Re: [OpenIndiana-discuss] ZFS remote receive
On 2012-11-01 01:47, Richard Elling wrote: Finally, a data point: using MTU of 1500 with ixgbe you can hit wire speed on a modern CPU. There is no CSMA/CD on gigabit and faster available from any vendor today. Everything today is switched. Ok then, I'll stand corrected by the practice, although my networking education speaks against this statement. We were taught that since GbE retains (at least formally) compatibility with older ethernet, and with halfduplex in particular, it is capable of hubbing, csma/cd, etc. These provisions are of course suboptimal (useless and harmful to performance) on point-to-point fullduplex links like host-to-host and host-to-switch. It may be a matter of particular modernized OS defaults however - to disable those obsolete beasts. But they should be there, and when Jumbo was introduced (for GbE, none other) - these beasts were known to cause these sorts of problems. Likely, interrupt coalescing, NIC offloads and CPU horsepower played their roles as well. And buffering, I've rather forgot about this. (Also note that buffers blindly thrown at any problem at all layers of the stack might only hide the performance degradations and disable protocol adaptability to reduced network abilities, as often happens with WiFi during interference, for example - draining the several hundred packets of the buffer over 1Mbit before even noticing that a problem exists, can take a good part of the second, if not more than one). My 2c, //Jim ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
[OpenIndiana-discuss] 64-bit PCI space
**Disclaimer: I've posted a similar question in illumos-team mailing list, however I'm reposting here since this mailing list seems much more active** Hello, I've been trying a new feature on my BIOS that allows to memory map PCI device registers to spaces above 4G. However, when I booted in Solaris, it failed to load some devices, and pci_boot also reported: unsupported 64-bit prefetch memory on pci-pci bridge [%d/%d/%d] (copy pasted that line from pci_boot.c) Digging further I also found some comments in npe.c in a register mapping function: case PCI_ADDR_MEM64: /* * We can't handle 64-bit devices that are mapped above * 4G or that are larger than 4G. */ I am in the process of rewriting the driver code right now, but I am wondering how deep will I have to go or if a fix already exists in OpenIndiana. If you are wondering why I need all this, it is because I have a device that requires a very large space of prefetchable memory (4G) and I need those BIOS settings to make it boot. Thanks ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Gnome and the future
Maybe I'm the oddball in the bunch. But I've been running OpenIndiana as my desktop OS on my laptop at work since OI-147 was released two years ago. Why? I don't care about shiny new UI gloss, but I have to have a stable system. What do I mean by that? One that can run stably without rebooting until I decide to install a newer version or switch systems and one that I'm not always having to wade through new non-security related updates just to keep the system secure. This is why I have run Slackware off-and-on for a couple decades. What brought me to OI from Slackware was ZFS. I could completely rave about ZFS, but one thing makes the point. I have to run a corporate Windows VM. Even pared down it craps itself every now and then. Previously this required a lot of pain and getting an approved reload from IT. Now all it requires is 'zfs rollback tank/vms/win-7@however-many-days-ago' and restart the VM. That's Winning! In the meantime I have discovered zones, crossbow and a few other things I absolutely love. I haven't gotten into dtrace yet, but see a lot of potential benefit there as well. All of this said, I'm about to switch to Slackware running ZFS on another machine. Why? I need the newer faster hardware with more cores for my job. There is a lack of support for the wireless and wired NIC's in the newer Dell laptop. I also seem to recall that the install disk would not fully boot on the system. This was the same reason that had me switching on and off of Slackware years ago, lack of hardware support. I know the core Illumos teams doesn't give a damn about DE or laptop support (half of them only run OI in VM's on their MacBook Pros) but I do. Had I the time and Solaris knowledge I would work to get the hardware support into Illumos, I've written a number of Linux network drivers over the years and would like to think I could figure out the superior Solaris internals. I am very sad about this. There is no reason OI could not make a good working desktop. It's been fine for my needs for two years now. I suspect given another 5 years or so we may see such. I just wish it was today. I don't need GNOME 3, KDE 4, Unity or Ubuntu. I do need modern laptop hardware to be supported. By the way I would like to congratulate Nvidia for providing good up-to-date drivers that work on OpenIndiana. The 30x series has been wonderful. Docking and undocking works flawlessly. Heck sometimes suspend works. (Not always the two together though). And yes, I too spend most of my day ssh'd to emacs'd into remote boxes, VMs, zones, chroots, etc. But I still need a DE for sanity's sake. Love my OI desktop, Ron Parker ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Gnome and the future
On 11/ 1/12 03:56 PM, Ron Parker wrote: Maybe I'm the oddball in the bunch. But I've been running OpenIndiana as my desktop OS on my laptop at work since OI-147 was released two years ago. ... There is no reason OI could not make a good working desktop. It's been fine for my needs for two years now. I suspect given another 5 years or so we may see such. I just wish it was today. I don't need GNOME 3, KDE 4, Unity or Ubuntu. I do need modern laptop hardware to be supported. You are not alone. We run all our workstations on OI-151a7, as well as all our file servers, and all our compute blades, in total 140 hosts of a couple of brands and different architectures, with Infiniband, CIFS, VLans (even on the workstations), and all that was spawned from carefully prepared single image based on OI! We are doing really serious science on these boxes (we are burning 15 Mio CPU hours per year constantly under 100% CPU load on the blades, handling about 400 TB data on the fileservers, and a similar amount offsite on PB-scale storage units, and absolutely need the desktop for expensive visualization and daily work, and have also written programs that need gtk2/Gnome. Not to speak of all the visualization packages in perl and python, wxWidgets, IDL, WebGL, that are not possible without a decent desktop, native 3D capable system. Gnome 2 is perfect (apart from a few annoying bugs), Gnome3/Unity would be a drawback indeed, and nobody here wants to go back to such basic stuff as xfce and other incapable desktop systems (KDE may would be an option). As long as gtk3 is not needed by any programs in use, there's no need to really do work in the gnome packages, apart from some fixes. Schillix OS shows that this can be handled even with just two people, so I don't see that this sucks up significant manpower. The current Gnome on OI works, there's absolutely no reason to delete that from OI. If anybody only needs a server, deactivate the desktop, maybe delete the gnome packages from your server, or install one of the other great server-only illumos-based distributions, but please don't obstruct the usability for other users by cutting of functionality just based on opinions. It's a strength and an unique feature of OI to be complete solution. We switched to OI because nobody else in the world (apart from Oracle) offers a real successor to OpenSolaris in it's full completeness. Nobody wants Linux, the apparent gain in sheer package volume comes with the drawback that you nearly always have to download and recompile this or that software yourself just because it's not compiled with all options needed, and often a whole chain needs to be deinstalled and recompiled, and so it makes no difference for us not to have a package offered or to have it offered in an unusable configuration. And Linux is clearly inferior in a couple of aspects that matter to us, apart from the fact that nobody here really has the time to administer two different flavours of Unix (with all the interoperability problems, e.g., in NFS), when one already does the job very well. Please accept that there are people that are doing much more with OI than just running a home NAS server with music and videos. We still feel that we are a legitimate part of the OI/illumos community, and not just a legacy (yes, I know, the desktop is dead, blablabla...). (And I just hate these unnecessary discussions!) -- Dr.Udo GrabowskiInst.f.Meteorology a.Climate Research IMK-ASF-SAT www-imk.fzk.de/asf/sat/grabowski/ www.imk-asf.kit.edu/english/sat.php KIT - Karlsruhe Institute of Technologyhttp://www.kit.edu Postfach 3640,76021 Karlsruhe,Germany T:(+49)721 608-26026 F:-926026 ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Gnome and the future
+1. There are alot of very useful desktop operating systems already, so i see no point in OI trying to compete with them. In business This way of thinking is responsible for what happened to Solaris. No OI should be a server and a desktop system. Otherwise, OI will die pretty soon. A.S. -- Apostolos Syropoulos Xanthi, Greece ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Gnome and the future
On Nov 1, 2012, at 12:10 PM, Udo Grabowski (IMK) wrote: Gnome 2 is perfect (apart from a few annoying bugs), Gnome3/Unity would be a drawback indeed, and nobody here wants to go back to such basic stuff as xfce and other incapable desktop systems (KDE may would be an option). As long as gtk3 is not needed by any programs in use, there's no need to really do work in the gnome packages, apart from some fixes. Schillix OS shows that this can be handled even with just two people, so I don't see that this sucks up significant manpower. This is a very informative contribution. Thank you for taking the time. I do want to ask, though: what is missing from Xfce that classifies it in your eyes as an incapable desktop? That's a pretty loaded statement and I think it deserves some explanation. My direct interest in your answer is actually because I'm cobbling together a package building system mostly for my OmniOS zones, but I expect to be running OI on my next laptop so I have a vested interest in having a stable and up to date desktop environment stack. It's been a couple of years since I ran anything but OS X on the desktop so while my recollection of Xfce is that it's a stable and complete DE, I'm ready to be corrected if there's an objective reason to not focus on bringing it to OI. The current Gnome on OI works, there's absolutely no reason to delete that from OI. Now that is something I can take issue with. While it may be perfectly usable, it's rather archaic. The DE is the first impression that many have of OI, and this is rather like going to Wal*mart (in the USA) and being greeted by an octogenarian in a blue smock. The senior citizen may do a fine job, but where's the appeal for newcomers? So you can put Lance Henrikson up front, or Miranda Kerr. I know which one I'd rather have greeting me. :) Nobody wants Linux I'll avoid the complete fallacy of that statement and merely point out that nobody is talking about Linux; we're talking about DE's. Please accept that there are people that are doing much more with OI than just running a home NAS server with music and videos. We still feel that we are a legitimate part of the OI/illumos community, and not just a legacy (yes, I know, the desktop is dead, blablabla…). Want to adopt the Gnome 2 collection? :) I'm building out some infrastructure for package building and will likely be putting together an Xfce collection. I think it would be really grand if more OI users figured out how to do this and started putting together (and maintaining!) collections for things like their favorite DE. -Magnus ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] How to disable local/remote login, still allowing access to smb share?
Thank you for the suggestion. I have not tried this yet, but I have tried to make user a role, which effectively disables login. Don't know whether smb share is still working in this scenario. Actually I am not able to connect to smb share from Windows machine in *any* case :( The http://wiki.openindiana.org/oi/Using+OpenIndiana+as+a+storage+server page gives too brief instructions. I have something missing. But I think I will figure out what's wrong, it should not be too hard. Dmitry. did you try locking the accounts (passwd -l/-N)? ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Gnome and the future
I do not feel similarly :) if it didn't feature a GUI we wouldn't notice or care. ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Gnome and the future
Yet I remember OpenWindows... Quite distinctive. WindowMaker sounds good. Used it for a while a few years ago. Been thinking about looking at it again. :) ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Gnome and the future
On Wed, Oct 31, 2012 at 2:35 PM, Jonathan Adams t12nsloo...@gmail.com wrote: from what I remember of the conversations of the time, we cannot move to Gnome 3 because of certain Linux dependencies ... Gnome 2 is no longer changing, and no longer being patched ... The problem with Gnome 3 is AFAIK mostly a problem of Illumos libc since Gnome 3 intends to move to use the ..._l() apis (e.g. strcasecmp_l(), isalnum_l(), ...) which are not present in Illumos libc. FreeBSD was confronted with the same problem (that major toolkits, including Gnome3 and Qt) and desktops are going to mandate the *_l() apis) and just did the work - see http://freebsdfoundation.blogspot.de/2011/09/following-project-update-was-written-by.html Irek ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Gnome and the future
On 11/ 1/12 02:20 PM, Irek Szczesniak wrote: On Wed, Oct 31, 2012 at 2:35 PM, Jonathan Adams t12nsloo...@gmail.com wrote: from what I remember of the conversations of the time, we cannot move to Gnome 3 because of certain Linux dependencies ... Gnome 2 is no longer changing, and no longer being patched ... The problem with Gnome 3 is AFAIK mostly a problem of Illumos libc That's just one of the problems. You'll also need better graphics driver support, unless you don't mind telling people to just stick to the Nvidia closed driver and ensure the illumos kernel remains 100% binary compatible with the private interfaces the nvidia driver uses from the Solaris kernel. -- -Alan Coopersmith- alan.coopersm...@oracle.com Oracle Solaris Engineering - http://blogs.oracle.com/alanc ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] ZFS remote receive
On Nov 1, 2012, at 1:24 AM, Jim Klimov jimkli...@cos.ru wrote: On 2012-11-01 01:47, Richard Elling wrote: Finally, a data point: using MTU of 1500 with ixgbe you can hit wire speed on a modern CPU. There is no CSMA/CD on gigabit and faster available from any vendor today. Everything today is switched. Ok then, I'll stand corrected by the practice, although my networking education speaks against this statement. We were taught that since GbE retains (at least formally) compatibility with older ethernet, and with halfduplex in particular, it is capable of hubbing, csma/cd, etc. These provisions are of course suboptimal (useless and harmful to performance) on point-to-point fullduplex links like host-to-host and host-to-switch. It does retain half-duplex compatibility. But when it is FDX, it does not use the HDX timing. I recall seeing a HDX GbE hub for sale once... but have never seen one in real life. It may be a matter of particular modernized OS defaults however - to disable those obsolete beasts. But they should be there, and when Jumbo was introduced (for GbE, none other) - these beasts were known to cause these sorts of problems. Jumbo preceded GbE, but late in the deployment of FastEthernet, so there weren't many FastEthernet switches that supported Jumbo frames, thus limiting their deployment. Everything converged with GbE -- inexpensive switches and Jumbo frame support. Likely, interrupt coalescing, NIC offloads and CPU horsepower played their roles as well. And buffering, I've rather forgot about this. (Also note that buffers blindly thrown at any problem at all layers of the stack might only hide the performance degradations and disable protocol adaptability to reduced network abilities, as often happens with WiFi during interference, for example - draining the several hundred packets of the buffer over 1Mbit before even noticing that a problem exists, can take a good part of the second, if not more than one). Indeed :-) -- richard -- richard.ell...@richardelling.com +1-760-896-4422 ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
