[osol-discuss] Migrating to ZFS from Gentoo Linux
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi everyone, I'm planning a migration of my fileserver from Gentoo Linux with samba to SCXE with ZFS and samba/cifs sharing. Being a linux junkie I don't have any previous experience in Solaris and there's a couple of questions I have about my migration: Using ZFS samba sharing, I have sharing setup but creation permission masks are a lil weird. For example, I browse to my SCXE server and get a login box, ok I setup smb auth in pam.conf. So I login and I'm presented with my shares. Now I create a file, but when viewing the perms from a console, I see that its got no perms at all. I know that samba has a creation mask option so I went looking in smb.conf, only to find that the only smb.conf is smb.conf-example so it's apparently not using it. I have to wonder, as my zfs pool is shared fine using zfs set sharesmb=name=public tank, is smb.conf in use at all? Is it even using samba or is it using CIFS? How can I go about setting up the finer points of the samba/CIFS sharing? Anything that can help a linux/samba geek migrate to solaris would be so helpful to me right now. Thanks in advance - -- Matt Harrison -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (MingW32) iEYEARECAAYFAkg9zB8ACgkQxNZfa+YAUWFnfgCg4HsYgcwcpAagQ0bK4XpQJIbB STUAnAxVzUbynSGCzEaXhv2TOv/XTZtH =Lu5R -END PGP SIGNATURE- ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Migrating to ZFS from Gentoo Linux
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Thommy M. Malmström wrote: | Matt Harrison wrote: | | Hi everyone, I'm planning a migration of my fileserver from Gentoo | Linux with samba to SCXE with ZFS and samba/cifs sharing. Being a linux | junkie I don't have any previous experience in Solaris and there's a | couple of questions I have about my migration: | | Using ZFS samba sharing, I have sharing setup but creation permission | masks are a lil weird. For example, I browse to my SCXE server and get a | login box, ok I setup smb auth in pam.conf. So I login and I'm presented | with my shares. Now I create a file, but when viewing the perms from a | console, I see that its got no perms at all. I know that samba has a | creation mask option so I went looking in smb.conf, only to find that | the only smb.conf is smb.conf-example so it's apparently not using it. | | I have to wonder, as my zfs pool is shared fine using zfs set | sharesmb=name=public tank, is smb.conf in use at all? Is it even using | samba or is it using CIFS? How can I go about setting up the finer | points of the samba/CIFS sharing? | | Anything that can help a linux/samba geek migrate to solaris would be so | helpful to me right now. | | Google is your friend. Here's a starter kit for ZFS/CIFS: | | http://opensolaris.org/os/project/cifs-server/Admin-Guide/p12.html | http://www.genunix.org/wiki/index.php/Getting_Started_With_the_Solaris_CIFS_Service Thanks for those links, I've managed to work out the majority of what I needed, including auto home mounting :) The one thing I can't find any reference to is the creation masks. I can write to the shares with an appropriate user, but after creation there are no permissions assigned. I've googled for this but all I come up with is samba's create mask, nothing to do with CIFS, which I now see I am using. - -- Matt Harrison [EMAIL PROTECTED] http://blog.genestate.com -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (MingW32) iEYEARECAAYFAkg/CfMACgkQxNZfa+YAUWEUogCfecr3lLtq2ziq4f1v1M6KhwLb kRMAoKdBWKysvkYET7ngSAYkuG2CFJVY =MseQ -END PGP SIGNATURE- ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Migrating to ZFS from Gentoo Linux
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Ok I got some help from someone off-list who directed me to ACLs and the ~ ZFS admin docs. I've tried to setup the ACLs as I need but it just doesn't seem to work how I want. Not having experience with ACLs I would be really grateful if someone could help me out with the ACLs. Here my setup: I have a pool (/tank) which contains a ZFS shared through CIFS. I want root to keep ownership of the ZFS but allow groups access as follows: group staff should have total control of ALL files in the pool, including inherited permissions for all newly created files. group users should have read/list/execute permissions to the files, and also any new files that are create by people in the staff group. I know this sounds simple but I've just spent 2 hours playing with ACLs and I haven't come up with anything that works. Thanks in advance - -- Matt Harrison [EMAIL PROTECTED] http://blog.genestate.com -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (MingW32) iEYEARECAAYFAkhBmY0ACgkQxNZfa+YAUWG6UgCgzKhsUA0M1GN82R0JEuUGA0GU GZgAoKzGW8/Wd7qxOvXzbZ9IYr/0qWcF =1xPp -END PGP SIGNATURE- ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Migrating to ZFS from Gentoo Linux
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Mark Shellenbaum wrote: | What ACL have you setup on the root directory of share? What does | ls -dV /tank show | | and what is the permission of files you have created via CIFS? What | does ls -V file show. Apologies this was originally posted to the previous poster and not the list. Getting used to a new MUA Ok I've tried to reduce this as simply as I can to see where my problems lie. If I reset the perms to trivial and apply the following: chmod -R A+group:staff:rwxpAWCo:allow tank Then it works as expected and the user in the staff group can perform actions with the same rights as the owner. However once a directory is created it assumes perms like so: d-+ 2 matthewh staff 2 May 27 07:44 test/subdir user:matthewh:rwxpdDaARWcCos:---:allow ~ group:2147483648:rwxpdDaARWcCos:---:allow Which means that administering the files in the future becomes very difficult. What I want to do is something like this: chmod A+group:staff:rwxpAWCo:fd:allow tank To allow all newly created files to be managed fully by the staff group. ~ This doesn't work and while I can create a directory, I can't set the name, rename it, delete it or anything else. This just doesn't seem to make sense. There must be a simple way to do what I want but ACLs make the job much more difficult than I think it should be. I've been reading through the zfsadmin pdf file and it looks like what I want to do should work, yet it doesn't. Can anyone please supply some examples of how they solved a problem like this? Thanks in advance - -- Matt Harrison [EMAIL PROTECTED] http://blog.genestate.com -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (MingW32) iEYEARECAAYFAkhB7u0ACgkQxNZfa+YAUWEMCgCg1EREytbhkXvGuirLD4lZqhCJ MdgAoIFWZUbgkK8OTSOmXBa4lcHlrngc =VYZe -END PGP SIGNATURE- No virus found in this outgoing message. Checked by AVG. Version: 8.0.100 / Virus Database: 269.24.4/1475 - Release Date: 30/05/2008 14:53 ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Migrating to ZFS from Gentoo Linux
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Mark Shellenbaum wrote: | You are connecting with user matthewh ? Yes I am, and matthewh's primary group is staff, and also in the users group | | Have you tried something like this? | | chmod A=user:matthewh:rwxpdDaARWcCos:fd:allow,group:staff:rwxpdDaARWcCos:fd:allow,group:2147483648:rwxpdDaARWcCos:fd:allow /tank No I hadn't previously but that has led me to experiment with using FULL permissions for the staff group, which seems to have done the trick | This will allow matthewh, group staff, and the system group full access | to every file/directory. | | another thing to experiment with would be something like this. I'm not | sure how the cifs server interprets the abstract owner@, and group@ | entries though. I'm not too fussed how the owner@ works through cifs as all root admin stuff will be done on the console or through ssh. the staff group would be used unless something went very wrong | chmod A=group:staff:rwxpdDaARWcCos:fd:allow,owner@:rwxpdDaARWcCos:fd:allow,group@:rwxpdDaARWcCos:fd:allow /tank | In summary, I believe my problem is fixed by using rwxpdDaARWcCos:fd perms for the staff group. I need to switch to a user who is only in the users group and make sure that read perms are ok and that write perms are blocked, but otherwise I seem to be working fine now. Thanks very much for the help, this is the last obstacle I was facing before my migration to ZFS next week. - -- Matt Harrison [EMAIL PROTECTED] http://blog.genestate.com -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (MingW32) iEYEARECAAYFAkhCBEkACgkQxNZfa+YAUWGL3QCeM/QX6LBVB8nNZDOlsYRrZXD6 idkAnA+TARXpwuMPoXhPNx+n667smess =0uU2 -END PGP SIGNATURE- No virus found in this outgoing message. Checked by AVG. Version: 8.0.100 / Virus Database: 269.24.4/1475 - Release Date: 30/05/2008 14:53 ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] genunix.org Down?
Sean Sprague wrote: The fact that the site is still down is a shame, I saw all the hubub about the site being down, and not knowing what it was i tried it. Its working from here perfectly and has since some of the first posts in this thread. I didn't have any dns cached so it looks like there's something more going on that simple dns changes. Hope someone find this of use -- Matt Harrison [EMAIL PROTECTED] http://blog.genestate.com No virus found in this outgoing message. Checked by AVG. Version: 8.0.100 / Virus Database: 270.2.0/1493 - Release Date: 09/06/2008 17:25 ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Extended Desktop?
On Mon, Jun 16, 2008 at 05:35:23AM -0700, Matt Parnell wrote: Hey Guys, its the Noob again. lol As you all know I'm an XP guy, attapting to OpenSolairs, so far its been great, a couple hickups that you'veset me in the correct directions, but i still am loving it. A Think which I've got used to on XP is the extended Desktop funtion. I am using a laptop, 15.4 1280x800. I have connected an external monitor, 19 1440x900. On XP, all i have to do is go to properties, enable the monitor drag it to where i want it, click extend and I can use both monitors. All i do i move my curse off the screen to the left (i made it to the left) and it's on me other monitor. Does OpenSolaris have an equivelant? When my Laptop is connected to the Monitor (as it always is, exept for when i'm out with it), OpenSolaris detects it and puts the screen resolution to match it: 1440x900. But that also isplays on my laptop LCD the same. The same thing is displayed, but because my laptop is set to 1280x800 I'm loosing a large block off my laptop screen. That doesn't matter aslong as there is an option, or a file that is editable for extended monitors. Can anyone help me out here? I've been browsing the NET for answers for ages, and i can't find anything. What do you think? (Some of you are probably thinking that im wrong to want it ... but it's very helpful to split your stuff up and have appications opening on diffent screens. I know thats what Work Stations are for, but they don't interface with scondary monitors) Hi, I haven't set this up myself, but I believe the /etc/X11/xorg.conf file should allow you to configure two monitors, their resolutions and where you want them in relation to each other. As I say I haven't done multiple monitors myself, but maybe a google search for multiple monitors X11 might turn up some useful information. HTH Matt Harrison pgpCHQn9ZBjGp.pgp Description: PGP signature ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] PANIC: VirtualBox on os2008-05 installing os2008-05
On Wed, Jun 18, 2008 at 12:00:12PM -0700, Glenn Lagasse wrote: Sure, anything is 'possible'. But in reality, none of the virtualization systems I've used (Xen, Vmware, VirtualBox, Parallels) actually support/work being run in a virtualized environment. I've certainly never gotten any sort of thing to work (and to be quite honest, I wouldn't expect it to work or work well). Could they be made to work? Possibly, I suppose. But the facts as they stand today are that they don't. At least in my experience. I'm no expert on most virtualisation s/w but I know with vmware, it has code in place to specifically stop you trying to run vmware inside vmware. This is presumably because the vmware guys know it isn't going to work and it shouldn't be tried. I know this doesn't help the conversation much but it's just my 2c :) Matt pgpBaQoNkvHR4.pgp Description: PGP signature ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] updating packages
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I'm running svn_89 and I've been trying to work out how to run updates. I found the updatemanager app, which just tells me there's nothing to update, so I did some googling. Apparently I can just upgrade from the latest disc, well I downloaded the svn_91 disc and ran that, but there's no option to do updates. I'm just wondering what the standard procedure for keeping my system up to date it as I've so far been unable to find this info myself. Thanks - -- Matt Harrison [EMAIL PROTECTED] http://mattharrison.org -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (MingW32) iEYEARECAAYFAkhjKygACgkQxNZfa+YAUWHPwwCg15AjinvssDJ/dAAXV7p46ELv CM0AoLCA5B6z4iisTprNGOf/h1FSOA2x =elhw -END PGP SIGNATURE- ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] updating packages
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Kristian Rink wrote: | Matt; | | I'm just wondering what the standard procedure for keeping my system up | to date it as I've so far been unable to find this info myself. | | Did you try | | pkg refresh --full | pkg image-update -v | | to get that job done? Or isn't that what you want/need? Thanks for the reply I have tried that, unfortunately the pkg command isn't on this system and it was installed with the most complete package selection. This is the problem I've had, everything I can find on google or the list archives refers me back to the pkg command, which I can't find. - -- Matt Harrison [EMAIL PROTECTED] http://mattharrison.org -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (MingW32) iEYEARECAAYFAkhjSVwACgkQxNZfa+YAUWHSbACfelpOK2wV0Y3/GyzbuUMTj0gI 4m4An3hrPUvnZPsNCdFSGvXs87Y6IQYE =FjKa -END PGP SIGNATURE- ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] updating packages
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Kristian Rink wrote: | Matt Harrison schrieb: | I have tried that, unfortunately the pkg command isn't on this system | and it was installed with the most complete package selection. | | This is the problem I've had, everything I can find on google or the | list archives refers me back to the pkg command, which I can't find. | | H, what kind of installation medium did you initially use to install | OpenSolaris from? IIRC the IPS facility (and thus the pkg command) | was/is just included in more recent OpenSolaris builds, especially the | Indiana TPs and OS 2008.05... Maybe your installation just was too old | to already have this tools included? This machine was installed from an svn_89 dvd image that I downloaded from the opensolaris.org site. I've got an svn_91 image here as well, maybe its included on that? - -- Matt Harrison [EMAIL PROTECTED] http://mattharrison.org -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (MingW32) iEYEARECAAYFAkhjTVwACgkQxNZfa+YAUWHBqQCeMb1enf6GYrb0RJCKpXmZ1CHi ccQAn0nmPFPBVj+ylifBN2/00GzNHrns =UQvc -END PGP SIGNATURE- ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] updating packages
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hugh McIntyre wrote: | Matt Harrison wrote: | | Kristian Rink wrote: | | H, what kind of installation medium did you initially use to | install | | OpenSolaris from? IIRC the IPS facility (and thus the pkg command) | | was/is just included in more recent OpenSolaris builds, especially the | | Indiana TPs and OS 2008.05... Maybe your installation just was too | old | | to already have this tools included? | | This machine was installed from an svn_89 dvd image that I downloaded | from the opensolaris.org site. I've got an svn_91 image here as well, | maybe its included on that? | | The pkg commands are only on OpenSolaris, e.g, the 2008.05 release. | | snv_89 or _91 is the classic Solaris version instead, so no pkg | commands. There's also no updatemanager, since this is for patches on | a released OS such as Solaris 10, not Nevada. Instead, for snv_91, | you're *supposed* to be able to either: | | - boot from a DVD of the snv_91 image and select an upgrade install. | This may fail though if it thinks your system is not upgradable for | whatever reason (which it may not explain). Such as low disk space, | corrupted package DB, for example. | | - or you can use live upgrade while the old OS is running. | | So I can't explain why the upgrade fails. Maybe ask on the | install-discuss mailing list, or opensolaris-help? But the pkg | command is definitely not going to work to upgrade snv_*. | | Hugh. Hehe, I seem to have got a bit confused between OpenSolaris the distribution and the SCXE which I'm running (just checked what I actually downloaded). I'm installing a new fileserver with CIFS/ZFS, is OpenSolaris (the distro) going to work just as well or better than Nevada? I'm only using a Nevada build as that is what was recommended on a ZFS blog, who's instructions I was trying to follow. Incidentally, I've managed to get the update option to work from snv_91 image, so it seems I do have an upgrade path of sorts should I decide to stick with Nevada. and for my previous posts s/svn/snv/g. Just out of interest what does snv stand for, sun version? I always thought it was the svn revision number. Thanks - -- Matt Harrison [EMAIL PROTECTED] http://mattharrison.org -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (MingW32) iEYEARECAAYFAkhjdZcACgkQxNZfa+YAUWEggACbBzq7YIDO2czgnShNSZH9Cbzp YckAoJHXuqxOyv0XUTcQ9++bL/ustzM+ =0LKW -END PGP SIGNATURE- ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] sxce94 : registration wizard still not working
john kroll wrote: I still like to register the software even though you guys don't really have organized support or updates ; Hoping the registration might collect something useful. This install is on dell e521 running bfe third party driver. Wizard claims no internet connection. I am posting this with the install so it has connection?? Had exactly the same problem since snv89, network was working perfectly with the exception of the registration wizard. It wasn't a big deal so I didn't persue it, although I'll be interested in hearing if there's a fix. Matt No virus found in this outgoing message. Checked by AVG - http://www.avg.com Version: 8.0.138 / Virus Database: 270.5.6/1578 - Release Date: 28/07/2008 17:13 ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] upgrading from 91 to 95
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Ron Halstead wrote: | I assume the zpool is on a second disk(s). If so, zpool export pool, do the installation of _95 and zpool import pool. Your data will be safe even if you don't export the pool, but, I'm paranoid. I used to do it that way before I decided to try Live Upgrade. | | --ron Excellent, nice to hear from someone who has used this method before. I'm going to test it all out on a virtual machine first so I know exactly what I'm playing with. I've made backups of all my system configs, so I think I'll try it over the weekend if all goes well. thanks Matt -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (MingW32) iEYEARECAAYFAkihzxYACgkQiMZP8RXkRgTT6ACffP2fUbKPk6wrbR952QrwPSVs UhgAn16ldcjH7GFvcLhNaxLmpH3cAg9j =6bmU -END PGP SIGNATURE- ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] upgrading from 91 to 95
Ben Rockwood wrote: Zpools are self contained, so as stated before its safe. In upgrades like this I often will actually pull or disconnect the zpool disks just for paranoia sake. As for the config of the existing machine, keeping a copy of /etc is never a bad idea. I typically will rsync a copy of it out somewhere, either to another server or in a case like this to the zpool... that way you have a fallback copy of the system config just in case. Upgrades get even easier if you put filesystems like /opt or /usr/local on the zpool. Post-install just mount /opt and /usr/local from the zpool (zfs set mountpoint=/usr/local pool/local zfs mount -a). I frequently upgrade my Nevada boxes (read: reinstall) and ZFS makes like a lot simpler Yes I'd done some test upgrades on systems with ZFS and it seems to go like a dream compared to UFS and messing about with slices etc. I will go ahead on friday and do the re-install of our fileserver, I did a test on VMware and it worked like a charm, my test pool was re-imported perfectly. I think I will follow your advice and pull the pool disks until the OS is re-installed, I guess you can never be too careful :) Once the system is on a ZFS root I will be confident to perform future upgrades with the live upgrade tools as I've tried that both on UFS and ZFS and the latter is a hundred times easier. Thanks for everyone's advice Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] missing lots of headers
I did a re-install of an SCXE machine to get it onto a zfs root and I've run into a problem. This time round I did an End-User install instead of Entire Distribution and it has left out installing a lot of things. The box is used to run mongrel server for Ruby on Rails websites. I have so far installed ruby, gmake, binutils and SUNWarc (whatever that is). The problem is now that I try to compile some Ruby gems, I get errors like this: /usr/ruby/1.8/lib/ruby/1.8/i386-solaris2.11/ruby.h:40:21: stdlib.h: No such file or directory /usr/ruby/1.8/lib/ruby/1.8/i386-solaris2.11/ruby.h:44:21: string.h: No such file or directory /usr/ruby/1.8/lib/ruby/1.8/i386-solaris2.11/ruby.h:53:20: stddef.h: No such file or directory /usr/ruby/1.8/lib/ruby/1.8/i386-solaris2.11/ruby.h:54:19: stdio.h: No such file or directory /usr/ruby/1.8/lib/ruby/1.8/i386-solaris2.11/ruby.h:71:20: alloca.h: No such file or directory /usr/ruby/1.8/lib/ruby/1.8/i386-solaris2.11/ruby.h:91:21: limits.h: No such file or directory In file included from /usr/ruby/1.8/lib/ruby/1.8/i386-solaris2.11/ruby.h:718, from fastthread.c:12: /usr/ruby/1.8/lib/ruby/1.8/i386-solaris2.11/missing.h:16:24: sys/time.h: No such file or directory /usr/ruby/1.8/lib/ruby/1.8/i386-solaris2.11/missing.h:25:25: sys/types.h: No such file or directory In file included from /usr/ruby/1.8/lib/ruby/1.8/i386-solaris2.11/ruby.h:719, from fastthread.c:12: /usr/ruby/1.8/lib/ruby/1.8/i386-solaris2.11/intern.h:219: error: syntax error before fd_set /usr/ruby/1.8/lib/ruby/1.8/i386-solaris2.11/intern.h:220: warning: struct timeval declared inside parameter list /usr/ruby/1.8/lib/ruby/1.8/i386-solaris2.11/intern.h:220: warning: its scope is only this definition or declaration, which is probably not what you want Obviously I am missing the package(s) which provide the headers needed by this and many other compilations. Can anyone tell me which packages I should install in order to get these headers onto my system? Many thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] missing lots of headers
Andras Barna wrote: probably SUNWhea excellent, that solved it, Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] bash line wrapping
I connect to my SXCE box regularly over ssh and I've got a bash line wrapping problem that is driving me mad. I googled about this for 2 weeks and haven't found any solution. If I type a command that exceeds 65 characters (including the shell prompt), it causes the line to wrap to the beginning of the same line, and if I go back to change the command, it litters up the entire terminal. Apart from making a mess it makes it really hard to work in the console. This problem manifests when connecting over ssh from either a linux machine with Konsole, or from Windows with PuTTY. I have tried various things, including changing terminal types, and using the shopt internal. In linux this problem occurred now and then but merely setting my term type to linux always sorted it out. Does anyone have any information relating to this, if I can find a solution it would make my life so much easier. On a side note, i have \w in my PS1 prompt to display the CWD. This is supposed to abbreviate the CWD with a tilde if working inside the home directory. Unfortunately, if working as any user other than root, the path is always displayed in full, as if outside the home directory. This makes the prompt very large and adds to the 65 character limit problem. Any ideas? Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] bash line wrapping
Shawn Walker wrote: Matt Harrison wrote: I connect to my SXCE box regularly over ssh and I've got a bash line wrapping problem that is driving me mad. I googled about this for 2 weeks and haven't found any solution. If I type a command that exceeds 65 characters (including the shell prompt), it causes the line to wrap to the beginning of the same line, and if I go back to change the command, it litters up the entire terminal. Apart from making a mess it makes it really hard to work in the console. This problem manifests when connecting over ssh from either a linux machine with Konsole, or from Windows with PuTTY. I have tried various things, including changing terminal types, and using the shopt internal. In linux this problem occurred now and then but merely setting my term type to linux always sorted it out. Does anyone have any information relating to this, if I can find a solution it would make my life so much easier. On a side note, i have \w in my PS1 prompt to display the CWD. This is supposed to abbreviate the CWD with a tilde if working inside the home directory. Unfortunately, if working as any user other than root, the path is always displayed in full, as if outside the home directory. This makes the prompt very large and adds to the 65 character limit problem. Any ideas? I've seen this behaviour as well -- it's annoying to no end. Especially since if you use the backspace key to backup beyond the second line of text, the display of the command line is corrupted and you can't actually see what was typed before. Exactly, if you do backspace too far and loose what you were typing, you end up counting the characters you typed in your head to get to the right point :P I'm glad I'm not the only one having this problem, hopefully someone will have an idea for a workaround and we can get back to typing commands without a backslash every 20 chars :) Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] bash line wrapping
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Frank Penczek wrote: Hi, On Sat, Aug 16, 2008 at 1:45 AM, Matt Harrison [EMAIL PROTECTED] wrote: I connect to my SXCE box regularly over ssh and I've got a bash line wrapping problem that is driving me mad. I googled about this for 2 weeks and haven't found any solution. If I type a command that exceeds 65 characters (including the shell prompt), it causes the line to wrap to the beginning of the same line, and if I go back to change the command, it litters up the entire terminal. Apart from making a mess it makes it really hard to work in the console. This problem manifests when connecting over ssh from either a linux machine with Konsole, or from Windows with PuTTY. I have tried various things, including changing terminal types, and using the shopt internal. In linux this problem occurred now and then but merely setting my term type to linux always sorted it out. Does anyone have any information relating to this, if I can find a solution it would make my life so much easier. On a side note, i have \w in my PS1 prompt to display the CWD. This is supposed to abbreviate the CWD with a tilde if working inside the home directory. Unfortunately, if working as any user other than root, the path is always displayed in full, as if outside the home directory. This makes the prompt very large and adds to the 65 character limit problem. Any ideas? Thanks Matt I've had a very similar problem. It turned out that it was related to escape sequences I've used for colours, more specifically, that bash didn't recognised these as non-printable characters (so I read). Embedding all sequences in \[ and \] solved it for me. For example, if your prompt command is \e[0;37m\W\e[0;34$ change it to \[\e[0;37m\]\W\[\e[0;34\]$ and see if it works for you as well. Cheers, Frank Excellent, I was using a lot of colour escapes in my profile and wrapping them with \[ and \] has solved the wrapping problem. Now I just have to find out why \w isn't being collapsed to a tilde when possible :P Thanks Matt -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (MingW32) iEYEARECAAYFAkipp/AACgkQiMZP8RXkRgRduQCgjFtNXtS4Cf56QndNyfVMwzkf 8BoAmwR50Di++8xZPor/VaNEt2X8fC5h =UAWq -END PGP SIGNATURE- ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] which package provides math.h?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I'm trying to get coloured syntax highlighting working in vim, and after a lot of work it looks like i'm going to have to compile my own vim, using ncurses. Ncurses requires math.h but I can't find which package provides it. A pointer to the package or some hints on syntax highlighting with colours would be great :) Thanks Matt -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (MingW32) iEYEARECAAYFAkisolAACgkQiMZP8RXkRgQi8QCgkW8A0LN1ZiJwhnC5K+XtX9rm MR4An2AGnj5XoayiQ7DD78Dh+qQB8gXz =u4Iw -END PGP SIGNATURE- ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] which package provides math.h?
I probably should have mentioned, I'm running SXCE (snv_95). I've tried all the TERM=xtermc, TERM=xterm-color stuff, and I'm still in black and white. Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] which package provides math.h?
Mike Gerdts wrote: On Wed, Aug 20, 2008 at 6:24 PM, Matt Harrison [EMAIL PROTECTED] wrote: I probably should have mentioned, I'm running SXCE (snv_95). I've tried all the TERM=xtermc, TERM=xterm-color stuff, and I'm still in black and white. I have the following in my .vimrc: :if has(terminfo) : set t_Co=8 : set t_Sf=ESC[3%p1%dm : set t_Sb=ESC[4%p1%dm : syntax on :else : set t_Co=8 : set t_Sf=ESC[3%dm : set t_Sb=ESC[4%dm : syntax on :endif Replace ESC with an escape character (^Vesc). Excellent :D I had found something through google that said to do this, but using different escape codes. It didn't work, but this does :) I now have coloured syntax hilighting over ssh and at the console :) many thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] which package provides math.h?
Glenn Lagasse wrote: * Matt Harrison ([EMAIL PROTECTED]) wrote: I probably should have mentioned, I'm running SXCE (snv_95). I've tried all the TERM=xtermc, TERM=xterm-color stuff, and I'm still in black and white. Works just fine for me on SXCE 92 (and thus it should work just fine on 95). I'm setting TERM=xterm-color and have the following in my .vimrc: syntax on set background=dark I log in to this box over SSH from OS X using Terminal.App. Vim on SXCE and OpenSolaris comes with syntax color support OOTB. Something else in your environment/setup is amiss (the question is, what). What terminal emulator are you using (xterm, gnome-terminal, etc)? Cheers, I was logging in from PuTTy or gnome-terminal on a linux box, my termtype was set to xterm, but I had tried it with xtermc and xterm-color. I had the same things in my .vimrc but until I tried Mike's suggestions, I was all in black and white. However Mike's .vimrc has fixed it nicely :) Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Bittorrent client for Solaris
Gizd wrote: I wanted to install libtorrent and I believe I did install it, but since it's the first time ever I'm installing a package I'm not entirely sure I did everything I had to. I downloaded libtorrent-0.11.4-sol10-x86-local.gz from SunFreeware, then what I did was: gunzip libtorrent-0.11.4-sol10-x86-local.gz pkgadd -d libtorrent-0.11.4-sol10-x86-local Got a message at the end that it was installed successfully, but now what? I know how it stupid sounds, but I can't figure out how to run the application. I believe that libtorrent is only the libraries and won't include actual applications for using torrents. I could be wrong but you'll probably have to install a torrent application and I have no idea what's available for solaris. HTH Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] dropping ethernet connection
Hi all, My OSOL fileserver has started dropping the ethernet connection every now and then. It's very annoying and I can't see what's going on. All I know is that network is disrupted and I get messages like this sometimes repeated continuously: Sep 10 19:03:01 exodus mac: [ID 486395 kern.info] NOTICE: nge0 link down Sep 10 19:03:06 exodus mac: [ID 435574 kern.info] NOTICE: nge0 link up, 100 Mbps, full duplex I've changed the cable and that helped, but then it started happening again, so I changed the cable again and it doesn't work at all. I've moved everything around that I can with the exception of changing the nic, as I don't know how to configure the other onboard nic and use that. Can anyone give me some ideas what to look for? fmadm faulty doesn't show anything, and network connectivity is perfect when it hasn't gone down. What could be happening? Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] dropping ethernet connection
James Carlson wrote: Matt Harrison writes: My OSOL fileserver has started dropping the ethernet connection every now and then. It's very annoying and I can't see what's going on. All I know is that network is disrupted and I get messages like this sometimes repeated continuously: Sep 10 19:03:01 exodus mac: [ID 486395 kern.info] NOTICE: nge0 link down Sep 10 19:03:06 exodus mac: [ID 435574 kern.info] NOTICE: nge0 link up, 100 Mbps, full duplex What version (nightly build) of code do you have? I used to see exactly this problem a long time ago -- it was bad enough that I plugged in an old 3c509 card to deal with it -- but it hasn't happened in quite a while. What could be happening? Most likely driver problems. Its running SXCE snv_95 and has been running since 95 was released with no problems whatsoever. It's just one of those problems that (for me) is really hard to track down but very annoying. I'll see if i can come up with any more info Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] dropping ethernet connection
It appears that for whatever reason, the interface refuses to enter the RUNNING state. No matter what I do I can't get nge0 RUNNING and that means I'm without network :/ This is really a big problem for me right now, any help would be greatly appreciated. I'm going to try putting in an old RTL8139 and see if I can get running on that for now. Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] dropping ethernet connection
James Carlson wrote: Matt Harrison writes: Its running SXCE snv_95 and has been running since 95 was released with no problems whatsoever. OK. All the problems I had predated build 95. It's just one of those problems that (for me) is really hard to track down but very annoying. I'll see if i can come up with any more info Make sure you file a bug on it. After messing around for a while with ifconfig I just tried: svcadm disable network/physical:default and svcadm enable network/physical:default and I seem to be working again, although for how long I have no Idea. Where would be the best place to report this so it can be looked at by the experts? Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] dropping ethernet connection
After messing around for a while with ifconfig I just tried: svcadm disable network/physical:default and svcadm enable network/physical:default and I seem to be working again, although for how long I have no Idea. Corrrection, I got a link light, but I'm not workingI can't find how to plumb this other card so I'm totally at a loss. I might have to do a re-install and see what I can do. Only problem is my install images were on that server so I'll have to download the dvd again on my 0.5Mb dsl...oh joy ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] dropping ethernet connection
James Carlson wrote: At a guess, doing this would have fixed it as well: # config=`ifconfig nge0 configinfo` # ifconfig nge0 unplumb # ifconfig $config That's a good clue, though I don't know enough about the driver to know why that would work. Yes I could have done that, i forgot about unplumb. Actually it didn't come back after i restarted the network service, but it did come back about 20 mins later whle i was fiddling again. While my connection is up I'm noticing some issues, video playback from the server is laggy, downloads on the server stall and then restart. bugs.opensolaris.org I will file a bug now, and when I've got some spare mins I'll upgrade to 96, just in case that has some fixes. Thanks so much for the replies, Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] dropping ethernet connection
James Carlson wrote: Matt Harrison writes: It appears that for whatever reason, the interface refuses to enter the RUNNING state. No matter what I do I can't get nge0 RUNNING and that means I'm without network :/ That means that the underlying driver is reporting link down. It's consistent with the log messages you're seeing. This is really a big problem for me right now, any help would be greatly appreciated. I'm going to try putting in an old RTL8139 and see if I can get running on that for now. Using some replacement interface is exactly what I was going to recommend -- along with filing a bug. (I would have recommended a trusty old Intel PPRO [e1000g] or 3COM [elxl] card, but that RTL8139 [rtls] should work as well.) I've put in another card and that's working fine, but how can I configure that card to be the default after reboot? I can manually use ifconfig to reconfigure but that is annoying as the faulty onboard NIC hangs when I try to unplumb it etc. If I can reconfigure sxce to use the other nic i should be alright. Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] autoconf and packages
Dennis Clarke wrote: I'm trying to install Ruby Enterprise Edition and it's failing because autoconf isn't installed. I've got automake installed ok but I can't find a package for autoconf anywhere unless it has a really cryptic name like most of the other packages. I've also tried to compile using Sun Studio but found there is no package for it on my CD. What are my options? does this help ? http://blastwave.network.com/testing/autoconf-2.62,REV=2008.09.09-SunOS5.8-i386-CSW.pkg.gz or just change that to sparc .. if needed Dennis Thanks Dennis, That certainly would help, but how do I go about installing it? Currently I copy needed packages from the CD to /var/spool/pkg and then run pkgadd. I don't have the simple pkg command as I'm running SXCE and not OpenSolaris the distro. Is this going to be a problem to me? ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] autoconf and packages
Alan Coopersmith wrote: Matt Harrison wrote: I'm trying to install Ruby Enterprise Edition and it's failing because autoconf isn't installed. I've got automake installed ok but I can't find a package for autoconf anywhere unless it has a really cryptic name like most of the other packages. BTW I'm running SXCE snv_97. Then you should have autoconf installed in /usr/bin/autoconf by the package SUNWaconf. It's been in SXCE for a few builds now - I see it on an snv_92 machine. Thats excellent, I've got automake installed, but now when I try to compile the package I get: autom4te: need GNU m4 1.4 or later: /usr/sfw/bin/gm4 /usr/sfw/bin/gm4 doesn't exist, however /usr/bin/m4 and /usr/xpg4/bin/m4 exist, although neither of them respond when I try to get their version. Every option I try is unrecognised. Any help appreciated Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] autoconf and packages
Alan Coopersmith wrote: Matt Harrison wrote: I'm trying to install Ruby Enterprise Edition and it's failing because autoconf isn't installed. I've got automake installed ok but I can't find a package for autoconf anywhere unless it has a really cryptic name like most of the other packages. BTW I'm running SXCE snv_97. Then you should have autoconf installed in /usr/bin/autoconf by the package SUNWaconf. It's been in SXCE for a few builds now - I see it on an snv_92 machine. Nevermind, I found the SUNWgm4 package and it all seems to be working now :) Thanks for the input Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Burn a bootable DVD
Leví Teodoro da Silva wrote: Hi guys I use OpenSolaris 2008.05 at my workstation and i will install Solaris 10 on my server. I am downloading the ISO for solaris 10. =) My question is simple : How can i do to create a bootable dvd with this ISO using OpenSolaris !? I do not know a program for it Thanks, Leví -- This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org If the image you're downloading is bootable then just burn it to a disc. I use the cdrw tool: cdrw -d your cdrom device, eg. cdrom0 -i /path/to/your/image This works for me when I burn the latest SXCE images. HTH Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] sudden huge slowdown
Everything was working fine on our snv_97 fileserver then one day our transfers had slowed to a crawl. I originally thought it was CIFS only but after some tests I can see that its all network transfers including scp and ftp. I've run some benchmarks like bonnie++ and disk IO seems to be ok. A zpool scrub doesn't bring any errors and takes around the same time as normal. The problem is that copying to the machine will run at about 20Kb/s and from the machine is faster, somewhere in the region of 60-100Kb/s. What on earth could be causing this massive slowdown? I've rebooted the machine, checked all the processes, checked for faulty devices, everything I can think of. Any ideas? Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] sudden huge slowdown
Bryan Allen wrote: How full is your pool? Service starts crawling as ZFS has to spin the platters to search for free blocks in disparate parts of the disks. (Perhaps someone else can explain that better, but that's my understanding as to why full pool = slow pool.) I try to keep my pools at 90% utilization. You'll see similar threads on [EMAIL PROTECTED] My pool is currently not even at 25% capacity, and has never been over 40% since it was installed. Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] sudden huge slowdown
Robert Milkowski wrote: Hello Matt, Tuesday, September 30, 2008, 10:52:57 PM, you wrote: MH Everything was working fine on our snv_97 fileserver then one day our MH transfers had slowed to a crawl. I originally thought it was CIFS only MH but after some tests I can see that its all network transfers including MH scp and ftp. MH I've run some benchmarks like bonnie++ and disk IO seems to be ok. MH A zpool scrub doesn't bring any errors and takes around the same time as MH normal. MH The problem is that copying to the machine will run at about 20Kb/s and MH from the machine is faster, somewhere in the region of 60-100Kb/s. MH What on earth could be causing this massive slowdown? I've rebooted the MH machine, checked all the processes, checked for faulty devices, MH everything I can think of. MH Any ideas? Network issues? Actually this could be related. About a week ago, the onboard NICs started playing up and would drop connection about 70% of the time. The only solution after talking with the list was to put in a PCI NIC, and it's possible that it might have some issues (it's really quite old). Could packet fragmentation on the NIC be causing this but not bringing down the whole connection? Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] sudden huge slowdown
Matt Harrison wrote: Robert Milkowski wrote: Hello Matt, Tuesday, September 30, 2008, 10:52:57 PM, you wrote: MH Everything was working fine on our snv_97 fileserver then one day our MH transfers had slowed to a crawl. I originally thought it was CIFS only MH but after some tests I can see that its all network transfers including MH scp and ftp. MH I've run some benchmarks like bonnie++ and disk IO seems to be ok. MH A zpool scrub doesn't bring any errors and takes around the same time as MH normal. MH The problem is that copying to the machine will run at about 20Kb/s and MH from the machine is faster, somewhere in the region of 60-100Kb/s. MH What on earth could be causing this massive slowdown? I've rebooted the MH machine, checked all the processes, checked for faulty devices, MH everything I can think of. MH Any ideas? Network issues? Actually this could be related. About a week ago, the onboard NICs started playing up and would drop connection about 70% of the time. The only solution after talking with the list was to put in a PCI NIC, and it's possible that it might have some issues (it's really quite old). Could packet fragmentation on the NIC be causing this but not bringing down the whole connection? Just replaced the old NIC with a nice shiney new RTL8139 and it works beautifully, speed to and from is now back to what it should be. I don't know why I didn't associate the slowdown with the NIC change in the first place, it should've been the first thing I looked at. Thanks for the suggestions. Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] output from fmadm faulty
Hi all, As I want to know when one of the disks in our fileserver is faulty, I have it run fmadm faulty through cron so I can get informed when something goes down (ie a disk). So far it hasn't produced any output until today when I get this: --- -- - TIMEEVENT-ID MSG-ID SEVERITY --- -- - Nov 03 11:40:16 9b069e20-9ea9-6169-ff93-ede7701ef800 SUNOS-8000-1L Minor Fault class : defect.sunos.eft.undiagnosable_problem Description : The EFT Diagnosis Engine encountered telemetry for which it is unable to produce a diagnosis. Refer to http://sun.com/msg/SUNOS-8000-1L for more information. Response: Error reports from the component will be logged for examination by Sun. Impact : Automated diagnosis and response for these events will not occur. Action : Run pkgchk -n SUNWfmd to ensure that fault management software is installed properly. Contact Sun for support. It obviously doesn't provide (me) with that much information, so does anyone know how or if I can trace this and replace hardware if necessary before the pool goes down? Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] PgSQL support in PHP
On Wed, Nov 12, 2008 at 12:44:47AM +, Matt Harrison wrote: Hi all, I'm running an snv_97 box here and trying to get a couple of apps working with PHP and postgres. From the compile options in phpinfo() it appears it was compiled with pgsql support but for some reason no matter what I do, I don't get any signs that PHP is using any pgsql stuff. It seems strange that support for a bundled app wouldn't be included so I must be missing something. Of course it helps if you install the pgsql packages for php. Thanks anyway ;) Matt pgpK7iP0hbwxD.pgp Description: PGP signature ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] CUPS errors
Hi all, I'm trying to get a printer to work for my network through CUPS. I've got the printer installed and looking ok, but when I try to print a test page from the CUPS interface, I get Unsupported format 'application/postscript'! Printing even a simple plaintext file results in Unsupported format 'text/plain'! I've done some googling and lots of people are having the same problem, but on linux distros like debian, ubuntu or RH. I can't see anything that would apply to opensolaris. I've tried fiddling with the mime.convs and the pstoraster filter but I can't get it to respond. I'm running on SXCE (build 97 currently but planning an upgrade to 101 some time soon). Also, once the printer is working locally, how can I set it up on the client machines? Since I'm using the in-kernel CIFS (in domain mode), I realise that the printer won't be accessible as it would with samba. I think I need to use IPP but no idea how :P If it would be possible to publish the printer, AND have it protected by the same domain auth that would be amazing. Any ideas on either part of the problem welcomed. Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] CUPS errors
Matt Harrison wrote: Also, once the printer is working locally, how can I set it up on the client machines? Since I'm using the in-kernel CIFS (in domain mode), I realise that the printer won't be accessible as it would with samba. I think I need to use IPP but no idea how :P I have managed to get the client connected to the printer via http://server:631/myprinter, but of course nothing is printing, due to the issue mentioned previously I assume. Any help on the format errors would be gratefully appreciated as it's my last obstacle. Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] CUPS errors
Norm Jacobs wrote: Perhaps you can provide a little information about your printer and how you have things configured. What kind of printer is it (HP Photospart C7280, Epson CX8300, ...)? How are you configured to communicate with it (USB, socket, ipp, lpd, ...)? How did you create your queue? Which PPD file are you using? Do you have GhostScript, hpijs, gimpprint, foomatic-rip, ... installed? Did you up the log level in cups and take a look at it ? It is an Epson DX4400 all-in-one, although I'm only interested in the printing right now. It's connected via USB. I used the CUPS interface and Find new printers. It found the printer connected via USB, however couldn't recommend an installed PPD. I did some googling and found the open printing database (linuxprinting.org?) which recommended the printer be used with gutenprint-5.2. I've installed gutenprint-5.2 from source and symlinked the ppds into place. I then had the option to use the PPD for the exact model of the printer. According to google, the same problem (with other printers) is down to missing types in mime.convs and/or mime.types. I've checked and the required types are there. As for the logs, I did ramp up to debug level, however the only relevant message was something like: Print-Job client-error-document-format-not-supported: Unsupported format 'application/postscript'! Replace the format above with any of your choice, the only format it will accept is a straight image (jpeg, png etc). The preceeding messages were the normal queuing and connection messages I see with any print job on other servers. Incidentally, I have tried the printer on a gentoo linux box, with the same cups config, and the same PPDs, and it works perfectly. Also the mime types appear to be identical on both. As for the access from a Windows box, since you are using the in-kernel CIFS support, you can't use SMB based printing at this time. You need to use IPP. On the Windows box, the printer uri for your queue will be something like http://hostname:631/printers/queue;. Yes I found this through trial and error, it even seems to use the windows authentication automatically :) Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] migrating issues from SXCE to 2008.11
m...@bruningsystems.com wrote: Hi Matt, Matt Harrison wrote: Hi all, I've got a filer using SXCE installed and I'm thinking about migrating to OpenSolaris due mainly to the improved pkg command. I've tried testing it on VMware (6.0.0 build-45731) but there is an immediate problem. The install went fine but on reboot it appears to fail to start X, dumping me back to a console login with some garbage characters printed over it. It tries to start X 3 times I think, each time coming back to the messed up prompt. I tried to login on the console to check the X logs but it just spams the screen with more garbage. Attached is a screenshot of the console after a login attempt. I have installed the previous release of OSOL before on the same version of VMware, and that didn't have this problem. Any ideas welcomed before I abandon it for SXCE again :) I see this a few times right now (I am developing an Xinput module). If you can telnet/ssh in, take a look at /var/log/Xorg.0.log. You should also try killing Xorg. I suspect it is running based on the screen snapshot you have attached. max Thanks for the reply Max, I've managed to ssh in and grepped /var/log/Xorg.0.log for EE: The two lines that show up are (EE) Unable to locate/open config file and (EE) AIGLX: Screen 0 is not DRI capable So presumably I need to fiddle a bit with the X config, unfortunately thats a job I always hate and haven't done in ages :P One thing I liked about SXCE was the way X seems to work better than on my linux boxes ;) I'll have a fiddle and see what I can come up with. Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] migrating issues from SXCE to 2008.11
m...@bruningsystems.com wrote: Hi Matt, Matt Harrison wrote: I'll have a fiddle and see what I can come up with. You might also want to look at ~/.xsession-errors max Well I couldn't find that file, presumably because I hadn't got to login yet. But, I ran xorgconfig as there is no default config and put in some sensible values for vmware, and now X is running :) So it seems the autodetection for this release isn't doing so well for vmware as it has before. Thanks for the input Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] diagnosing server hang
Hi all, We've got an SXCE snv_97 filer that has been working ok for a few months now. Recently, it has started falling over when large amounts of data are being copied from it. There is nothing printed to the console and I can't find anything related in the logs. The machine doesn't respond via the network or the console. Copying data to the machine is fine, but from (usually) results in a hang. There is a fault reported by fmadm faulty which has been asked about on fault-discuss but they couldn't tell me what it was related to. Attached is the fault report. After a reboot (ouch), the data is apparently fine and scrubbing shows no problems, the machine continues to work perfectly until the next hang. I have no idea how to continue investigating this, so any help would be much appreciated. Thanks in advance Matt Harrison Nov 03 11:40:16 9b069e20-9ea9-6169-ff93-ede7701ef800 SUNOS-8000-1L Minor Fault class : defect.sunos.eft.undiagnosable_problem Description : The EFT Diagnosis Engine encountered telemetry for which it is unable to produce a diagnosis. Refer to http://sun.com/msg/SUNOS-8000-1L for more information. Response: Error reports from the component will be logged for examination by Sun. Impact : Automated diagnosis and response for these events will not occur. Action : Run pkgchk -n SUNWfmd to ensure that fault management software is installed properly. Contact Sun for support. ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] diagnosing server hang
Ian Collins wrote: Matt Harrison wrote: Hi all, We've got an SXCE snv_97 filer that has been working ok for a few months now. Recently, it has started falling over when large amounts of data are being copied from it. There is nothing printed to the console and I can't find anything related in the logs. The machine doesn't respond via the network or the console. I'd guess a hardware fault. Swap out the NIC and run a thorough memory check (memtes86 on an x86 system). Thanks for the reply, the NIC is a brand new intel pro 1000 server card, which was replaced after talking on the zfs-discuss list (we were having performance issues over cifs). The memory certainly could be a problem. Seeing as this is a 64bit AMD x2 system, should I be looking for a 64bit version of memtest? Thanks -- Matt Harrison ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] diagnosing server hang
Ian Collins wrote: Matt Harrison wrote: Ian Collins wrote: Matt Harrison wrote: Hi all, We've got an SXCE snv_97 filer that has been working ok for a few months now. Recently, it has started falling over when large amounts of data are being copied from it. There is nothing printed to the console and I can't find anything related in the logs. The machine doesn't respond via the network or the console. I'd guess a hardware fault. Swap out the NIC and run a thorough memory check (memtes86 on an x86 system). The memory certainly could be a problem. Seeing as this is a 64bit AMD x2 system, should I be looking for a 64bit version of memtest? Just grab a copy of the ultimate boot CD, every home should have one! thanks Ian, I'll look into this tomorrow. -- Matt Harrison ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] diagnosing server hang
Matt Harrison wrote: thanks Ian, I'll look into this tomorrow. Well I'm not sure if it's good news or not. I've got the machine running memtest86+ with the standard tests and so far it's done 2 passes (3 hours runtime) without a single error. I'm going to leave it running overnight but does it seem there could be another problem other than memory? Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] diagnosing server hang
m...@bruningsystems.com wrote: Hi Matt, Matt Harrison wrote: Matt Harrison wrote: thanks Ian, I'll look into this tomorrow. Well I'm not sure if it's good news or not. I've got the machine running memtest86+ with the standard tests and so far it's done 2 passes (3 hours runtime) without a single error. I'm going to leave it running overnight but does it seem there could be another problem other than memory? Have you gotten anywhere yet with this hang? Have you tried set snooping=1 in /etc/system? How about booting with kmdb and forcing a dump? I'm not sure why this is necessarily hardware related... max Unfortunately not yet...I was forced to bring the server back up to get some files from it, and I haven't had a chance to take it down again yet. I still need to make sure it can survive 24h solid of memtest, but I am happy to try other things. I'm not familiar with the snooping variable, nor with kmdb, although I have read about it being used here and there. I'll go ahead with the memtest when I can and report back. Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] diagnosing server hang
Matt Harrison wrote: m...@bruningsystems.com wrote: Hi Matt, Matt Harrison wrote: Matt Harrison wrote: thanks Ian, I'll look into this tomorrow. Well I'm not sure if it's good news or not. I've got the machine running memtest86+ with the standard tests and so far it's done 2 passes (3 hours runtime) without a single error. I'm going to leave it running overnight but does it seem there could be another problem other than memory? Have you gotten anywhere yet with this hang? Have you tried set snooping=1 in /etc/system? How about booting with kmdb and forcing a dump? I'm not sure why this is necessarily hardware related... max Unfortunately not yet...I was forced to bring the server back up to get some files from it, and I haven't had a chance to take it down again yet. I still need to make sure it can survive 24h solid of memtest, but I am happy to try other things. I'm not familiar with the snooping variable, nor with kmdb, although I have read about it being used here and there. I'll go ahead with the memtest when I can and report back. Ok, sorry it's taken a while but I've had the server run memtest for 24 hours and it hasn't found any errors whatsoever. Does anyone have an idea as to what I could try next? Thanks -- Matt Harrison ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] diagnosing server hang
m...@bruningsystems.com wrote: I would try booting with kmdb (or, alternatively, load kmdb once the machine is up but before it is hung). You can do this from command line console login (no graphics) by running: # mdb -K -- this will load kmdb and drop into it :c -- this will continue If you must have a windowing system to reproduce the hang, you can still use kmdb, but, unless you can redirect console input/output from/to a serial port, you won't be able to see what you are doing. But, it is ok. You type: # mdb -K -F -- again, loads kmdb and drops into it. The machine will appear hung. Now, carefully with no typos: : c -- and enter (that's colon c enter (3 key strokes)) the machine should come back (unless you have a typo). Now, do whatever you are doing that causes the machine to hang. When the machine is hung, type F1-a (that is function key f1 and a together. Unless the machine is hard hung, this will put you into kmdb. Again, you won't be able to see what is happening if your console is on a windowing system. Then type (again, no typos): $systemdump-- this will give you a panic dump and reboot. If the above doesn't work, you either made typos, or your machine is hard hung. If it is hard hung, add this line to /etc/system (of course, you'll have to bounce the machine to get it back up to do this): set snooping=1 Then reboot. This sets a deadman timer. Again, do your thing to cause the hang. If the scheduling clock does not run for (by default) 50 seconds, the machine will panic giving you a dump. If neither of these work, it implies that the real time clock is blocked out. This is highly unlikely, but can occur. Once you have the dump, report back... Thanks Max, I can follow that and I will try it very shortly. I am happy to move list, I'll subscribe to mdb-discuss and post my next reply there. I'll try what you've given me and post back in an hour or two. Thanks again -- Matt Harrison ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] diagnosing server hang
Matt Harrison wrote: I'll try what you've given me and post back in an hour or two. Wellnot exactly good news, but it's something. For whatever unfathomable reason, the server has stopped hanging when we copy large amounts of data from it. It has been doing it steadily for at least 2 weeks and nothing has changed except I rand the mdb commands as directed. Now, instead of a hang, the network stops. dladm shows that the link is still there, ifconfig shows it's still running but nothing is going either way. If I unplumb and re-plumb the interface it works fine again for anything between 50 and 500mb, then it drops again. Previously, the machine would just hang but now it appears to be an interface problem. I've tried rebooting to clear the mdb command, but that has made no difference at all. I'm getting to my wits end with this machine, I'm thinking about RMAing the entire box and trying something different, which is silly because I know others with exactly the same hardware that works perfectly. Incidentally, I previously had problems with the onboard realtek NICs, and only managed to get round it by replacing them with a PCI-X Intell pro 1000. Now this new NIC is giving problems, or something related to it. Grateful for any insights and sorry that this is turning into a bit of a circus. -- Matt Harrison ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] worrying hangs
Hi all, We've got a filer built on consumer hardware running SXCE snv_97, holding a small (1.4TB) raidz array. It's been going great for the last 6 months or so, but recently its started misbehaving. We use in-kernel CIFS for most of our needs and it works perfectly when playing media or mounting backed-up CD images. The problem comes when we try to explicitly copy something from it. When you actually try a direct copy via CIFS, HTTP, SSH or FTP, the transfer has about a 70% chance it will hang the machine. The larger the file, the more probable it is. I have no idea how to start investigating this as the network is inaccessible, the console is frozen and there are no hints left behind in the logs. My only way to recover the server is to shutdown (with the soft-off button on the case) and bootup. I can tell the machine isn't totally hung as it will apparently do a proper shutdown procedure. We're really out of ideas and worried that there could be a problem with our raidz array, even though there are no errors logged concerning it. As far as we can tell, there is no problem with any data (yet), just the system itself. Any ideas how to go about investigating this further? Many thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] worrying hangs
Che Kristo wrote: If you are concerned about the availability maybe move it to at least a stable opensolaris release, running of SXCE in production seems to be asking for trouble Thanks for the reply. I am aware that SXCE isn't the best release. I've been planning a move to 2009.06, but I thought if this is a hardware problem, then I should fix that first. I'll try exporting the data pool, installing 2009.06 onto the root disk and re-importing. I understand this should preserve the data perfectly. Of course, that should provide a more stable platform and I'll report back if I'm still having the hangs. Thanks ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] worrying hangs
Che Kristo wrote: If you are concerned about the availability maybe move it to at least a stable opensolaris release, running of SXCE in production seems to be asking for trouble Thanks for the push, I forced myself to make time and reinstalled the machine with 2009.06. I managed to get the pool back, setup CIFS and get everything going again in a couple of hours. So far I haven't managed to make the machine hang again, but I'll keep trying. I guess it could have been a buggy driver or something that didn't like my disks/controller. Fingers crossed it's all over. Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Some Why?-Questions
On Tue, Dec 08, 2009 at 06:10:40AM -0800, J??rg Stephan wrote: Hi there, at the weekend i installed OpenSolaris 2009.06 on an AMD Athlon64 X2 5200+ with 1GB DDR Ram. I wanted to try virtualization with VirtualBox. But i there where some things i really dont like, maybe you can help me to understand why it is and when it will work 1. Why uses OpenSolaris 95% of Ram directly after install (By the way same machine other OS takes 30%) I believe OpenSolaris caches everything it can to improve performance. As it has been discussed before, if the memory is desperately needed by another application it will be freed. 2. What about LVM and ext2,3,4 Support? I t would look better if i could use my old linux storage discs by just plugging them in. It might make migration easier, but then we wouldn't be able to use ZFS, which would be a crying shame. 3. Why does the boot of the System takes so long? I could drink a coffee. I mean the mashine isnt that slow. This can depend on several things. As someone pointed out already, the initial boot takes longer due to building the smf database. If you have a large number of ZFS filesystems, it can also increase the boot time dramatically. 4. Why doenst Alt+F1 switch to an console? Or better question... Where is the console? I'm not sure sorry, i don't run X on any of mine. After all. I really like Solaris. But this 4 Points make it a bit bad. Point 2 is the biggest problem. First i could solve by buying ram which isnt expencive, but my storage data would be necessary. Copying 400GB Data just takes too long, even when i must insert it into a different machine to make scp or so. I agree that copying massive amounts of data can be a pain, but it's not somthing that should need to be done every day. Most people would be happy to spend a day copying if it meant they could move from ext* to ZFS. I think, while some people might find a use for linux FS support, it won't be enough for it to be included in the kernel, unless you fancy writing it yourself of course :) As far as the RAM usage goes, do you find you are actually running out of RAM? Are applications dying because there is not free memory? Or is it just the free memory statistics that have you worried? Hope some of this helps you. Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] terrible idmap blunders and confusion
Hi list, A couple of years ago I set up a zfs/smb/idmap server talking to my active directory. Everything works well and except for the occasional reboots, I don't usually have any problems. I had to sync my clock with the domain controller (300 seconds out) but after that I was getting idmap errors (RPC connection problems). I followed the steps posted in a messages I found through google, which ended up removing my mappings. I thought I could remember them. Now I have idmap running, but adding some maps results in a constant GENESTATE\matt: idmap failed. I've tried maps for my main user such as: # idmap add winuser:*@genestate.com unixuser:* # idmap add winuser:m...@genestate.com unixuser:matt # idmap add winuser:*@genestate.com unixuser:* and various other permutations. Originally, I had a wildcard map like the first line above, plus a map for Domain Admins and one for Domain Users. Well nothing I can pull from my memory, or from google will allow me access to my fileserver again. I did have it all written down but that is now lost (of course). Nor can I find the orignal guides I made my notes from. Can someone please shed some light as I'm currently stuck without any access at all. Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] [SOLVED] Re: terrible idmap blunders and confusion
On 02/03/2011 00:53, Matt Harrison wrote: Hi list, A couple of years ago I set up a zfs/smb/idmap server talking to my active directory. Everything works well and except for the occasional reboots, I don't usually have any problems. I had to sync my clock with the domain controller (300 seconds out) but after that I was getting idmap errors (RPC connection problems). I followed the steps posted in a messages I found through google, which ended up removing my mappings. I thought I could remember them. Now I have idmap running, but adding some maps results in a constant GENESTATE\matt: idmap failed. I've tried maps for my main user such as: # idmap add winuser:*@genestate.com unixuser:* # idmap add winuser:m...@genestate.com unixuser:matt # idmap add winuser:*@genestate.com unixuser:* and various other permutations. Originally, I had a wildcard map like the first line above, plus a map for Domain Admins and one for Domain Users. Well nothing I can pull from my memory, or from google will allow me access to my fileserver again. I did have it all written down but that is now lost (of course). Nor can I find the orignal guides I made my notes from. Can someone please shed some light as I'm currently stuck without any access at all. Ok, turns out it was a misconfiguration of my DNS server which had only shown up after idmap was restarted/the machine was rebooted. Idmap now seems to work fine and I have access again. They key was to turn on idmap debug logging so I noticed it wasn't finding a PDC. Thanks ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] machine drops off the network
Hi all, Things have been going pretty well with my home fileserver, but recently and I think increasingly, the system is dropping off the network for just long enough to disrupt connections. It only really happens when copying large amounts of data, for example an 8 gig folder of mixed file sizes. Client: Windows 7 x64 Server: zfs/cifs on SunOS exodus 5.11 snv_133 i86pc i386 i86pc Solaris I dont run any other client operating systems so I can't test with any others. In the middle of a copy (or right at the beginning, its random) i'll get a message saying the network path couldn't be found. After I convince windows to cancel and try to access the share myself, it's all fine. My NFS shares however are a bigger problem, I've never found NFS to be very tolerant of things like this and its causing some big headaches for me. I'm no expert in tracking down this kind of stuff so any pointers are much appreciated. Thanks Matt ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] hangs all over
On 10/04/2011 23:31, Ian Collins wrote: On 04/11/11 10:29 AM, iwasinnamuk...@genestate.com wrote: On Mon, Apr 11, 2011 at 10:20:38AM +1200, Ian Collins wrote: On 04/11/11 10:06 AM, iwasinnamuk...@genestate.com wrote: Hi list, I've got a machine running OSOL...can't remember the exact version :/ Hasn't been updated in a while as I couldn't get access to the repositories. I normally runs just fine (except for a small problem probably related to the network card) however today it's really let me down. I tried to destroy some datasets but the command hung. Couldn't kill (-9) it. I was also trying to delete some files on another dataset (same pool) but that eventually hung too. There were a number of issues with zfs destroy hanging in older builds. It can still take an age if you are using dedup. Try booting from a live CD, import the pool, scrub and export. Thanks for the quick reply. I'm downloading the image again, the original was on the machine that is down of course :D I'll try to clean it up from the livecd, but out of interest: I did some reading since I posted and would I be right in thinking that it might be trying to continue the destroy, and that it might actually finish (eventually)? Possibly, or it could have hung. You could check for activity with zpool iostat, or truss/dtrace the process doing the delete. Apologies my last message went off-list. I'm trying to do too many things at once. Well I've got the livecd running and trying to import the pool. It's been running for at least 40 mins now, initially iostat reported lots of reads but its down to relatively few now. zpool iostat hangs, truss on the import process doesn't output anything at all. Disk light is still on solid. I'm not very familiar with dtrace but i'll have a read and see if I can find anything. I should probably be posting to the zfs-discuss list now I'm sure it's a ZFS problem. Maybe I'll do that if things don't look brighter soon :P Thanks ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
