Re: [osol-discuss] External disk 1TB
This suggestion of relabeling the 1TB disks using an EFI label doesn't work for me --- when OpenSolaris produces the disk has 2930277168 blocks, which is too large for a 32-bit kernel error, it doesn't register any block devices for that disk, so there's no way to specify to format -e which disk you'd like to relabel. I was able to temporarily use a 64bit machine to relabel disks with EFI --- this didn't work either; the 32bit machine still produces the same error message after the disk has been relabeled with EFI. According to http://docs.sun.com/app/docs/doc/819-2723/disksconcepts-14#disksconcepts-22 , the only way to support 1TB disks is with a 64bit kernel. -- This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Opensolaris roadmap
Greetings, I've been reading a lot of the threads on here about the future of opensolaris, and it seems that there *might* be a future, depending on whose post you read. While I truly hope for that, I would like a definitive answer, or at least one I could work off of in the future. Right now, my organization would like to implement a long-term data storage solution (particularly with the advantages of ZFS (dedup, snapshotting, ease of management, etc). Given the current circumstances with Oracle, and the lack of mostly 'official' information, we've begun to reach the point where we might decide to implement this service as a linux based solution, and leverage a linux filesystem, especially since the hardware is coming very soon, and we need to get working on it. Now, the idea right now might be to leverage a linux solution in the short term until everything is 'worked out' with opensolaris/solaris, with the possibility of moving to a better platform long-term. So here's my question: What should be expected out of opensolaris (or even solaris) in the future? I realize the possibility of forking the project has been discussed, but also there has been information that internal builds are being worked on, etc. Would it be advantageous to *wait* for an official answer from Oracle, as well as watch to see what happens with opensolaris? The issue is enterprise backing, because if we're going to spend money on hardware, we'd like to have enterprise support/etc. Given that ZFS dedup and other performance (and bug) fixes won't be out until Solaris 11/Next/whateverthehellORaclewantstocallit, would it be a smart bet to stick around and wait? PS: sorry about starting another thread about this, but some of the threads with similar questions have turned into semantics, flame wars, FUD, etc, and all I'd like to see is a least the opinion of a few individuals with some actual insight on the matter... Thanks, I appreciate it; stay thristy... -- This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] slow CIFS performance
Is the server Windows 2003 or 2008? If 2003 both these are likely related: http://www.bitshop.com/Blogs/tabid/95/EntryId/56/OpenSolaris-NAS-problems-with-SQL-Server-backups-on-Windows-Server-2003.aspx http://www.bitshop.com/Blogs/tabid/95/EntryId/78/Bug-in-OpenSolaris-SMB-Server-causes-slow-disk-i-o-always.aspx If Windows 2008 I haven't seen the issue - I haven't ever checked what XP / Vista / Win 7 versions demonstrate the same issues. -- This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Asterisk
I don't believe you HAVE to run zap; you can run a timer module if i remember right. I haven't done a lot with Asterisk in a few years beyond usage.. We were a very early adopter of it and liked it a lot. The zaptel drivers were used early on for several timing functions, especially related to conference calls / meetme. Our Asterisk runs inside an OpenVZ container on Linux and I don't see any zap modules loaded at the host nor container levels. We don't use a lot of the functions any more, generally speaking we've migrated more to skype / IM for most things we initially used Asterisk for - once we have an established relationship with a customer many of them are preferring other solutions than a traditional phone... -- This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Running Large Zones From Earlier Release On Later Released Global Zone
Hello All, Firstly apologies as I am going to use Sun Solaris 10 terminology as this is what I am more familiar with. Is it possible to run large zones that are have been built using Solaris 11/06 update 3 within a global zone of Solaris 05/08 Update 5 and above? Kind Regards, SRG -- This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] eVGA GTX 295 CO-OP version Isn't Recognised Properly
John Martin wrote: Steve wrote: Hi all, Using snv_119 (clean install) also tried under OpenSolaris 2009.06 I used to have a LeadTek WinFast GTX 295 graphics card. It was recognized correctly as a dual GPU nvidia card in nvidia X Server Settings and ran Compiz beautifully. Unfortunately, the card had other issues and I had to exchange it for an eVGA GTX 295 CO-OP (single PCB) card. Now when I log in, (with Compiz enabled) all I get is a white screen. When I spin the desktop cube, it turns and I can see it spin but the faces of the virtual desktops and all windows are white and it spins very slowly. Also, nvidia X Server Settings gives the following error: /You do not appear/ to be using the /NVIDIA/ X /driver/. Please edit your X configuration file (just run `/nvidia/-xconfig` as root), and restart the X server . pci bus 0x0004 cardnum 0x00 function 0x00: vendor 0x10de device 0x05eb nVidia Corporation Device unknown CardVendor 0x10de card 0x0705 (nVidia Corporation, Card unknown) STATUS0x0010 COMMAND 0x0047 CLASS 0x03 0x00 0x00 REVISION 0xa1 Remove any /etc/X11/xorg.conf file you have created by hand or with nvidia-xconfig and do: # update_drv -a -i 'pci10de,5eb' nvidia # reboot -p (or just reboot if you haven't updated to the /dev repository). Thank you very much. # update_drv -a -i 'pci10de,5eb' nvidia was exactly what I needed. Can this be added to future releases? Regards, Steve ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] eVGA GTX 295 CO-OP version Isn't Recognised Properly
LATENCY 0x00 CACHE 0x08 BASE0 0xec00 SIZE 8 I/O BASE1 0xe882 SIZE 1 I/O BASE2 0xe800 SIZE 8 I/O BASE3 0xe482 SIZE 1 I/O BASE4 0xe400 SIZE 16 I/O BASE5 0xfeaffc00 SIZE 1024 MEM BASEROM 0x addr 0x MAX_LAT 0x00 MIN_GNT 0x00 INT_PIN 0x01 INT_LINE 0x0a pci bus 0x0007 cardnum 0x00 function 0x00: vendor 0x168c device 0x0013 Atheros Communications Inc. Atheros AR5001X+ Wireless Network Adapter CardVendor 0x1186 card 0x3a13 (D-Link System Inc, D-Link AirPlus DWL-G520 Wireless PCI Adapter(rev.B)) STATUS0x0290 COMMAND 0x0006 CLASS 0x02 0x00 0x00 REVISION 0x01 BIST 0x00 HEADER 0x00 LATENCY 0xa8 CACHE 0x08 BASE0 0xfebf SIZE 65536 MEM BASEROM 0x addr 0x MAX_LAT 0x1c MIN_GNT 0x0a INT_PIN 0x01 INT_LINE 0x0a pci bus 0x0007 cardnum 0x03 function 0x00: vendor 0x11c1 device 0x5811 Agere Systems FW323 CardVendor 0x1043 card 0x8294 (ASUSTeK Computer Inc., Card unknown) STATUS0x0290 COMMAND 0x0006 CLASS 0x0c 0x00 0x10 REVISION 0x70 BIST 0x00 HEADER 0x00 LATENCY 0x40 CACHE 0x08 BASE0 0xfebef000 SIZE 4096 MEM BASEROM 0x addr 0x MAX_LAT 0x18 MIN_GNT 0x0c INT_PIN 0x01 INT_LINE 0x0f j...@opensolaris:~# Any help would be much appreciated. Regards, Steve ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] snv_120 on a laptop - no keyboard?
I had this exact problem with 119 on several machines. There is a fix that Alan Coopersmith pointed me to. I booted into Failsafe mode and then created a file in /a/etc/X11 called xorg.conf with the three lines Section ServerFlags Option AllowEmptyInput False EndSection Then I rebooted and my keyboard and mouse were OK. The details are at http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6861767 Steve Christensen -- This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] evga GTX 295 CO-OP not recognized as an nVIDIA card
Hi all, Using snv_117 (clean install) also tried under OpenSolaris 2009.06 I used to have a LeadTek WinFast GTX 295 graphics card. It was recognized correctly as a dual GPU nvidia card in nvidia-config and ran Compiz beautifully. Unfortunately, the card had some (other) issues and I had to exchange it for an evga GTX 295 CO-OP card. Now when I log in, all I get is a white screen. When I spin the cube, it turns and I can see it spin but the faces of the virtual desktops and all windows are white and it spins very slowly. Also, nvidia-config thinks it's not an nvidia card and GLX isn't loaded. I've even tried using my own /etc/X11/xorg.conf file (Xorg -configure and tailored manually as well as with nvidia-xconfig) and that doesn't work. My keyboard ends up screwed up typing a random number of jiberish with every keystroke. I've also downloaded the latest nvidia drivers from their Website. Can anyone shed some light on my predicament? Regards, Steve ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] in.ftpd does not show ls output if connected with browser
Philip Torchinsky wrote: I successfully configured ftp service on OpenSolaris (snv_111b x86). It runs, I can connect to an ftp server, and if I connect with command-line ftp client, everything works fine. However, if I connect with browser using URL ftp://u...@my.ftp.server ftp server does not show me directory content (just first line - Index of ftp://u...@my.ftp.server; and the empty page then). It does not depend on browser, I tried Firefox 2.0 and 3.0, Opera 9.64. I connected from OpenSolaris only. in.ftpd is run as in.ftpd -a -d -L and log file always shows that in.ftpd starts ls -lA all the time, regardless of connecting method. What can be an issue? The first thing I'd do is run tcpdump or WireShark to see what's going on, on the wire. Keep in mind, with ftp, the command channel is port TCP21. Data transfers occur on a higher arbitrary port selected by the server. May be something to check out. Cheers, Steve ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Bad Memory
Hi all, I've got a Dell GX280SD that had 2 GB of RAM in it and all was well. Today, I installed 4 GB of RAM (2 x 2 GB). The 4 GB shows up in the BIOS and Windows Vista can see the 4 GB (multibooting) but FreeBSD and OpenSolaris can only recognize 3 GB. I put the snv_115 install CD in and booted from it and even GRUB thinks I've only got 3 GB (3143195K). Can anyone shed some light as to what might be going on? $ uname -a SunOS apollo 5.11 snv_115 i86pc i386 i86pc $ prtconf | awk '/^Memory/{print}' Memory size: 3071 Megabytes Regards, Steve ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] startx with root rights
philo neo wrote: hello, j' have to install opensolaris, everything works well, how to make to have the rights root under gnome. I demare with a traditional user that j' have to create during l' installation, but I n' is not rights on the file, I would like to change the rights of nsswitch.conf, j' made a chmod 666 nnsswitch.conf but he asks me a password root, but Ca did not go, j' have tests with sudo , how to make to start gnome with the rights root philo try opensolaris as the password ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] grub //menu.lst
philo neo wrote: yes i have /rpool/boot/grub/menu.lst --- j' have to install the order sudo to change the rights of menu.lst but I am on a traditional user admin, the sudo t' has; it of the parrametres? i try sudo chmod 666 menu.lst i write my password and i have this message : admin is not in the sudoers file. this incident will be reported SY philo that's because you're not in the sudoers file. As root, you need to run visudo and you can add your regular account to it like: your-user-nameALL=(ALL) NOPASSWD: ALL alternatively, just su to root and use chmod as root. Better to add yourself to the sudoers file and use sudo though. ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] iSCSI + VMWare ESX 3.5
Folks, I've tried finding a definitive answer to this, and have not been able to - please forgive if I'm posting in the wrong area... I have tried in the last week 2008.11, 2008.5, and the most recent version of the Community Edition. In all cases, I have followed the many documents out on the 'net for configuring an iSCSI target server for an ESX cluster I have. In addition, I have also tried an NFS server. In all cases, network performance is bottlenecking for some reason. With iSCSI, it's between 1-4MB/sec. With NFS, it's around 35MB/sec. I've seen various posts with folks having similar problems, but no resolutions. I have ZFS configured underneath, and have tried this on three different high end Dell server systems, all with the same result. I've changed network cards, tried different switches, changed flow control and jumbo frames in places, and tried at least 10 different changes to iSCSI recommended in the various community forums and mailing lists. Is iSCSI performance just broken? Is NFS really limited to 35MB/sec, even without ESX? (I've tested with a local machine mounting NFS). I've isolated that it's not my hardware, not my network, and not my clients (they're performing far better against several other iSCSI implementations - both hardware and software, as well as other NFS implementations). I really, really, really want to use ZFS and all of the other benefits for an 8T array that I need to deploy in a week or so, but not if I can't get iSCSI or NFS performing better. Any resources are appreciated... -s ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] iSCSI + VMWare ESX 3.5
I have many iSCSI volumes on my X4540, on a stock configuration, the performance sucks. You really need a DRAM based ZFS log device, or to disable the ZIL (which is what we did). Otherwise, NFS and iSCSI performance will be awful. Since my last set of testing, I ended up installing NexentaStor on the same hardware, disabled the ZIL, and did some other tweaks, and am seeing performance on par or *exceeding* my much more expensive purpose-built Equallogic array. Even with disabling the ZIL with OpenSolaris, I still had far less performance than I'm seeing here. I have yet to use the COMSTAR iSCSI version in NexentaStor as it's not working in 1.1.5 (supposed to be fixed next week in 1.1.6) but that's supposed to be even faster. Considering trying NexentaCore to see if the same performance can be had there as well. ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] wireless browsing is slow
We'd need a little more information about your setup and testing results in order to provide some form of response: Q) What OS are you running? (uname -a) Q) What system do you have (Laptop/Desktop Make Model) Q) What network cards do you have in the system (ifconfig -a) Q) What wifi router/modem do you have? When you say Wifi is slow please define what you mean by slow? What tests are you doing to quantify the 'slowness'? As you probably know, wifi depends on the environment and distance from the router/wifi hot spot. Any obstacles will reduce the signal strength thus giving you degraded performance. [pfexec] dladm show-wifi will give you the current wifi speed between you and your wifi hotspot. Have you tried doing some speed tests when connected via wifi and when connected via the cable? If so what differences do you see? http://www.speedtest.net/ is just one of many out there. Another test would be to download a large file (DVD ISO for example) either from the internet or from another system on the local network. This will test sustained throughput and give you a good indication of the true speed of your network. Obviously when using a local host you'd expect faster speeds using the LAN connection Vs the Wifi. ping -s remote_host is always a good test to understand the round trip times of packets. If you choose a host out on the internet and run the ping tests over your wifi and LAN connections what response times do you get? Any dropped packets? What response times do you get when pinging your default gateway? traceroute remote_host is another good test to show where the delays are. netstat -in will show any errors on the interfaces so it's worth keeping you eye on that whilst doing the tests to see if any of the error counts are increasing. If you're not already on the latest OS release I'd recommend upgrading and see if the problem remains. Hope this helps. -- This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Dell 2950 + MD 1000 + iSCSI and NFS
Thanks for your reply. Actually NFS would be a much more elegant and simple solution, if your clients are NFS capable, even for example for database workloads (did you know that Oracle has a built-in NFS client/accelerator of her own?) This is for a VMware ESX cluster, so iSCSI is the primary method of connecting to it. I've used NFS with ESX clusters as well, but it's just not as fast as iSCSI, and I need fast... Good approach. Try to configure MPxIO, so you have load balancing and multipathing over both controllers. See http://docs.sun.com/ for details on how to configure MPxIO. I will look into that - I'm not sure how that would help, though, if each controller port is isolated and connected to it's own set of disks... RAID-Z. Actually, in your case, it should be RAID-Z2. As to why, read below... Thanks for the links and background - RAID-Z2 does seem like a good choice. -- This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Dell 2950 + MD 1000 + iSCSI and NFS
Hello, I have a set of questions as to a design for an iSCSI and NFS system on the following hardware: Dell 2950 with Dual Quad Core processors, 8GB RAM, 2x 146GB disks with RAID1. Opensolaris is loaded on the 2 RAID1 disks, and works fine. Add-on 4 port network card for trunk, etc. I have two PERC5/E controllers, and two MD1000 arrays, each with 15x 400GB 15K RPM SAS drives in them. I would like to serve primarily iSCSI from this to a VMware cluster, and maybe some NFS, although that is not as important. I would like either RAID5 or RAID10 - I do need to maximize my space while still having good redundancy. I've used Equalogic, NetApp, and many other purpose built arrays, and also have a Dell NX1950 storage array with an MD3000 and multiple MD1000's and have a decent amount of experience with iSCSI and all the RAID/IO/MULTIPATH world, just not with Opensolaris... I've read hundreds of posts out there about approaches, but many are hardware specific, and I haven't found any specific examples for my hardware *exactly* I think the best approach would be: Use both PERC/5E controllers. Setup both MD1000 as JBOD and not use any of the horrible RAID capabilities Connect all four ports from the PERC/5E's individually to the MD1000's Use software RAID (is this raidz?) in Opensolaris and create some RAID sets The question then becomes - how many RAID sets - should I not span the individual ports from the PERCs to the MD1000's - I believe that port 0 on the MD1000 addresses drives 1-7, and port 1 addresses 8-15. I assume it's best not to span a RAID set across those boundaries for performance reasons. Anything else that I should be aware of? Thank you -- This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] how to register new install?
I have installed zero patches on this box - should I install the Recommended Solaris Patch Clusters If yes which one? Solaris 10 x86 05/08 bash-3.2# cat /etc/release Solaris Express Community Edition snv_99 X86 Copyright 2008 Sun Microsystems, Inc. All Rights Reserved. Use is subject to license terms. Assembled 23 September 2008 bash-3.2# showrev Hostname: rincon Hostid: 38b2308a Release: 5.11 Kernel architecture: i86pc Application architecture: i386 Hardware provider: Domain: ** Kernel version: SunOS 5.11 snv_99 bash-3.2# cat registration.properties userName=** password=** hostName= subscriptionKey= portalEnabled= proxyHostName= proxyPort= proxyUserName= proxyPassword= bash-3.2# /usr/sbin/sconadm register -a -r /registration.properties sconadm is running Authenticating user ... failed registration! bash-3.2# tail -5 /tmp/basicreg20081018124353757.log java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:907) java.lang.Thread.run(Thread.java:619) 18-Oct-2008 12:44:01 PM com.sun.cns.basicreg.cacao.SWUPOMCacaoAdapter getRegisterResult INFO: SWUPOM Fault: Failed to register asset with AMS. Error message is :com.sun.scn.offering.prom.AssetRegistrationException: RemoteException bash-3.2# ping sunsolve.sun.com sunsolve.sun.com is alive -- This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] sun qfe card
I recently built a new peecee with the intention of running Solaris 10. It seems that Solaris 10 x86 does not include support for the Sun Quad Fast ethernet card. Is this card supported under OpenSolaris or any of the variants? Thanks -- This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Best practice recommendations for ZFS + Virtualbox
Greetings. I'm currently looking into creating a better solution for my combination of Sun xVM Virtualbox and ZFS. I have two 500g sata drives configured into a zpool. I've used virtualbox for awhile, as well as zfs, so I am familiar with their functionalities. My main question, is more of a theory or 'best practice' based concern with the two entities. Right now I have a zfs filesystem that holds all of my virtualbox vdi files, as well as the machine files. This really isn't a space saving situation, because the use of zfs clones/snapshots encompasses the entire zfs filesystem (tank/xvm), so I was wondering if any of you had a better solution. I usually create 'gold' images of virtual machines after successful OS installation + vbox tools. I would then use the virtualbox clonevdi commands to create more vdi files when I'd like to test different situations. I don't like the virtualbox snapshots as much as fresh vdi files. As of right now, I was thinking of creating a 'gold image' zfs filesystem (tank/xvm/VDI/gold), and then the snapshot+clone techniques to create something like: tank/xvm/VDI/clones. The problem is that if I ever add a new 'gold vdi file', it does not effect the clones, so I won't be saving the disk space when I clone that vdi file. I am certain I'll be testing more OS's than the current ones, so scalability is a concern. Anyone have a similar setup, but with better results, or maybe something I could use as an idea? Thanks in advance, Steve -- This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] installation experience
I installed nevada build 91 x86 last night and I wanted to share my experience. I have an old Asus A8N-SLI deluxe motherboard with a plextor sata dvdrom drive. I had read that the neither built-in sata controllers worked well (or at all) with opensolaris so I bought a controller with sil3124 chipset which was said to work with opensolaris. I downloaded the dvd image burned it to dvd, and went to work. While I had the hard drives connected to the sil3124 controller, I had the dvd plugged into the nv sata onboard controller, just to see how far it would get in the install. I set the bios to boot from the plextor, it booted, grub told it to start the install, it reported an error about not being able to find the java software and dropped to a prompt. I looked up the error and found that it was a problem with loading the x86 miniroot. Not a big surprise since I had read about issue with the built-in controller. I next tried the other built-in controller, which was a sil3114 chipset. This also was supposed to have problems, but I wanted to see for myself. So, I boot up, grub tells it to install, it reports the error about not finding java software, and dropped me to a prompt. Not a surprise. I then plugged the plextor into the sil3124 controller that I had bought. I booted up, grub told it to try to install, it again gave me an error about not finding java software, dropped me to a prompt. This was a concern, since this chipset was supposed to work. I then thought about putting the install files onto a usb flash drive and seeing if I could install that way. Unfortunately, grub's menu.lst file posed a problem. the install_media is set to cdrom and I didn't now what to call the media for a flash drive, so that idea went out the window, and I next tried an external usb connected dvdrom. I plugged the dvdrom into the usb, set the bios to boot from it, and booted up. This didn't get far at all. I got dropped to a grub prompt. I assume grub thought install_media=cdrom meant the internal device, tried to use it, and gave up, but I don't know for sure. I had installed RHEL5 on a different machine using this external dvdrom, but I don't know what that menu.lst file looked like. Anyway, after scratching my head, I decided to yank out an older internal ide dvdrom drive from another computer and try that. I installed the dvd drive, told the bios to use it to boot from, and booted up. Instead of giving me that java software not found error, it told me that this wasn't a hsfs filesystem and dropped me to a prompt. I used fstyp to query it and fstyp reported that it was hsfs. I then tried mount -F hsfs and mount said it wasn't hsfs. Not to be undone, I decide to download the cdrom image files and burn them to cds instead of the dvd image. So, 6 cds in hand, I started to install opensolaris for the upteenth time. I stuck cd1 in the drive, it booted up, I selected solaris interactive, and instead of giving me java software not found or this is not hsfs, it actually started the install. 5 more cds and a reboot later, I had opensolaris finally installed. I was excited until I found that / was ufs and not zfs. The reason I installed build 91 was for the zfs root install. I looked it up and found that to get zfs root installed, you need to either select the text install or jumpstart. So, once more into the breach. I stuck cd1 back in, this time I select 3 (text based install), and it starts up the java gui and had the old familiar blue and red text install in a window. Why does the text install require java? Why can't it be full screen text like it was when I was installing solaris 2.6? Anyway, I selected zfs instead when asked what filesystem I wanted (I don't understand why I get the prompt in the text install but not the gui one), and 5 more cdroms later, I have opensolaris with zfs /. What fun. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] open solaris installed to native linux partition?
Question 1: I have been searching for information as to weather open solaris will install to a native linux partition, or ext2 file system? Reason is so i can setup to read/write from either linux or solaris. As is, i have only been user of linux for 12mths, so please excuse my ignorance. Question 2: Will open solaris detect the other os (pclinuxos)? At present i am dual booting from grub (pclinuxos 2008) and have windows xp as the other os. I intend to install to the present ´c´drive (first slice), and understand the bootloader (grub) will be overwritten. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Zimbra on Solaris now available
New Update: TEST8 released today! This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Zimbra on Solaris now available
Just thought I would recognize the efforts of someone who has decided to build Zimbra on Solaris and it's working great! Site can be found here: http://varlogmessages.vroomvroom.org Hope this will become supported officially by Zimbra. If so, this could bring some good competition for Exchange. S This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Storage driver developer
Hi Artem, I have a technical question here but can't find an answer. In Linux, it provides spin_lock_irq() and spin_unlock_irq(). Does Solaris provide similar functions? Thanks. Steve Chang HighPoint Technologies, Inc. 408-240-6115 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Thursday, January 24, 2008 6:33 PM To: Steve Chang Cc: opensolaris-discuss@opensolaris.org Subject: Re: [osol-discuss] Storage driver developer Steve Chang wrote: I am new to Solaris. I am looking for some information of how to develop a company's SATA RAID product driver for Solair. Is there any one acn point out where can I find those documents - compile, driver model, driver update ..? It's quite simple, you'll only need three things: 1. Study the Writing Device Drivers book: http://docs.sun.com/app/docs/doc/816-4854 2. Inspect code for similar drivers: http://src.opensolaris.org/source/ 3. Take your time doing 1 2. If you still have questions after that, send them to [EMAIL PROTECTED] -Artem ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Storage driver developer
I am new to Solaris. I am looking for some information of how to develop a company's SATA RAID product driver for Solair. Is there any one acn point out where can I find those documents - compile, driver model, driver update ..? Thanks This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Solaris 10 Update 4 questions
On 7/6/07, Shawn Walker [EMAIL PROTECTED] wrote: Those aren't there yet as the new installer has not been adopted yet. It's safe to say those are a future release. Plus, since I know SPARC support isn't ready, I know Sun wouldn't do that :) Does anyone know if Live Update will support zfs rooted zones? TIA Steve ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Kernel message 'out of disk space' ?
Hello All Here is some more info as I am back on the server now. The issue is with [b]/app[/b] [b]/var/adm/message[/b] Jun 26 09:34:18 server ufs: [ID 845546 kern.notice] NOTICE: alloc: /app: file system full Jun 26 09:35:00 server ufs: [ID 213553 kern.notice] NOTICE: realloccg /app: file system full [b]#df -k[/b] . . /dsk/c0t2d0s745851027 41140424 471060390%/app [b] # df -h[/b] ... ... /dsk/c0t2d0s7 44G39G 4.5G90%/app [b]# df -F ufs -oi[/b] /dsk/c0t2d0s0 236169 113247117% / /dev/dsk/c0t2d0s4 17402 483462 3% /var /dev/dsk/c0t2d0s6 10889 249847 4% /home /dev/dsk/c0t2d0s71228172 429298022% /app [b]# df[/b] (/dev/dsk/c0t2d0s7 ): 9421206 blocks 4292980 files Additional information . I can touch a file I can make a file no larger than 7k but can create multiple files examples :- [b]# mkfile 8198k testf[/b] Could not set length of testf: No space left on device [b]# mkfile 7k testf # mkfile 7k testf1 # mkfile 7k testf2 # mkfile 7k testf3 # mkfile 7k testf4[/b] [b]# ls -lart[/b] -rw--T 1 root root7168 Jun 26 10:02 testf1 -rw--T 1 root root7168 Jun 26 10:02 testf2 -rw--T 1 root root7168 Jun 26 10:02 testf3 -rw--T 1 root root7168 Jun 26 10:02 testf4 Not sure if this is connected but the 'T' sticky bit is on which restricts execution right? But on running the following command the /app directory does not have a sticky bit set ? [b]# find / -type d -perm -1000[/b] I also executed fsck which reported no errors:- [b]# fsck /dev/dsk/c0t2d0s7[/b] Regards SRG This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Kernel message 'out of disk space' ?
Thanks for your help Casper . Sorry about the verbose post but I thought as much info as possible would help! Thanks Again! SRG This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Kernel message 'out of disk space' ?
Hello all, I am running Solaris 10 and am getting 'out of disk space' messages from the kernel but a df -k show that only 90% is used and that around 4Gb is available. I cannot write to this partition though! What could be the cause of this? Thanks SRG This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Kernel message 'out of disk space' ?
Thanks Casper... I did a basic fsck but it did not report any errors. I also upped the file descriptors in an attempt to resolve the issue. As you mentioned I have the former situation where I cannot create a file which does sugest an inode issue. I only got to look at this system today (new job :-) ) I am wondering if there are two many files 4,100,000 for standard block size but have not done the math as yet. Oh and it is UFS. Thanks Again SRG This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Unable To Install DBI::mysql on Solaris 10
Thanks for your replies I am still having trouble with this. Here is what I have done Downloaded and installed perl / mysql / gcc / make and associated support packages from sunfreeware.com. These all install into /usr/local I have set up ${PATH} as follows :- [b] /usr/local/bin:/usr/local/mysql/bin/:/usr/sbin:/usr/bin:/usr/ccs/bin[/b] And ${LD_LIBRARY_PATH} as :- [b]/usr/local/mysql/lib/mysql/:/usr/local/lib[/b] I can make the DBD:mysql with no errors but all tests fail! When using force directive to instal DBD:mysql and executing a test script I get a Segmentation Fault (core) when it atempts to connect to the database. Bellow is some output from make test with a debug of 12 set :- export DBI_TRACE=12 [b]# make test[/b] PERL_DL_NONLAZY=1 /usr/local/bin/perl -MExtUtils::Command::MM -e test_harness(0, 'blib/lib', 'blib/arch') t/*.t t/00base.DBI 1.56-nothread default trace level set to 0x0/12 (pid 561) at DBI.pm line 271 via 00base.t line 28 install_method DBI::db::get_info, flags 0x2a00, usage: min 2, max 2, '$info_type' install_method DBI::db::take_imp_data, flags 0x1, usage: min 1, max 1, '' install_method DBI::db::disconnect , flags 0x10c00, usage: min 1, max 1, '' install_method DBI::db::selectrow_array, flags 0x2000, usage: min 2, max 0, '$statement [, \%attr [, @bind_params ] ]' install_method DBI::db::tables , flags 0x2200, usage: min 1, max 6, '$catalog, $schema, $table, $type [, \%attr ]' install_method DBI::db::quote_identifier, flags 0x0430, usage: min 2, max 6, '$name [, ...] [, \%attr ]' install_method DBI::db::clone , usage: min 1, max 2, '[\%attr]' install_method DBI::db::quote , flags 0x0430, usage: min 2, max 3, '$string [, $data_type ]' install_method DBI::db::type_info , flags 0x2200, usage: min 1, max 2, '$data_type' install_method DBI::db::statistics_info, flags 0xaa00, usage: min 6, max 7, '$catalog, $schema, $table, $unique_only, $quick, [, \%attr ]' install_method DBI::db::selectrow_arrayref, flags 0x2000, usage: min 2, max 0, '$statement [, \%attr [, @bind_params ] ]' install_method DBI::db::begin_work , flags 0x0400, usage: min 1, max 2, '[ \%attr ]' install_method DBI::db::last_insert_id, flags 0x2800, usage: min 5, max 6, '$catalog, $schema, $table_name, $field_name [, \%attr ]' install_method DBI::db::foreign_key_info, flags 0xaa00, usage: min 7, max 8, '$pk_catalog, $pk_schema, $pk_table, $fk_catalog, $fk_schema, $fk_table [, \%attr ]' install_method DBI::db::primary_key , flags 0x2200, usage: min 4, max 5, '$catalog, $schema, $table [, \%attr ]' install_method DBI::db::commit , flags 0x0c80, usage: min 1, max 1, '' install_method DBI::db::ping, flags 0x0404, usage: min 1, max 1, '' install_method DBI::db::selectall_arrayref, flags 0x2000, usage: min 2, max 0, '$statement [, \%attr [, @bind_params ] ]' install_method DBI::db::type_info_all, flags 0x2a00, usage: min 1, max 1, '' install_method DBI::db::do , flags 0x3200, usage: min 2, max 0, '$statement [, \%attr [, @bind_params ] ]' install_method DBI::db::selectcol_arrayref, flags 0x2000, usage: min 2, max 0, '$statement [, \%attr [, @bind_params ] ]' install_method DBI::db::prepare_cached, flags 0xa200, usage: min 2, max 4, '$statement [, \%attr [, $if_active ] ]' install_method DBI::db::rows, flags 0x0004 install_method DBI::db::rollback, flags 0x0c80, usage: min 1, max 1, '' install_method DBI::db::column_info , flags 0xaa00, usage: min 5, max 6, '$catalog, $schema, $table, $column [, \%attr ]' install_method DBI::db::table_info , flags 0xaa00, usage: min 1, max 6, '$catalog, $schema, $table, $type [, \%attr ]' install_method DBI::db::primary_key_info, flags 0xaa00, usage: min 4, max 5, '$catalog, $schema, $table [, \%attr ]' install_method DBI::db::prepare , flags 0xa200, usage: min 2, max 3, '$statement [, \%attr]' install_method DBI::db::preparse install_method DBI::db::connected , flags 0x0004, usage: min 1, max 0, '' install_method DBI::db::data_sources, flags 0x0200, usage: min 1, max 2, '[\%attr]' install_method DBI::db::selectall_hashref, flags 0x2000, usage: min 3, max 0, '$statement, $keyfield [, \%attr [, @bind_params ] ]' install_method DBI::db::selectrow_hashref, flags 0x2000, usage: min 2, max 0, '$statement [, \%attr [, @bind_params ] ]' install_method DBI::dr::default_user, usage: min 3, max 4, '$user, $pass [, \%attr]' install_method DBI::dr::data_sources, flags 0x0800, usage: min 1, max 2, '[\%attr]' install_method DBI::dr::disconnect_all, flags 0x0800, usage: min 1, max 1, '' install_method DBI::dr::connect_cached, flags 0x8000, H 3, usage: min 1, max 5, '[$db [,$user [,$passwd [,\%attr' install_method DBI::dr::connect , flags 0x8000, H 3, usage: min 1, max 5, '[$db [,$user [,$passwd [,\%attr' install_method DBI::st::more_results, usage: min
[osol-discuss] Re: Unable To Install DBI::mysql on Solaris 10
[b]Update :--[/b] I thought I'd try to statically link the DBD::mysql module. Not sure if this is the correct way to do this but heres what happened :- [b]# perl Makefile.pl[/b] Edited Makefile and changed LINKTYPE from dynamic to static [b]#make # make test_static[/b] rm -f blib/arch/auto/DBD/mysql/mysql.so LD_RUN_PATH=/usr/local/lib:/usr/lib:/usr/local/ssl/lib /usr/local/bin/perl myld gcc -G -L/usr/local/lib dbdimp.o mysql.o -o blib/arch/auto/DBD/mysql/mysql.so \ -L/usr/local/lib -R/usr/local/lib -R/usr/lib -L/usr/lib -R/usr/openwin/lib -L/usr/openwin/lib -L/usr/local/ssl/lib -R/usr/local/ssl/lib -L/usr/X11R6/lib -R/usr/X11R6/lib -L/usr/local/mysql/lib/mysql -lmysqlclient -lz -lposix4 -lresolv -lgen -lsocket -lnsl -lm -L/usr/local/ssl/lib -lssl -lcrypto \ chmod 755 blib/arch/auto/DBD/mysql/mysql.so make -f Makefile.aperl perl make[1]: Entering directory `/.cpan/build/DBD-mysql-4.004-GwLB5D' make[1]: `perl' is up to date. make[1]: Leaving directory `/.cpan/build/DBD-mysql-4.004-GwLB5D' PERL_DL_NONLAZY=1 ./perl -MExtUtils::Command::MM -e test_harness(0, 'blib/lib', 'blib/arch') t/*.t t/00base.ok t/10connect..ok t/20createdrop...ok t/30insertfetch..ok t/35limitok t/35prepare..ok t/40bindparamok t/40bindparam2...ok t/40blobsok t/40catalog..ok 16/77 skipped: various reasons t/40listfields...ok t/40nullsok t/40numrows..ok t/41bindparamok t/41blobs_prepareok t/42bindparamok t/50chopblanks...ok t/50commit...ok 14/30 skipped: various reasons t/60leaksskipped all skipped: $ENV{SLOW_TESTS} is not set t/70takeimp..skipped all skipped: test feature not implemented t/75supported_sqlok t/80procsok t/insertid...ok t/multi_statementok t/param_values...ok t/prepare_noerrorok t/texecute...ok t/utf8...ok All tests successful, 2 tests and 30 subtests skipped. Files=28, Tests=578, 4 wallclock secs ( 2.19 cusr + 0.40 csys = 2.59 CPU) So what the hell is going wrong? I guess it points to dynamic libraries being the issue so if so is this a Solaris 10 issue? Any help appreciated !! SRG This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Unable To Install DBI::mysql on Solaris 10
Hello All I appreciate that this may be off topic but I am only having this issue with Solaris 10 and quite frankly I appear to becoming to the end of my rope. So in desperation I am posting this here in the hope that someone can help me. I have been trying to install DBI::mysql on a Solaris 10 box. Using MCPAN :- [b]bash-3.00# perl -MCPAN -e shell[/b] CPAN: File::HomeDir loaded ok (v0.65) cpan shell -- CPAN exploration and modules installation (v1.9102) ReadLine support enabled [b] cpan[1] install DBD::mysql[/b] Running install for module 'DBD::mysql' Running make for C/CA/CAPTTOFU/DBD-mysql-4.004.tar.gz Has already been unwrapped into directory /.cpan/build/DBD-mysql-4.004-kMsgI8 Could not make: Unknown error Running make test Can't test without successful make Running make install Make had returned bad status, install seems impossible Manualy : - [b] bash-3.00# pwd[/b] /.cpan/build/DBD-mysql-4.004-kMsgI8 [b]bash-3.00# make[/b] cc -c -I/usr/perl5/site_perl/5.8.4/i86pc-solaris-64int/auto/DBI -I/usr/sfw/include/mysql -xstrconst -mt -DDBD_MYSQL_INSERT_ID_IS_GOOD -g -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -D_TS_ERRNO -DVERSION=\4.004\ -DXS_VERSION=\4.004\ -I/usr/perl5/5.8.4/lib/i86pc-solaris-64int/CORE dbdimp.c dbdimp.c, line 498: warning: implicit function declaration: parse_number dbdimp.c, line 3215: warning: implicit function declaration: mysql_warning_count dbdimp.c, line 3429: prototype mismatch: 3 args passed, 4 expected dbdimp.c, line 3593: prototype mismatch: 3 args passed, 4 expected dbdimp.c, line 3931: prototype mismatch: 3 args passed, 4 expected dbdimp.c, line 3945: prototype mismatch: 3 args passed, 4 expected dbdimp.c, line 4264: prototype mismatch: 3 args passed, 4 expected dbdimp.c, line 4291: prototype mismatch: 3 args passed, 4 expected dbdimp.c, line 4303: prototype mismatch: 3 args passed, 4 expected dbdimp.c, line 4465: prototype mismatch: 3 args passed, 4 expected cc: acomp failed for dbdimp.c *** Error code 2 make: Fatal error: Command failed for target `dbdimp.o' Anyone have any ideas as I am pulling my hair out ;-) Regards SRG This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Unable To Install DBI::mysql on Solaris 10
Is this bad coding etiquette ? [u]dbdimp.c[/u] 4459if (!my_login(h, imp_dbh)) 4460{ 4461 [b]do_error(h, mysql_errno(imp_dbh-mysql), mysql_error(imp_dbh-mysql)[/b] 4462 #if MYSQL_VERSION_ID = SQL_STATE_VERSION 4463 [b], mysql_sqlstate(imp_dbh-mysql));[/b] 4464 #else 4465 [b] );[/b] 4466 #endif This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: RE: backspace key not working on Java
On 5/25/07, Anne [EMAIL PROTECTED] wrote: The sad thing about this is that I still cannot get the backspace working on the Java environment. UGH what a major pain. And the DEL key acts the same way. The ONLY thing that works for backspace is continually hitting CTRL+H The Java interface is so much nicer than the CDE, but just can't use it without backspace. Thanks to everyone for trying to get this working. Anne I really think this serves as a good reminder to those who are against providing a method of changing backspace behavior for users which expect PC like semantics. In the end, it really doesnt matter what is technically correct, it comes down to user expectation and needs. Expecting the millions of PC keyboards out there to be considered obsolete when a very large chunk of Solaris users are running on PC hardware is just silly. Steve ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: And that would break... what, exactly? (Re: Sun to make Solaris more
On 5/21/07, Brian Gupta [EMAIL PROTECTED] wrote: You are missing something very significant. There are times when doing so, might give you a significant competitive advantage. Whether that is time to market, efficiency gains, or some other advantage, there are many cases that make strong business sense for having paid staff do customization of third party source. If you forget that business needs are the primary driver of IT specifications, you are going to eventually find yourself out of employment. (You aren't just doing technology for what you think is cool. You have to be doing what the business needs, not just what you *think* the business needs). -Brian Perhaps I am being a little dense here, but what does any of this have to do with breaking existing standards and compatibility? Isn't this what businesses are subscribing to when they pay for a given piece of software and/or support? ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Inexpensive/quiet server hardware for running OpenSolaris.
On 5/18/07, Brian Gupta [EMAIL PROTECTED] wrote: http://www.tyan.com/product_board_detail.aspx?pid=181 Questions: Is this unit 64 bit? Assuming you put in a 64bit processor, yes. This is a socket 479 motherboard, so you will need to look at mobile cpu's. I would suggest a lowend core 2 duo if you are looking for a 64bit processor. Also, is their a supplier where I can purchase a barebones system? (Or if not what rackable case should I use?) Look around. The board is a FlexATX - I decided to go with a simple low profile case instead of a rack mount (I needed to support 5 1/4 bays), but I was able to find a few decent rack mountable cases that ranged from 1U to 3U. Chenboro and Supermicro are both good vendors for cases (rackmountable and low profile) Thanks, Brian ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Inexpensive/quiet server hardware for running OpenSolaris.
Chenboro and Supermicro are both good vendors for cases (rackmountable and low profile) Sorry for the typo, that should have read: Chenbro http://www.chenbro.com/ ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Re: Sun to make Solaris more Linux like
On 5/18/07, Brian Gupta [EMAIL PROTECTED] wrote: Why don't we first focus on those applications that don't exist in Solaris currently? Also, in parallel, we should work on an updated packaging format. Whoa... slow down Tiger. SVR4 is by far a more complete and cohesive packaging format than is available on any other UNIX or Linux distribution. Essentially there is no technical merit in tossing SVR4 packaging - not by a long shot. That said, SVR4 could certainly use a little help with respect to usability (pkgadd -D anyone?). This was posted earlier, but please take an indepth look into the packaging guide and see what is provided out of the box - I think you will agree this beats the pants off of RPM, .deb, and POTS (plain old tarballs). These two things will do much to lay the foundations for addressing the out of box completeness gap. I would like to invite you to take a look at what Blastwave has accomplished with respect to integrating a GNU runtime. You might be pleasantly surprised to see your work already done for you. -Brian Steve ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Displaying Primary Network Interface?
Hello All, In the situation that a server has multiple physical interfaces how can one display the Primary Interface NAC only? I could use :- # ifconfig -a (but this shows all interfaces including lo | virtual) # netstat -i (Again multiple and does not show MAC) # dladm show-link (but again no primary ether and displays multipe NICS) From OpenBoot i can issue :- OK banner (but I need to do this from a running Solaris box I thought /etc/path_to_inst may have help ... but not really. # showrev (gives me hostid / os / arch but no ether) Cheers This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Displaying Primary Network Interface?
The following works but I am sure ther must be a command :- Ignore lo print first interface name plus ether address and ignore any that follow. # ifconfig -a |grep -v lo | perl -ne 'if (/(^\w+0):\s+/) {print $1} if (/\s+ether(.*)/) { print $1.\n; exit} ' Cheers. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Sun to make Solaris more Linux like
f) leave Solaris precisely how it is, and add additional personalities (ie: /usr/gnu) to support disparate runtimes with sensible aliases for new users. On 5/17/07, Doug Scott [EMAIL PROTECTED] wrote: [EMAIL PROTECTED] wrote: I agree with much of your post. Though rather than using the execution PATH, why not use the function call 'isatty' to change the behavior only in an interactive session, and leave scripts un-touched. As it this is already used by several commands in Solaris, where is it a problem with simple commands like df? But that means that you can't easily write a script which expects new output. Ok before the call was for compatibility to not break existing scripts. Now people want to not break new scripts. Maybe we need a vote. Do we want - a) Status Quo - Break new scripts, command line needed arguments for humans. b) Linux Version - Possibly breaks old scripts. Human friendly. c) Alternative 1 - Old scripts work. New scripts need to be adjusted. Human friendly output. d) Alternative 2 - All of above. Binaries to suite everybody. New scripts need to change for new path. Possible path conflicts. A mixture of Paths and Environment variables fixes all peoples problems. Diskspace is cheap! e) Distribution dependant. i.e. Open Solaris keeps code for all alternatives. f) ??? Please make a choice. The solution is only code :) Doug ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Sun to make Solaris more Linux like
I have perhaps a more general question. Why not fully flesh out the /usr/gnu personality and just provide users with either a default .profile/.cshrc (or *gasp* new user documentation) rather than wreaking havoc on 10+ years of compatibility? Why not provide default aliases for these new users as well? Its pretty simple to change the default behavior of a given command without actually impacting the rest of the system. Personally, I am not terribly worried about GNU/Linux users who have a hard time moving over to Solaris. I find the goal to be laughable and I would far more upset if my OS of choice was diluted to support a marketing initiative. GNU/Linux has had a long and illustrious history of ignoring standards when it suits them. I have no interest in seeing Solaris travel down that same path. IMHO - Solaris is not Linux, nor should it ever be. Steve On 5/16/07, Nils Nieuwejaar [EMAIL PROTECTED] wrote: On Tue 05/15/07 at 17:58 PM, [EMAIL PROTECTED] wrote: My larger point: There's a big difference between ABI breakage (something Linux does regularly, and you're absolutely right to criticize it for doing that--but that's OPPORTUNITY) and tweaking the default output of a utility that may be used in a handful of scripts. No, there isn't. Whether it's a C interface, a system call interface, or a scripting interface, the impact is exactly the same. In each case you are changing the behavior of the system in an unexpected way, causing new failures for end users. The specific example that has been brought up many times now (I think originally by you, but I'm not certain) is making df -h the default output for df. You/we have no idea how many scripts this tweak will break, how important they are, or how deeply buried inside a customer's infrastructure they might be. The man page for df indicates that this is a Standard interface, which means that we have promised customers that the interface will not change incompatibly unless we rev Solaris' major version number. If we want to change the default behavior of our Standard interfaces, we either have to change Nevada's release number from 5.11 to 6.0, or we're telling customers that they can't trust our interface stability guarantees. If you want to provide a different Solaris environment that explicitly disavows all of these guarantees, there's nothing inherently with that - it just can't be the default environment. Nils ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Sun to make Solaris more Linux like
On 5/16/07, Brian Gupta [EMAIL PROTECTED] wrote: Personally, I am not terribly worried about GNU/Linux users who have a hard time moving over to Solaris. I find the goal to be laughable and I would far more upset if my OS of choice was diluted to support a marketing initiative. How exactly did you learn Solaris? (It's easy to forget). 1) I installed Solaris on my shiny new Sparcstation. 2) I read documentation, I subscribed to mailing lists, and I joined IRC channels. 3) Profit. In essence, I did what (I believe) most other users do when they decide to take on a new operating system - I took upon myself the responsibility of learning something new. Maybe I am in the minority here. Steve GNU/Linux has had a long and illustrious history of ignoring standards when it suits them. I have no interest in seeing Solaris travel down that same path. The opposite could be said about Solaris. A) Solaris has an illustrious history of adopting useless standards, and b) solaris has an illustrious history of not meeting a needs if there is any conflict with a standard. Others have addressed this already on the the list... IMHO - Solaris is not Linux, nor should it ever be. This we agree on. ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] unable to bring up nfs server on solaris 10
Have you tried issuing 'svcadm enable -r nfs/server' ? There may be a dependency lurking somewhere you could have missed. On 5/15/07, jason jin [EMAIL PROTECTED] wrote: solaris 10 11/06 on SUN V240 # cat /etc/dfs/dfstab share -F nfs -o ro,anon=0 /jumpstart #svcadm enable svc:/network/nfs/server:default # svcs -x svc:/network/nfs/server:default svc:/network/tftp/udp6:default has no restarter property group; ignoring. svc:/network/nfs/server:default (NFS server) State: disabled since Tue May 15 14:28:17 2007 Reason: Disabled by an administrator. See: http://sun.com/msg/SMF-8000-05 See: nfsd(1M) Impact: This service is not running. # svcs -d svc:/network/nfs/server:default STATE STIMEFMRI disabled 14:28:15 svc:/network/rpc/keyserv:default online 14:28:19 svc:/milestone/network:default online 14:28:42 svc:/system/filesystem/local:default online 14:28:42 svc:/network/rpc/bind:default online 14:29:30 svc:/network/rpc/gss:default online 15:48:27 svc:/network/nfs/mapid:default offline15:48:26 svc:/network/nfs/nlockmgr:default #svcadm enable svc:/network/rpc/keyserv:default then # svcs -xv svc:/network/rpc/keyserv:default svc:/network/tftp/udp6:default has no restarter property group; ignoring. svc:/network/rpc/keyserv:default (RPC encryption key storage) State: disabled since Tue May 15 14:28:15 2007 Reason: Disabled by an administrator. See: http://sun.com/msg/SMF-8000-05 See: man -M /usr/share/man -s 1M keyserv Impact: This service is not running. svcs -d svc:/network/rpc/keyserv:default STATE STIMEFMRI online 14:28:37 svc:/system/identity:domain online 14:28:42 svc:/network/rpc/bind:default and # cat network-rpc-keyserv:default.log [ Jan 22 16:15:00 Disabled. ] [ Jan 22 16:15:00 Rereading configuration. ] so what is wrong here? how to re-enable it? thanks. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Inexpensive/quiet server hardware for running OpenSolaris.
I have a network attached tape device I built using this board from Tyan. It is a FlexATX so its quite small (only marginally larger than a mini-itx) and is dead silent (under 17db) in full operation and makes use of the newer core 2 chipset (vt-x instruction support) so vmware would run at near native speeds on it. http://www.tyan.com/product_board_detail.aspx?pid=181 On 5/15/07, Brian Gupta [EMAIL PROTECTED] wrote: I am looking for a relatively inexpensive server(s), that can run in my (family) living room. (IE: That don't sound like jet engine's) Ideally one of them would be beefy enough to run vmware server on, so I can have multiple dev environments running concurrently. (If they are cheap enough I could probably forgo this requirement. Does anyone have any suggestions? Thanks, Brian P.S. - It can be Sparc or x86, new or used. ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Which reason will cause thi error?
It sounds like you are having runtime lookup issues. If you are compiling something by hand, make sure that you are passing a -R flag with the library paths you wish to link to at runtime. Do *not*, I repeat, do *not* use LD_LIBRARY_PATH. Hint: This should look a lot like your -L paths. On 5/14/07, Brian Gupta [EMAIL PROTECTED] wrote: Check your LD_LIBRARY_PATH and PATH environmental variables. It generally means that something in the dependancy tree can't be found. On 5/14/07, snow [EMAIL PROTECTED] wrote: Somebody can help me? This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Which reason will cause thi error?
Serves me right for not actually reading the original email... What flags are you passing to modload... -p ? On 5/14/07, Steve Stallion [EMAIL PROTECTED] wrote: It sounds like you are having runtime lookup issues. If you are compiling something by hand, make sure that you are passing a -R flag with the library paths you wish to link to at runtime. Do *not*, I repeat, do *not* use LD_LIBRARY_PATH. Hint: This should look a lot like your -L paths. On 5/14/07, Brian Gupta [EMAIL PROTECTED] wrote: Check your LD_LIBRARY_PATH and PATH environmental variables. It generally means that something in the dependancy tree can't be found. On 5/14/07, snow [EMAIL PROTECTED] wrote: Somebody can help me? This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Sun to make Solaris more Linux like
Well it didnt take long for this to escalate to into an all out flame. Remember Gentlemen, pinkies up! On 5/11/07, Brian Gupta [EMAIL PROTECTED] wrote: On 5/11/07, UNIX admin [EMAIL PROTECTED] wrote: Not at all, but Sun execs have a long history of telling the press long before they talk to the engineers, and half of what the press reports bears no relation to what actually ends up happening in the long run. And why have they not been reprimanded for doing such things? I told the management in my company to go pound sand when they started talking BS. They moved me around, but the difference is, I can easily switch companies, while they can only switch departments. And they know it. Why don't you guys push back??? Management has absolutely no business sticking their noses where they don't belong! You speak with such lofty ideals. You are asking the Sun employees why don't they push back. I'll tell you why. Because their livelihood depends on it. You on the other hand have no such restrictions. I think it would be very noble of you to set an example and stick it the Sun executives, in a way that the Sun employees can't. Those execs will be so screwed when you leave the OpenSolaris community, since you can easily switch communities, while they can only switch departments!! -Brian ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] which iso is for production server
The ISO you should use for your production environment should be the one that passed your testing in your staging environment ;) ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Is the cplus_demangle call mt-safe or async-signal-safe
The cplus_demangle call is mt-safe, guarded by a mutex at the outer level. There is no external state, so I don't see how asynchronous signals could cause a problem. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Getting rid of CD images for future SX:CE deliveries
Getting rid of CDs is not a good idea at least for a while. I deal with users from around the world everyday and there are many who do not have either DVD reading or writing capabilities. (Some don't have CD burning hardware either, but that is getting to be rare.) Surely many of these will not be moving to Opensolaris or even Solaris 10 on their hardware and so maybe the OpenSolaris community should not care, but by not offering CDs you are guaranteed to keep them from trying OpenSolaris or contributing to the effort in any way. Some of these people have only one Solaris system, so using another system to install might not work. Live Upgrade could work though there may be limits to that as well. It is also my experience that there are organizations (oddly enough some very large ones) and individuals in the US who would fit the third world category. Steve Christensen Jason J. W. Williams wrote: Hi Alan, We've removed all of our v20z's so its not an issue for us...but if we still had them it would be painful in extreme circumstances. The ILOM and ELOM both support remote ISO mounting, and the basic BMC on the v20z doesn't. So in an emergency you're stuck with CDs. Best Regards, Jason On 2/23/07, Alan Coopersmith [EMAIL PROTECTED] wrote: Ignacio Marambio Catán wrote: On 2/23/07, Rich Teer [EMAIL PROTECTED] wrote: On Fri, 23 Feb 2007, Alan Coopersmith wrote: Would you have a problem with DVD-only images for SX:CE? Do you have systems you use for OpenSolaris that don't have DVD drives and that you can't netinstall from another system or Live Upgrade from a mounted image? Speaking personally, no I have no issues with this. -- That would probably be a show stopper in many third world countries, but then again, as long as the developer edition has cd images, there shouldn't be a problem. Getting rid of them for Nevada will affect the Developer Edition as well. -- -Alan Coopersmith- [EMAIL PROTECTED] Sun Microsystems, Inc. - X Window System Engineering February 2007 Selection: LSARC Chair of the Month Club ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] (no subject)
set anthenticate ezifna set digest plain ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Project Proposal: OpenSolaris Ports System
All of this certainly up for discussion. Ive registered a channel on irc.freenode.net named #opensolaris-ports so we can discuss this outside of the discuss list for now. Over the next couple of days, I'll get a ports list setup on opensolaris.org.This goes for anyone else who is interested, hop on to the channel and lets starting getting this stuff ironed out.Steve On 10/11/06, David Kleiner [EMAIL PROTECTED] wrote: Steve,What's your take on the metadata (versions, pkg dependencies, stable vs. -devel, that sort of stuff).As an example, is putting all this into mercury (Hg) an option? I know it's python but python is now in /usr/sfw, so it may not be so bad.Or you'd rather keep it all in C? Then there is distribution - svn / Hg / cvs / cvsup (modula) / csup (c)...Where can I help?Cheers,DavidThis message posted from opensolaris.org ___opensolaris-discuss mailing listopensolaris-discuss@opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Project Proposal: OpenSolaris Ports System
Sorry about that, Im a bit guilty of using /usr/sfw as shorthand for the consolodation - i'll qualify next time :)On 10/11/06, Darren J Moffat [EMAIL PROTECTED] wrote:David Kleiner wrote: Steve, What's your take on the metadata (versions, pkg dependencies, stable vs. -devel, that sort of stuff).As an example, is putting all this into mercury (Hg) an option? I know it's python but python is now in /usr/sfw, so it may not be so bad.Or you'd rather keep it all in C? Nit pick:python is in /usr/bin not /usr/sfw/.It is in the SFW consolidation butnot the /usr/sfw file hierarchy.--Darren J Moffat___opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Proposal: OpenSolaris Ports System
Erast,I am more than willing to develop and support this project, however it will depend on the rest of the community to make use of it. I think that would be a good goal to have ultimately. I have had a couple of ideas to support specific dist revisions of the ports repository, which would make official support and patches pretty easy to maintain. SteveOn 10/9/06, Erast Benson [EMAIL PROTECTED] wrote: On Fri, 2006-10-06 at 18:04 -0700, Steven Stallion wrote: Greetings all, One of the largest headaches of running a desktop solaris box is the lack of support for source built/binary packages. Blastwave.org is the current favorite (or so it seems), but it still very lacking in terms of variant support and several packages have fallen by the wayside (the last vlc package was created in 2004). Something I would very much like to see (and I know others would as well) is a BSD-like ports system geared towards a solaris environment. As a long time BSD user, I am of course biased towards the FreeBSD ports system and pkgsrc, but in all honesty, i find darwinports to be far and above the best ports system available. I would like to propose a solaris based ports system based on the darwinports system modified to support sun tools and packaging. I have a significant chunk of time I can donate to this project and I know of others who are willing to participate as well. Thoughts?+1 from me too.The more OSS consolidations the better. Packaging itself is not so muchof the problem, but quality ports are. I'm also hoping to see a distrobuilt on top of it. Any plans? --Erast ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Project Proposal: OpenSolaris Ports System
Danek,Thats great to hear. What else needs to happen make this project official?SteveOn 10/10/06, Danek Duvall [EMAIL PROTECTED] wrote:On Tue, Oct 10, 2006 at 07:28:19AM -0500, Steve Stallion wrote: I most definitely agree - however, I would like to see the project get an actual start before these discussions commence. To be perfectly honest, Im feeling a little guilty for all the traffic I have caused on the -discuss list the last few days ;)Hey, at least it's been largely quite positive!I think people are justexcited by the prospect; I know I am.:)Danek ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Project Proposal: OpenSolaris Ports System
Dave,This may be ignorance on my part, but AFAIK, there are no TCL bindings to sun api's. In my experience, more developers (including myself) are more proficient in C than TCL. While TCL is in /usr/sfw, I would like to see the ports system available to any solaris user whether or not they have the software companion installed (read: binary distribution). SteveOn 10/9/06, Dave Marquardt [EMAIL PROTECTED] wrote: Steven == Steven Stallion [EMAIL PROTECTED] writes: From above:Steven I would like to propose a solaris ports system loosely based Steven on darwinports functionality. There are of course changes ISteven would like to see made: support for different compiler setsSteven (gcc vs. sun), support for solaris packages, and a shift from Steven TCL to C, and the abillity to rely on package dependenciesSteven that were not built with the ports system (ie:Steven SUNWgnome-base-libs).First, I know nothing about darwinports, so I'm acting from ignorance in that respect.I'm more familiar with FreeBSD ports, having been auser of that some years back.But why the proposed switch from Tcl to C?I'm a big Tcl fan, and amalways curious when people propose replacing Tcl and wonder why?Thanks. --Dave MarquardtSun Microsystems, Inc.Austin, TX+1 512 401-1077 (SUN internal: x64077) ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Project Proposal: OpenSolaris Ports System
I completely agree. It does make the dependency list much bigger but there are a couple of options (at least as I see it):1) /usr/sfw is moved into the ports collection (eventually of course), and we all live in a big happy world. 2) We build in alternate dependencies on /usr/sfw package in the event a companion package is not found on the system.This really requires more explanation:lets say that normally we depend on SUNWfoo to build a given package which just so happens to live in the companion. We can build the ports system in such a way that if the SUNWfoo dependency is not found on the system, we can build out an alternate package in the ports collection. Also, we could build in the option of upgrading the revision in place or support multiple revisions (again, more work, but would make users much happier in the long run). Now here is the kicker. Ultimately I would like to see our ports package be used interchangeably throughout the entire system. Yes boys and girls, this means the possibility of installing to /usr or /usr/gnu or any other prefix we desire for a given package. Standards will most definitely need to be set, but imagine being able to actually use a hier that makes sense without require users to add yet more entries to their path. I would really like to see this done in such a way that once a port is built, it works in every way just as if the package came straight from sun. This most definitely needs sun endorsement, but in my mind, this would be one of the greatest strengths to obtaining third party software on solaris. This does complicate things a bit, but imagine having the abillity to upgrade your version of JDS in place without having worrying about existing packages and revisions.Admit it, its a nice thought ;) Cheers!SteveOn 10/9/06, Dennis Clarke [EMAIL PROTECTED] wrote: Dave, This may be ignorance on my part, but AFAIK, there are no TCL bindings to sun api's. In my experience, more developers (including myself) are more proficient in C than TCL. While TCL is in /usr/sfw, I would like to see the ports system available to any solaris user whether or not they have the software companion installed (read: binary distribution).You know, that is one of the key ideas behind Blastwave.If you have a core install of Solaris 8 ( or 9 or 10 or ... ) then you can install the binarypackages and not be dependant on some other thing that may or may not be in/usr/sfw or /usr/local or /usr/foo.The problem is that people have fits when they see that a package CSWfoo drags in the national science museum history of dependencies *because* wecan NOT depend on any damn thing in /usr/sfw or whereever.Totalindependance can mean that you drag a LOT of stuff around.Hence the vast size of the Blastwave package dependency list.Dennis ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Proposal: OpenSolaris Ports System
Agreed. Build recipies are cheap to host, Sun or genunix or wherever would have the master repository, but it would be very easy to build a mirror system with so little space requirements.Admittedly, I would be thrilled to see the Companion folks make use of this once we are ready. Just as Dennis mentioned, I am more interested in software that works (and subsequently makes my life easier) rather than waging a politcal battle. Solaris is a great product. Lets make it even better.SteveOn 10/9/06, Dennis Clarke [EMAIL PROTECTED] wrote: On Fri, 6 Oct 2006, Steven Stallion wrote: Greetings all, One of the largest headaches of running a desktop solaris box is the lack of support for source built/binary packages. Blastwave.org is the current favorite (or so it seems), but it still very lacking in terms of variant support and several packages have fallen by the wayside (the last vlc package was created in 2004). Something I would very much like to see (and I know others would as well) is a BSD-like ports system geared towards a solaris environment. As a long time BSD user, I am of course biased towards the FreeBSD ports system and pkgsrc, but in all honesty, i find darwinports to be far and above the best ports system available. I would like to propose a solaris based ports system based on the darwinports system modified to support sun tools and packaging. I have a significant chunk of time I can donate to this project and I know of others who are willing to participate as well. Thoughts? Steve To me this proposal sounds like a good start, the only big thing it's lacking -- IMO -- is some consideration and discussion of relative positioning vis-a-vis the Companion CD project/forum. (This is especially important because it's being proposed as an opensolaris.org based project.) For example, should this project be housed by the CCD project, or should it be separate? If separate, why?Etc. I don't recall that the man wanted a political battle.More like a thislooks like a good idea, let's do it.So I offered resources for him to getstarted and then if he wants .. he can, well, you know, get started. Who cares where its housed.It can be offshore in Burma or KalamazooMichigan for all we care.Dennis ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Project Proposal: OpenSolaris Ports System
Thats an excellent point Alan - it sounds like /opt/sfw should probably be left on its own. This may be a huge stretch, but what would the implications be of having the consolodation managed and distributed via the ports system? This is how other free OS's typically handle non core packages (ie: FreeBSD et al). Given this, a user would simply need to issue a 'port upgrade' command to download and install the latest patches/software for a given port. There is of course the option of maintaining multiple revisions of the ports repository. ie: stable (sun supported), vs. unstable (unsupported). Cheers! SteveOn 10/9/06, Alan Coopersmith [EMAIL PROTECTED] wrote: Steve Stallion wrote: I completely agree. It does make the dependency list much bigger but there are a couple of options (at least as I see it): 1)/usr/sfw is moved into the ports collection (eventually of course), and we all live in a big happy world.I think there's confusion here, mostly caused by Sun's poor choice of names.The Companion, which installs in /opt/sfw, is the unbundled/unsupported add-oncollection of freeware. It's a poor choice to base dependencies on since it's traditionally only been updated in sync with new Solaris releases, andgets no security or urgent bug fixes outside of that - so Solaris 8 users arestill stuck with a Companion last updated in 2002, bugs, security holes and outdated packages in all.The SFW consolidation, which currently installs much of it's software in/usr/sfw but is moving to /usr/bin /usr/lib during the Nevada developmentcycle, is a fully supported/updated part of Solaris. Depending on that for software tied to a specific Solaris release is not that risky (though there'salways some risk when dealing with upstream communities without compatibilityas a core goal - for instance, see the recent havoc caused by OpenSSL's compatibility-breaking upgrade), but since the older releases of Solaris onlyget security and critical bug fixes, may not have all the features you wantout of newer versions.---Alan Coopersmith- [EMAIL PROTECTED] Sun Microsystems, Inc. - X Window System Engineering ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Net install onto an Ultra 5 with a Linux DHCP server
I just finished installing NV 33 onto an old Ultra 5 without a cdrom. The only real snag was getting the DHCP server to feed it the correct information, because the DHCP server is on a Linux box. I have a Netra T1 acting as the install server and as the boot server. But I didn't want to try running to DHCP servers on the same subnet. Here's what I ended up with in the dhcpd.conf file on my Linux box. Heimdall is the name of the install and boot server, at 10.10.1.131. option spaceSUNW; option SUNW.SrootOptcode 1 = text; option SUNW.SrootIP4code 2 = ip-address; option SUNW.SrootNM code 3 = text; option SUNW.SrootPTHcode 4 = text; option SUNW.SswapIP4code 5 = ip-address; option SUNW.SswapPTHcode 6 = text; option SUNW.SbootFILcode 7 = text; option SUNW.Stz code 8 = text; option SUNW.SbootRS code 9 = integer 16; option SUNW.SinstIP4code 10 = ip-address; option SUNW.SinstNM code 11 = text; option SUNW.SinstPTHcode 12 = text; option SUNW.SsysidCFcode 13 = text; option SUNW.SjumpsCFcode 14 = text; option SUNW.Sterm code 15 = text; option SUNW.SbootURIcode 16 = text; option SUNW.SHHTPProxy code 17 = text; class vendor-classes { match option vendor-class-identifier; } subclass vendor-classes SUNW.Ultra-5_10 { vendor-option-space SUNW; option SUNW.SbootURI= tftp://10.10.1.131/inetboot.SUN4U.Solaris_11-1 ; option SUNW.SinstIP4 10.10.1.131; option SUNW.SinstNM = heimdall; option SUNW.SinstPTH = /export/solaris11/install; option SUNW.SrootIP4 10.10.1.131; option SUNW.SrootNM = heimdall; option SUNW.SrootPTH = /export/solaris11/install/Solaris_11/Tools/Boot ; } host chimera { hardware ethernet 08:00:20:a2:22:66; option domain-name sdowney.org; option host-name chimera; next-server 10.10.1.131; fixed-address 10.10.1.132; } http://www.sdowney.org/2006/03/solaris-network-install-using-linux.html All the information is around the 'net, but it took me some time to get it all together. Hopefully this will be of use to someone else. -SMD This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] OpenSolaris Crypto Code
What is the current status of the Cypto Code integration with the main sources? Are they still seperate or are they merged and part of nightly builds now? Thanks, -STEVEl __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org