. I expect there is some library difference somewhere. > I've been poking at tcsetattr to tweak this, and the best I can do is > see 4 characters get queued up before the program gets them (S10U8 does > that with different termios flags, too). read() returning 0 instead of > EAGAIN is interpreted as EOF and the program quits, rather clumsily. > I don't suppose you have the source? (I guess you may not have, as if you did then you could just recompile it.) Where does stdin come from, because it's different in the 2 cases (under OI it's largefile enabled, from your truss output)? -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] How to manage a send over ssh
On Thu, Mar 23, 2017 at 11:19 PM, jason matthews wrote: > > > On 3/23/17 2:56 PM, Timothy Coalson wrote: > >> The main advantage with this way is that you don't leave root ssh exposed >> for people to try to break into, and the special user for receiving can't >> get higher privileges for anything other than "zfs". >> > > With the notable exception that your unprivileged user can destroy your > file systems :) > Which is why I logged this bug: https://www.illumos.org/issues/5989 -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Any 32-bit users?
On Sat, Feb 11, 2017 at 6:02 PM, Nikola M wrote: > On 02/11/17 05:29 PM, Peter Tribble wrote: > >> Something that came up in discussion while we were at FOSDEM was >> whether it's safe to ship 64-bit only applications in OpenIndiana. >> >> Which begs the obvious question - is anyone running OpenIndiana on >> hardware that isn't 64-bit? Please reply if so. >> > > I have Asus eeepc that is 32-bit only, That was a fairly popular platform back in the early OpenSolaris days. Although (depending on model) it's not just the bitness of the cpu that's the problem, most of the eeepc and similar had pretty small RAM and storage provision. That's going to be a general problem - 32-bit systems may not have enough memory or disk to run a fairly heavyweight OS like OpenIndiana successfully. > that I installed Linux onto on notice that 32bit is not supported for > illumos for the future. > I haven't seen any suggestion that illumos drop 32-bit support. I mean, clearly it will eventually (if nothing else, it's the easiest way to fix time_t in 2038). That said, there's a clear shift in a lot of software (and other mainstream distros) that 64-bits is becoming the norm. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Any 32-bit users?
On Sat, Feb 11, 2017 at 4:52 PM, Aurélien Larcher < aurelien.larc...@gmail.com> wrote: > On Sat, Feb 11, 2017 at 5:29 PM, Peter Tribble > wrote: > > > Something that came up in discussion while we were at FOSDEM was > > whether it's safe to ship 64-bit only applications in OpenIndiana. > > > > Maybe we could identify a subset of applications, libraries for which it > makes sense to ship 32-bit and 64-bit, like server software, while desktop > could move to 64-bit only. > I would expect it to be the other way round. Server stuff tends to be for 24/7 or business use; there is no economic sense in running antique hardware for that, the power bill alone encourages replacement with newer (and much more capable) systems. Whereas having an old desktop or laptop that you might want to get a bit more service out of seems much more likely. But that's an opinion and has some anecdotal support from people I meet in the pub, it's not fact. Hence why asking the question is necessary. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Any 32-bit users?
On Sat, Feb 11, 2017 at 5:06 PM, ken mays via openindiana-discuss < openindiana-discuss@openindiana.org> wrote: > Peter, > 1. Yes. Many end users still own old 32-bit Pentium 4-based desktops and > laptops. > While true, that wasn't the question. The question is whether people are actively using OpenIndiana on said hardware. > 2. Yes, OI could ship 64-bit only applications/drivers/etc. > And such software exists. There are some applications that are 64-bit only by design - Go and anything written in it is an example here, and that's getting to be quite a large ecosystem. Then there are things that are shipped 64-bit by default - JDK8 is the obvious one, you can build it either 32 or 64 but on Solaris the default build is 64-bits. And then there's the fact that the upstream Oracle useland gate is aimed at a platform (Solaris) that no longer supports 32-bit hardware, so they can (and are doing so) choose to remove 32-bit support for certain applications. Hence OI can either follow them (and drop 32-bit support), or take on the maintenance burden of building a different bitness or bi-arch packages. OI has recent binary core snapshots and FOSS packages supporting 32-bit > platforms. Having a 64-bit only core distro snapshot and packages is just > forward thinking. Package maintainers can support either selection - but it > is no longer the norm to support both selections (32/64-bit packaging) if > the demand is not justified). > - Ken > > > > On Saturday, February 11, 2017 8:29 AM, Peter Tribble < > peter.trib...@gmail.com> wrote: > > > Something that came up in discussion while we were at FOSDEM was > whether it's safe to ship 64-bit only applications in OpenIndiana. > > Which begs the obvious question - is anyone running OpenIndiana on > hardware that isn't 64-bit? Please reply if so. > > To check, use the isainfo command. If "isainfo -k" returns "amd64" > (or "isainfo -b" returns "64") then you're running in 64-bit and should > happily ignore this question. If you get "i386" or "32" respectively > then it would be nice to know. > > Thanks, > > -- > -Peter Tribble > http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
[OpenIndiana-discuss] Any 32-bit users?
Something that came up in discussion while we were at FOSDEM was whether it's safe to ship 64-bit only applications in OpenIndiana. Which begs the obvious question - is anyone running OpenIndiana on hardware that isn't 64-bit? Please reply if so. To check, use the isainfo command. If "isainfo -k" returns "amd64" (or "isainfo -b" returns "64") then you're running in 64-bit and should happily ignore this question. If you get "i386" or "32" respectively then it would be nice to know. Thanks, -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Is it possible to arrange for dual booting oi hipster and Solaris 11.3 on a single disk?
On Thu, Jan 19, 2017 at 9:18 PM, Hans J Albertsson < hans.j.alberts...@gmail.com> wrote: > I'm just wondering. I would say not. If you just have the one disk, then you're limited to them sharing the same disk partition (there are limitations about there only being one solaris partition on a disk) and thus they have to be in the same zfs pool. Which means you have to create the root pool with zfs version 28 and make sure neither OS does a zpool upgrade. But the likely killer is that they use different bootloaders. S11.3 is GRUB2 and OI is either legacy grub or the new loader. If they were on different disks you would be OK, because you just chain one from the other, but needing two different bootloaders on the same disk would be a bit of an insurmountable problem. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] The end is near
On Wed, Jan 18, 2017 at 8:13 PM, Apostolos Syropoulos via openindiana-discuss wrote: > > Why "The End is near"? Sounds the SPARC platform will be continued > > beyond 2020, right? > > Obviously I do not care about SPARC machines... But I see no Solaris 12... > It seems that Solaris 11.4 will be the last release of Solaris. > Version numbers have specific meanings here. Having a 12 implies that it's incompatible with 11 at some level (that's precisely what the version number bump would mean). I haven't seen anything coming down the track that requires such a bump, it's far better for customers and support (and vendor qualification) to stick with update releases of 11. I guess they've just rebranded 12 as 11.next and about the only difference will be that all the legacy junk scheduled for removal in 12 will stay around a bit longer. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] How many versions of libs/apps should OI provide?
On Thu, Dec 22, 2016 at 6:14 PM, Jay F. Shachter wrote: > > Now, imagine a png package that contains the old header files, call it > png-devel version 1, and a png package that contains the new header > files, call it png-devel version 2. The discussion on this mailing > list seems to be whether hipster should publish more than one version > of a library. But this question can only arise with respect to > binaries, because libraries contain version numbers, so that > whizbang.so.1 and whizbang.so.2 can coexist on the same system. With > respect to header files, two versions cannot coexist on the same > system. Version 1 of the png-devel package will contain the file > /usr/include/png.h and version 2 of the png-devel package will also > contain the file /usr/include/png.h and consequently installing one of > the versions will overwrite a file that belonged to the other version. > Not so. In the case of libpng, the header files are also versioned so that you can tell it which version to compile against. The bare /usr/include/png.h is a symlink to whatever you've chosen as your default version. (And the libraries for different versions have different names.) So to build against, say libpng version 1.2.x, you would have -I/usr/include/libpng12 and -lpng12 This shows, by the way, that the possible solutions depend on which package you're discussing. (It's not plain sailing, even with something like png. You can end up with something pulling in both libpng12 and libpng14, for example, if you haven't updated all the dependent libraries as well.) -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] How many versions of libs/apps should OI provide?
[trimmed] On Thu, Dec 22, 2016 at 12:22 AM, Jim Klimov wrote: > > We had a sort of a discussion on IRC today, which ended up in a choice > better made or at least discussed by community than by the back-alley > dealers ;) > > So, as software evolution marches on, some projects' newer releases bring > new features, bugs and ABI/API incompatibilities that make them different > from older releases - and sometimes not backwards compatible (most projects > try to make updates easy for their consumers, some don't). > > As a result, the distro, such as OI/Hipster (and especially Hipster, > posing as bleeding-edge rolling-release), has to balance and make a choice > between: > > a) providing only the newest versions of libs/apps > > b) also provide older releases at least partially > > c) don't rush for newest releases - ain't broke don't fix; > > All of these variants (and probably more can be thought of) have their > valid pro's and contra's so it is not easy to pick one as an apparently > best choice. > Also, the correct choice may vary depending on the package in question, as they vary in their approach to stability. Some - gtk for example - are explicitly designed for major versions, such as gtk2 and gtk3, to coexist. Others have versions that you can overlap, such as libpng. In Tribblix I ship a compat package for libpng, that I can install for old packages that are linked against older versions. Anything new gets built against the current version. I use a similar approach for tiff. I originally needed to do this for the transition period of rebuilding everything I could, but the compat packages still exist just in case. So I favour keeping everything as current as you can, having compatibility packages for older release of libraries at least. I thoroughly dislike the "stable" approach of backporting patches to older versions. In my experience this ends up being expensive to maintain compared to just building current, ties you up in knots because other things require newer versions, and give you bigger and harder jumps (and more breakage) when you finally do upgrade. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] p7zip
On Thu, Dec 8, 2016 at 7:05 PM, Tim Mooney wrote: > In regard to: Re: [OpenIndiana-discuss] p7zip, Alexander Pyhalov said > (at...: > > >> Hi. Yes, we missed this fix. I've just committed it. >> Unfortunately, pkg info is quite useless in determining, which security >> fixes are applied to the package. >> > > Yeah, we talked about that issue last year around this time. This > post from Peter is from the middle of the long thread, but it captures > one of the most interesting ideas: > > https://openindiana.org/pipermail/openindiana-discuss/2015- > December/018370.html That's one way to do it. My point there was really though that if you were going to add metadata to IPS, then someone had already done it and it would make sense to follow in their footsteps rather than reinvent the wheel. The other option that I've been vaguely looking at is vuxml. That's a separate database (where the output from the database is just an xml file). Used by FreeBSD, and they have some tooling around it so you wouldn't be starting from scratch. And it may be possible to save a lot of work by importing their database. In any event, tracking all this is a lot of work. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Σχετ: FlashPlayer 23 NPAPI for Linux comeback
On Thu, Sep 8, 2016 at 6:32 AM, Apostolos Syropoulos via openindiana-discuss wrote: > we do not need flash plugins. This is a server OS > A server OS is probably more likely to be a sensible place to have flash, as it may be properly secured, and may be configured to limit you to accessing only the business-critical sites you need to get the job done rather than leaving you open to the whole unsavoury internet. Not having flash is, in most cases, a feature not a bug, and has been for a very long time. One of the longer-standing irritations I've had with losing flash is that the BBC relied on it. Now at least they've moved iPlayer to HTML5 so that works fine in Firefox on illumos without flash. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] VirtualBox v5.1.0 on hipster lastest
On Thu, Sep 8, 2016 at 1:35 PM, Predrag Zečević - Unix Systems Administrator wrote: > QT5 is required for VirtualBox 5.1 (https://www.illumos.org/issues/7145) > Not so. VirtualBox 5.1 includes QT5; the problem is that it doesn't work right. I'm wondering if this is another case of 5709/7165. It looks like the VirtualBox binaries were built on S10U10, as they're looking for asprintf. https://www.illumos.org/issues/5709 -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Building handbrake 0.10.5
On Sun, Aug 21, 2016 at 2:25 PM, Aurélien Larcher < aurelien.larc...@gmail.com> wrote: > On Sun, Aug 21, 2016 at 3:11 PM, russell > wrote: > > Hi, > > > > If you want to run HandBrake on OpenIndiana Hipster, you need to follow > the > > steps below after downloading the 0.10.5 source and extract to > > /tmp/HandBrake-0.10.5 > > > > 1. ensure that your path is set to use /usr/gnu/bin > > $ PATH=/usr/gnu/bin:$PATH ; export PATH > > > > 2. Next as root in seperate terminal, go to /bin > > # ls -l sh > > rwxrwxrwx 1 root root 9 Aug 21 13:55 sh -> i86/ksh93 > > > > Now remove this link > > > > # rm sh > > > > Replace this link with one pointing as /usr/gnu/bin/sh > > > > # ln -s /usr/gnu/bin/sh sh > > You should never do this. > If shell scripts are made for bash and incorrectly use /usr/bin/sh you > should modify the shebang. Should, yes. In this case Handbrake is downloading source archives itself, then unpacking them and running various scripts. So there's no easy way within the framework of the automated build process to modify any of the files, which is presumably why this game is necessary. An alternative might be to create fixed tarballs of affected packages and drop them into the download directory. I don't think Handbrake bothers verifying the tarball checksums, so it might work. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] OpenIndiana Drivers For My Wireless Network Device
On Fri, Aug 19, 2016 at 12:46 PM, the outsider wrote: > And this one seems to have it working too: > http://www.renatomorano.net/?p=291 > > Funny thing is that Peter dribble is mentioned in that page. He's on this > list too? > Tribble. Ahem. The NDIS stuff is, in general, *really* old. My experience was mixed at best. I think it got me through an OpenSolaris summit, but it never worked in the office, and was pretty unstable in busy environments (eg the airport, where it would kernel panic in under a minute). > And I found this: > https://thestaticvoid.com/post/2011/06/09/wireless-802- > 1x-support-in-solaris > / > > ( I can remember that I had a laptop once where I installed OI on. It > worked > with WPA out of the box for as far as I can remember) > > -Oorspronkelijk bericht- > Van: Carsten Grzemba [mailto:grze...@contac-dt.de] > Verzonden: vrijdag 19 augustus 2016 13:17 > Aan: Discussion list for OpenIndiana >; > openindiana-discuss@openindiana.org > Onderwerp: Re: [OpenIndiana-discuss] OpenIndiana Drivers For My Wireless > Network Device > > > > On 19.08.16 12:20, Jean-Pierre André > wrote: > > > > Aurélien Larcher wrote: > > >On Fri, Aug 19, 2016 at 11:13 AM, jay wrote: > > >> > > >>Esteemed Colleagues, > > >> > > >>Yesterday I booted my new laptop (well, new to me) from the latest > > >>hipster iso, intending to proceed immediately therefrom to an > > >>installation. I noticed, alas, that the system had no knowledge of > > >>my wireless network device. This makes the system useless for my > > >>computer, it is, after all, a laptop, it is a portable device that > > >>on occasion literally sits atop my lap, it has to be able to connect > > >>to a network without there being an Ethernet cable stuck into it. > > >>The Device Driver Utility (or whatever it's called, I don't have it > > >>on the screen anymore, otherwise I would be running OpenIndiana and > > >>would therefore be unable to send this e-mail) noted, correctly, the > > >>existence of a > > >> > > >> Broadcom Corporation BCM4312 802.11b/g LP-PHY > > >> > > >>but it had no driver for it. > > > > > >There has been some work to user NDIS wrapper: > > > > > >https://www.illumos.org/issues/3367 > > > > > > > This can only use WEP encryption, because the interface to NDIS5 is > > not compatible with the WPA four-step handshake (some user level > > replies from the access point have to be redirected to the > > supplicant). > > > > I would volunteer to interface to the Broadcom supplied driver > > (hybrid-port) for which an unterface to Linux is available, or to a > > more recent open source driver for Linux (which supports newer > > Broadcom hardware, though the BCM4312 is said to be poorly supported). > > > > But I have (again) to ask for help for doing that. The relations > > between net80211, mac, dladm and wpad are opaque. > > There has been a Google summer of code about WPA2, but I could not get > > anything useful from it (apparently WPA Enterprise requires a > > significant reworking of the relations between these modules). > > > > Jean-Pierre > > > > > > > > > > ___ > > openindiana-discuss mailing list > > openindiana-discuss@openindiana.org > > https://openindiana.org/mailman/listinfo/openindiana-discuss > > > I tried to merge the GSoC WPA-Enterprise stuff in illumos but it did not > work. Because I couldn't contact the maintainer I stopped my effort. > > https://github.com/cgrzemba/illumos-gate > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > https://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org https://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] setting autofs-props/environment variable
On Mon, Aug 8, 2016 at 5:03 PM, Udo Grabowski (IMK)
wrote:
> Hi,
> does anyone know how to set more than one environment variable
> with sharectl autofs-props/environment ?
>
> Tried several variants (all one-liners):
> sharectl set -p environment=
>... "ENV1=VAL1 ENV2=VAL2"autofs
>... "ENV1=VAL1\nENV2=VAL2" autofs
>... "ENV1=VAL1:ENV2=VAL2"autofs
>... "ENV1=VAL1^MENV2=VAL2" autofs
>
> where ^M is a literal CR, which was accepted, but gave funny
> output on svcprop and didn't exactly do what I wanted...
> The autofs manpage only says:
>
> AUTOMOUNTD_ENV==
>
> Environment variables. Each environment variable-value
> pairing must be on its own line. You can specify multi-
> ple such pairings. There are no environment variable
> settings supplied. For example: AUTOMOUNTD_ENV=DAY=TUES
>
That's odd. The autofs manpage on my systems says:
environment=name=value[,name=value]...
Specifies a comma separated list of environment variables. If
an
environment variable has more than one value, those values
should
be separated with a comma, preceded by a backslash as an escape
character ("\,"). For example:
VAR1=val1,VAR2=val2\,val3
So I think you just want commas.
--
-Peter Tribble
http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/
___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] [oi-dev] OpenIndiana Code of Conduct
Adam, Thanks for doing this. In my view, it's long overdue. Comments below. As part of a larger effort at providing a more formal governance structure > for the OpenIndiana project, > So, there is a wider context to this. Which is a good thing, but it would be good to have that fleshed out. Not necessarily conclusions, but it would be good for the community at large to understand what's missing and what's being done about it. > I’d like to announce on the behalf of OI developers the adoption of an > OpenIndiana Code of Conduct. The draft text for this new document can be > found at http://www.openindiana.org/community/code-of-conduct/. > > We would also like to point out this draft document is open to discussion > and acceptance by the community. > Our desire is for the discussion to be civil and for it to center around > the verbiage of the Code of Conduct. > We do not wish for the discussion to go off topic, or question the need > for such a document. > > Please compose your thoughts and comment with as few replies as necessary > so the community may solidify the final text of this document. > Overall, I'm happy with most of this. It's about the right length, as well. I think you need an explicit reporting mechanism. Something like "email ab...@openindiana.org" or the like. (Of course, you need that to be routed somewhere.) And the first statement about confidentiality means that the 2nd part about not being made public is redundant. There's also no mention of consequences. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Installing Hipster
On Thu, Jun 9, 2016 at 3:46 PM, Udo Grabowski (IMK) wrote: > A seemingly very simple question: > > - How to install a fresh (nonupdate) complete new Hipster > instance from an old /dev OI into a new BE ? > > Tried hard, but it seems that this is impossible to do without > rebooting into the new BE. All attempts end with pkg referring > to installed packages of the running system (and denying because > of conflicting packages, etc.pp). We just want a pristine > /hipster on its own zfs filesystem without changing or halting > the running system. I've implemented a couple of variants of this idea for Tribblix. 1. Boot from live media and have an alternative installer that differs from the normal installer in that it uses the existing rpool rather than creating a new one. The advantage of this method is that you're using the same bits to do the install that you're actually installing, which greatly simplifies matters. 2. Install a zone from an arbitrary ISO. (In practice, the script has to know how each distro builds the ISO, so it's not entirely arbitrary.) The script I use is at the URL below, you could probably modify it to install into an alternate BE - about the only change you would need is to not delete /boot /kernel and /devices. You then need to fiddle the grub menu. And I'm not sure whether the manual removal of the live image piece is correct for hipster, as the last time I ran this for OI was pre-hipster. Note that I never invoke pkg at all. https://github.com/tribblix/tribblix-zones/blob/master/usr/lib/brand/alien-root/live-iso-unpack So it's possible, but will probably involve some fiddling. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] [oi-dev] OpenIndiana Docs - updated
On Mon, May 16, 2016 at 5:39 PM, Alexander Pyhalov wrote: > ken mays via oi-dev писал 16.05.2016 18:19: > >> Hello, >> The 'default' minimal hardware requires =>768MB RAM. The achieved >> hardware minimum recently with the latest illumos kernel is 48MB RAM >> (i.e. in CLI mode). This was recently achieved by Peter Tribble. >> > > I suspect IPS alone will eat 1GB RAM without any issues... So if we speak > about minimum usable system, I'd say 768 RAM is not enough. > Quite, 1GB seems more realistic as a bare minimum. The current Oracle Solaris minimum has risen to 2GB. Those of us messing around at the 48MB point aren't actually expecting people to use such a system! Currently available new hardware is mostly 4GB or more, with a handful of systems (mostly cheap laptops) at 2GB. All my 2007/2008 vintage test boxes have at least 4GB. The interesting thing is people trying to re-use 10-year old boxes. It's probably worth clarifying 32-bit support. I recall Alexander saying recently that it had been effectively dropped, I'm not sure exactly what that means. (I know that OI hasn't really supported 32-bit for a while, I think 64-bit only X was the tipping point.) -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] New box/new motherboard suggestions
On Fri, May 13, 2016 at 3:53 AM, Jerry Kemp wrote: > I am at a point where I need to build a new box for $HOME, and I believe > that unless something else pops up, I'm looking at a SuperMicro ATX > motherboard + some Intel CPU's. > > Right now, I'm looking at these motherboards, but I'm open to something > else:: > > < > https://www.cdw.com/shop/products/SUPERMICRO-X10DAL-i-motherboard-ATX-LGA2011-v3-Socket-C612/4047227.aspx?pfm=srh > > > > < > https://www.cdw.com/shop/products/SUPERMICRO-X9DRL-3F-motherboard-ATX-LGA2011-Socket-C606/2783863.aspx?pfm=srh > > > > < > https://www.cdw.com/shop/products/SUPERMICRO-X10DRL-i-motherboard-ATX-LGA2011-v3-Socket-C612/3550684.aspx?pfm=srh#PO > > > > Planed usage: > > This will be a desktop system, not a server but I want dual CPU's and lots > of memory to have lots of zones and VM's. > > > Plus some new mid range NVIDIA frame buffer and a new/higher capacity > power supply. > > Staying with ATX size as this is all going into one of my spare Sun Ultra > 20 cases. I will probably uses those forever, or at least till I can't use > them anymore. > > Comments on the above motherboards are appreciated. If you have a better > suggestion, that is appreciated. > I don't have any direct experience with those, but the one thing I notice is that the X10DAL is a workstation (or tower) motherboard; the others are server models. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] command will not run from cron but will from CLI
On Mon, May 9, 2016 at 8:01 PM, wrote: > I have not been able to get even a simple command to run from cron. For a > test I created this script: > > #!/bin/sh > > export PATH="/usr/bin:/usr/sbin:/root" > /bin/echo "test" > /root/test.log > > It runs from the command line but does not from cron: > > 55 11 9 * * * /root/test > > The log shows: > > > CMD: * /root/test > > root 17987 c Mon May 9 11:55:00 2016 > < root 17987 c Mon May 9 11:55:00 2016 rc=127 > It's not the script, it's your crontab entry. The command it's trying to run (see the CMD in the log) is * /root/test and, indeed, where there should be 5 time specifiers in the crontab entry you have 6: 55 11 9 * * * -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Fwd: Xview in OpenIndiana
On Thu, Mar 10, 2016 at 1:20 PM, Juanjo Presa wrote: > Hi all, > > I have some problems with migration of a legacy Solaris x86 application to > a newer OpenIndiana operating system. > The problem is with the libxview and libolgx libraries. > The app (and the libs) works perfectly in Solaris 8, 9 and 10. > > Can you give some hints about how to compile this libs for OI? > I'm not sure you would need to recompile them. I'm not aware of any significant reason why copying the application and the libraries onto OpenIndiana wouldn't work. We should, in general, be compatible with applications from Solaris 10 and earlier. What are the problems you're having? -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] masquerade
On Thu, Mar 10, 2016 at 2:11 AM, wrote: > > This should be a simple and short thread. > > How do I configure packet filter on my computer, with two network > interfaces, to masquerade from my private LAN to the outside world, so > machines on my private LAN can have conversations with machines that > have public IP addresses? Astonishingly, search engines have not led > me swiftly to the solution (lots of stuff about sendmail masquerading > though, in case anyone cares about that), nor can I find helpful > documentation on the Oracle documents website. I have done my best to > read the fabulous manual, but I am confused. > I wrote a blog entry that covers this: http://ptribble.blogspot.co.uk/2015/11/zones-behind-zones.html It's in a different context (zones wired with crossbow rather than hosts on a LAN) but should cover it. So if you ignore the dladm bits to set up crossbow then it boils down to ipadm set-ifprop -p forwarding=on -m ipv4 net0 ipadm set-ifprop -p forwarding=on -m ipv4 net1 and map net0 172.16.0.0/16 -> 0/32 portmap tcp/udp auto map net0 172.16.0.0/16 -> 0/32 where you use the interface on the *public* side of the network. I think, looking at what you have below, that just changing net1 to net0 in ipnat.conf is what you need to do. You can omit telling me about routeadm, I've already done that. The > computer is already set up to route IP datagrams, I just need to get > the packet filtering right. > > Here is the state of my router machine at present: > > > / # ipadm show-addr > ADDROBJ TYPE STATEADDR > lo0/v4static ok 127.0.0.1/8 > net0/dhcp dhcp ok 99.140.186.69/30 > net1/v4 static ok 192.168.1.42/24 > net1/v4a static ok 172.16.1.1/16 > lo0/v6static ok ::1/128 > / # ndd -get /dev/ip ip_forwarding > 1 > / # cat /etc/ipf/ipnat.conf > map net1 172.16.0.0/16 -> 0.0.0.0/32 > map net1 192.168.1.0/24 -> 0.0.0.0/32 > / # ipnat -l > List of active MAP/Redirect filters: > rdr * 0.0.0.0/0 port 21 -> 0.0.0.0/32 port 21 tcp proxy ftp > map net1 172.16.0.0/16 -> 0.0.0.0/32 > map net1 192.168.1.0/24 -> 0.0.0.0/32 > > List of active sessions: > MAP 172.16.1.1 53<- -> 192.168.1.4253[172.16.1.3 56138] > MAP 172.16.1.1 53<- -> 192.168.1.4253[172.16.1.3 61524] > MAP 172.16.1.1 53<- -> 192.168.1.4253[172.16.1.3 55160] > MAP 172.16.1.1 64496 <- -> 192.168.1.4264496 [172.16.1.3 22] > > > I can ssh in to machines (e.g., the abovementioned 172.16.1.3) on my > home network, but once logged in, I cannot access the outside world > therefrom (e.g., "ping 8.8.8.8" times out). Needless to say, > 172.16.1.1 is the default router for 172.16.1.3, so that is not the > problem. And, if further proof be needed, 172.16.1.3 can easily ping > 99.140.186.69. So the masquerading is the problem, not the routing. > As I indicated, probably an extremely easy question to answer if you > know the answer. I'm sure it's something simple, like maybe the zeros > are supposed to be on the left rather than the right, in ipnat.conf. > Thank you in advance for any and all replies. > > > Jay F. Shachter > 6424 N Whipple St > Chicago IL 60645-4111 > (1-773)7613784 landline > (1-410)9964737 GoogleVoice > j...@m5.chicago.il.us > http://m5.chicago.il.us > > "Quidquid latine dictum sit, altum videtur" > > > ___ > openindiana-discuss mailing list > openindiana-discuss@openindiana.org > http://openindiana.org/mailman/listinfo/openindiana-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] [oi-dev] Updated OpenIndiana FAQ (3rd draft)
> "At this time, the following distributions are known to support the SPARC > platform:.." > I think it would be good to reorganize the list in alphabetic order. I should have chosen a different name, I always seem to end up last on the lists :-( More seriously, I wouldn't necessarily call out Tribblix as experimental, it's not that different really to the others (it's just that my x86 port, like others, is rather more mature than the SPARC port). But I would remove Oracle Solaris from the list. It should still be mentioned, but not in a way that implies it's an illumos distribution. Something like a separate line: "For production use on modern SPARC hardware, there is also Oracle Solaris." -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
[OpenIndiana-discuss] FOSDEM
(First, apologies for cross-posting.) Who's going to be at FOSDEM this weekend? I'll be there. I know Dan's giving a talk. And we have an illumos booth - who's doing that? We should try and arrange something. If nothing else I'll swing by the booth whenever I can. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] OI roadmap (for production)
On Fri, Dec 4, 2015 at 3:01 PM, Alexander Pyhalov wrote: > > As for my personal plans, I'd like to look at porting Mate now. > Just as a matter of interest, are you thinking of MATE as an additional desktop environment, or as a replacement for JDS/Gnome? I found it pretty easy to get going, and it doesn't have that many dependencies. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Update of JPEG libraries
On Mon, Dec 14, 2015 at 9:14 PM, Aurélien Larcher < aurelien.larc...@gmail.com> wrote: > Just FYI: > > http://www.openindiana.org/2015/12/14/update-of-jpeg-libraries/ I believe that libjpeg-turbo ought to be a straight drop-in replacement for the older ijg jpeg library. That's the claim, and the limited testing I've done hasn't thrown up any obvious problems. Are you aware of any reason not to simply replace the old ijg libjpeg with the one from libjpeg-turbo? -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] ACL problem
On Sun, Dec 13, 2015 at 3:02 PM, Michelle wrote: > > All I'm trying to do, is ensure that all files written within a > directory, have the same ownership as the directory itself, no matter > what account actually writes them. > Unfortunately, that's something you can't do. There's the setgid bit you can set on a directory, which ensures all newly created files have the same group as the directory, but there's no corresponding facility for the file owner. Using ACLs can allow you limited control over permissions - and you can extend permissions to multiple users and groups (rather than the single group that you normally use with traditional unix permissions). But even with ACLs you can only set the initial permissions on newly created files. If the application you're using to add the files starts to manipulate permissions itself (and many do) then even ACLs can't help you. What I normally do is: 1. Use basic unix groups to control access 2. Use the permissions of the top-level directory to act as a primary access filter - make it group owned by the appropriate group, usually setgid, and don't allow any access to other if you want to keep it private (mode 770). 3. If you can, control the application(s). For instance, you can tell samba what the default permissions should be. 4. For when all else fails, have a cron job that fixes up anything that's become broken. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] OI roadmap (for production)
On Tue, Dec 8, 2015 at 11:14 AM, Jim Klimov wrote: > > >From: Tim Mooney [mailto:tim.moo...@ndsu.edu] > > > > >I'm trying to find a way to verify component security that doesn't rely > >on more work from the few people that are already doing the security > >work, but it's not clear what a good method is to perform that > >verification. > > > >Tim > > Might it make sense to use some pkg(5) metadata to list the cve's known > covered by a particular release+patch recipe used in the build? I know i'd > quickly stop maintaining such data though, but there may be even pedantical > people than mysekf out there ;) And for a commercialized or otherwise paid > effort, someone could be doing this sysiphus task. Anyhow, someone has to > revise if a cve applies to our code and write down the inspection results > somewhere - might as well accompany the relevant code snapshot. > > reminds me sort of like sun's patch readmes with lists of changelogs and > bugids and errata... > You mean like the way Oracle Solaris has additional IPS metadata to track CVEs? https://blogs.oracle.com/darren/entry/cve_metadata_in_solaris_ips -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] using touch ... whats wrong here
On Thu, Dec 3, 2015 at 7:52 AM, Harry Putnam wrote: > Not particularly and OI question but that is where I need this > technique to work. I'm hoping some of you old shell hands will > see the problem. > > I hope not to be exibiting true ignorance... but here I go. > > I keep messing with one forumlation of find cnmd that seems like it > aught to work... What am I doing wrong or not understanding? > > I'm trying to use finds' -c newer technique > > Check on just two of many files with following dates. > > Two of quite a few files in this location: > -rwxr-xr-x 1 harry nfsu 1768 2014-07-22 20:18 scripts/perl/getad > -rwxr-xr-x 1 harry nfsu 6429 2014-07-22 20:18 scripts/perl/rnami.pl > > To my eye, those dates fall within the targetting files > > Now create our targeting files with target date inbetween > > touch NewerThan -t 201406141130 > > touch ButNotNewerThan -t 201408141130 > > Now check my targeting files: > > ls -l NewerThan ButNotNewerThan > -rw-r--r-- 1 harry nfsu 0 2014-08-14 11:30 ButNotNewerThan > -rw-r--r-- 1 harry nfsu 0 2014-06-14 11:30 NewerThan > > Ok, now try it up > > find scripts/perl/ -cnewer NewerThan -and ! -cnewer ButNotNewerThan > > (Nada) > > Where is the flaw? > You're mixing the various timestamps. A file has three timestamps: access, shown by ls -lu modify (of data), shown by ls -l creation (or modify metadata), shown by ls -lc Using find -cnewer (a gnu extension) looks at the creation time. The creation time of the files you're using as the comparison will be just now, so won't work. If you want to compare against the regular modification time (which is the time shown in your ls output), then just use -newer. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
[OpenIndiana-discuss] 640k should be enough for anybody
For anybody who's free in London next Wednesday 18th, I'm going to be talking at the UKOUG Solaris SIG on "640k Should be Enough for Anybody: Experimentation on the Fringes of Solarishness" (Sort of based on some of my recent blog posts.) Details of the event and registration here: http://www.ukoug.org/events/ukoug-solaris-sig-nov/ To be clear - it's free to register, free to attend. The website just doesn't have the logic to switch off the prompts when the charge is zero. As I understand it, this is the last SIG meeting in the old style, with it being replaced next year. Which is a bit of a shame, but we've had a good run. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Newer versions of Firefox -- What About Chrome?
> I have 31.8.0esr in Tribblix, which works fine for most things. Anything >> else >> I run 40.whatever under Wine, which actually does a very good job. >> > > Hello. > Does FF plugins work for you (IcedTea/Flash)? Is it gcc-compiled? > I'm just using the Oracle builds. (Actually, the S10 build which is more compatible - the SS11 build requires a bit more of gnome than I have.) I have IcedTea if I need it. Not having flash isn't a great loss ;-) (It's the only thing I have that uses pangox, which got removed from pango a while back. OK, I could add pangox-compat, but I'm not sure it's worth it. Not having flash - or plugins generally - makes the browser much more stable, and there isn't that much that needs flash [and there are fairly aggressive moves to get rid of flash anyway). The flash plugin is pretty old anyway, so you wouldn't really want it. I have looked at lightspark and gnash, but they are pretty messy. Or there's Shumway. But it's so much easier to run a Windows version of Firefox under Wine if you actually need some of the horrors.) -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Newer versions of Firefox -- What About Chrome?
On Mon, Aug 24, 2015 at 7:23 AM, Alexander Pyhalov wrote: > Hello. > > j...@m5.chicago.il.us писал 24.08.2015 04:43: > >> >> >> I am running Solaris 11.2 on one of my computers, the latest version >> of Solaris available for download from Oracle (at least, as of a month >> ago, when I downloaded and installed it). It comes with Firefox ESR >> 17.0.11. As far as I can tell, it works fine, I am completely >> satisfied with it. Why is it important to get Firefox 40.0 (this is >> not a rhetorical question, I want to know)? What I would really like >> to get -- not as a replacement for Firefox, but to supplement it -- is >> Chrome. How come I never see anyone discussing porting Chrome? It >> is, after all, an open-source product. Hasn't anyone else been >> interested in getting Chrome working on OpenSolaris-derived systems? >> > > I think that updating Firefox to more or less fresh version is much easier, > as there are at least Oracle patches and pkgsrc work. > Still quite a bit of work, though. I have 31.8.0esr in Tribblix, which works fine for most things. Anything else I run 40.whatever under Wine, which actually does a very good job. > As for porting Chromium, I'd estimate it would be a good thing to do, > but I fear we just lack men skilful and motivated enough to do it > (briefly: it's unreal). > It might be easier to get Webkit running properly - it certainly has been available on OpenSolaris in the past. That would open the door to a number of other browsers that use Webkit, as well as a number of applications that use it for rendering. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ ___ openindiana-discuss mailing list openindiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
