Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
Am Sun, 24 Jul 2011 20:08:02 -0500 schrieb Nathan Whitehorn nwhiteh...@freebsd.org: It's a change from before, but a normalization with respect to most Linux distributions, since we are now using the same dialog as, e.g., Debian and Ubuntu. -Nathan Hi, yes. And I want to thank you (and everyone) for this change in libdialog. This was something I immediately liked especially while configuring ports. It saves many key presses and is perfectly logical and provides more usability. I've done a CURRENT installation, too. And it looks like you did a good job. I've liked what I've seen so far. I have one request though: Please provide more recent snapshots for more platforms. It's very difficult to find an acceptable one. It would be nice when you get some early feedback from users instead of making a typical dot-zero release that needs to be fixed and no one will be able to accept. -- Martin signature.asc Description: PGP signature
Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
On 25/07/2011 06:01, Freddie Cash wrote: Thank goodness. The worst thing about sysinstall was that it tried to be a Swiss Army knife doing everything, yet not doing any one thing well. It made a royal mess of rc.conf if you tried to use it to configure a system. Usually the first time someone mentions they use it for post-install configuration, the recommendation is to stop doing that! An os installer should do just that: install the os and nothing else. I tend to disagree with this. For people unfamiliar with FreeBSD using it as a systems administration tool can be really useful, at least until they understand where all the various configuration files are and how they work. Having recently switched to opensuse from Ubuntu I know I find the YaST tool incredibly useful, and probably wouldn't have continued using SuSE if it hadn't been there. Its installer mode is one of the better installers I've come across, and lets you fine-tune the configuration. -- Bruce Cran ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: chromium port causing massive I/O faults
On Mon Jul 25 11, Adrian Chadd wrote: Is it perhaps doing disk IO using mmap? how can i check, whether that's the case or not? adrian On 25 July 2011 05:25, Alexander Best arun...@freebsd.org wrote: hi there, i noticed that chromium, expecially in combination with nspluginwrapper and flash, is causing a lot of I/O faults. i ran 'top -mio -I -n 99' and after only ~ 4 hours of running chromium (most of the time not loading any new pages), i got the following data: last pid: 39976; load averages: 0.37, 0.26, 0.19 up 3+02:38:30 23:15:26 72 processes: 2 running, 70 sleeping Mem: 755M Active, 662M Inact, 447M Wired, 51M Cache, 212M Buf, 45M Free Swap: 10G Total, 159M Used, 10G Free, 1% Inuse PID UID VCSW IVCSW READ WRITE FAULT TOTAL PERCENT COMMAND 39908 1001 7409 51112 0 0 4 4 0.00% /usr/local/lib/nspluginwrapper/i386/linux/npviewer.bin --plugi 39605 1001 598315 233115 11 0 3 14 0.01% /usr/local/lib/nspluginwrapper/i386/linux/npviewer.bin --plugi 1752 1001 22292378 29644471 138 0 696 834 0.38% /usr/local/bin/Xorg -nolisten inet6 1756 1001 1551733 2002630 480 0 455 935 0.43% /usr/local/bin/awesome 39140 1001 10672291 1240670 0 0 6522 6522 2.97% chrome: --type=zygote (chrome) 39116 1001 5967965 3237798 8249 20401 136394 165044 75.14% chromium-browser: (chrome) 39138 1001 6436642 994546 0 0 1785 1785 0.81% chrome: --type=zygote (chrome) 39135 1001 4334272 169320 0 0 1723 1723 0.78% chrome: --type=zygote (chrome) 39133 1001 4321593 169574 1 0 1717 1718 0.78% chrome: --type=zygote (chrome) 39132 1001 4292029 164913 6 0 1766 1772 0.81% chrome: --type=zygote (chrome) 39137 1001 4152284 139225 1 0 1762 1763 0.80% chrome: --type=zygote (chrome) 1629 560 356784 70399 25 0 40 65 0.03% /usr/local/sbin/hald 1767 1001 355603 87998 32 0 0 32 0.01% /usr/local/libexec/gam_server 39144 1001 2659919 409841 0 0 3578 3578 1.63% chrome: --type=plugin --plugin-path=/usr/home/arundel/.mozill 10217 1001 472898 258689 601 1 8 610 0.28% /usr/local/bin/musicpd /usr/local/etc/musicpd.conf 39121 1001 552140 44286 1 0 181 182 0.08% chrome: --type=zygote (chrome) 39358 1001 103237 20357 223 1479 211 1913 0.87% /usr/local/bin/sakura 39119 1001 91173 58899 2 0 14795 14797 6.74% chrome: --type=zygote (chrome) 39846 1001 275524 51575 0 0 7085 7085 3.23% chrome: --type=zygote (chrome) 39120 1001 60470 18204 0 0 22 22 0.01% chrome: --type=zygote (chrome) 1538 0 53910 6390 0 0 1 1 0.00% sendmail: accepting connections (sendmail) 39363 1001 33822 9157 1 1113 3 1117 0.51% /usr/local/bin/sakura 39805 1001 55542 43060 0 0 2787 2787 1.27% chrome: --type=zygote (chrome) 39117 1001 2935 13041 156 0 155 311 0.14% chromium-browser: (chrome) 39902 1001 43829 31005 0 0 4477 4477 2.04% chrome: --type=zygote (chrome) 362 0 28923 1878 1 0 5 6 0.00% /usr/sbin/wpa_supplicant -s -B -i wlan0 -c /etc/wpa_supplicant 1548 0 5122 672 11 0 0 11 0.01% /usr/sbin/cron -s 1217 0 13118 676 21 39 0 60 0.03% /usr/sbin/syslogd -s 39907 1001 16179 6366 0 0 2 2 0.00% /usr/local/lib/nspluginwrapper/i386/linux/npviewer.bin --plugi 39118 1001 976 716 90 0 81 171 0.08% chrome: --type=zygote (chrome) 1345 0 1362 201 1 0 2 3 0.00% /usr/local/sbin/smartd -p /var/run/smartd.pid -c /usr/local/et 1685 1001 180 22 52 0 30 82 0.04% -zsh (zsh) 1458 65534 512 62 2 0 0 2 0.00% /usr/local/bin/mpdscribble --daemon-user nobody 39360 1001 394 287 14 0 5 19 0.01% /usr/local/bin/zsh 1636 0 184 181 8 0 0 8 0.00% hald-runner 39365 1001 98 113 18 0 0 18 0.01% /usr/local/bin/zsh 1633 0 648 133 29 0 5 34 0.02% /usr/local/libexec/polkitd 1631 0 608 71 15 0 24 39 0.02% /usr/local/sbin/console-kit-daemon --no-daemon 39931 1001 53 81 1 0 1 2 0.00% /usr/local/bin/zsh 1713 1001 21 16 0 0 2 2 0.00% ssh-agent 1352 556 176
Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
on 25/07/2011 07:47 Warren Block said the following: 2. The options don't always really apply. Create when ad0 is highlighted leads the user to think they can create a new device, like ad1. But it will really create another partition. Delete on ad0 deletes all the partitions, not ad0. No warning, either. Are you sure about this one? I have never expected that any installer would be able to create or delete hardware (a hard disk) in my computer. -- Andriy Gapon ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: chromium port causing massive I/O faults
Am 25.07.2011 09:21, schrieb Alexander Best: On Mon Jul 25 11, Adrian Chadd wrote: Is it perhaps doing disk IO using mmap? how can i check, whether that's the case or not? Use truss(1) for instance. However, unless there are *practical* problems, a high number of page faults is not an indication for problems. Although it may sound scary, page faults are a feature of the memory management. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Heavy I/O blocks FreeBSD box for several seconds
on 12/07/2011 11:05 Andriy Gapon said the following: I think that the best thing you can further provide (as objective evidence for the problem at hand) is ktr(4) traces for at least KTR_SCHED mask. Perhaps you even already have them from your previous sessions with Jeff. P.S. This is not a promise to actually debug this issue based on the traces :-) So do you have an opportunity to provide this kind of information? Actually I would like KTR_SCHED|KTR_INTR|KTR_PROC|KTR_SYSC mask. Also, sysctl kern.sched output would be useful too. This is for the ULE case, of course. -- Andriy Gapon ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
On 07/25/2011 02:56, Nathan Whitehorn wrote: On 07/24/11 19:11, Claude Buisson wrote: On 07/24/2011 23:33, Nathan Whitehorn wrote: On 07/24/11 16:29, eculp wrote: I have been hearing about a new installer but I obviously have not payed enough attention, I am afraid. I started running freebsd at 2.0 and never really had a problem with understanding the installation program. There is always a first time, I guess. ftp://ftp.freebsd.org/pub/FreeBSD/snapshots/201105/ When booting I seem to get a screen that makes me remember installer screens of the 1980s. (They were not exactly intuitive.) I somehow got the idea that the new installer was graphic. Maybe something like PCBsd that is not bad at all. I use it on all our employees computers. Actually, after seeing this, I would love to have the old installer back. Is their an option for that? Does this new ASCII installer have a how to with a bit of information on the flow of the installation. Thanks, Can you please describe what you didn't like about it, and what you would prefer be changed? Reminiscent of the 1980s is not really helpful, especially given that the new installer in fact looks very much like sysinstall, which you seemed to like. -Nathan Recently I installed a system from the official memory stick May snapshot (FreeBSD-9.0-CURRENT-201105-amd64-memstick.img). here are a few remarks: Thank you for testing! My intent was not to test the installer, but I needed to install a recent 9.0-CURRENT with gpt on a brand new hardware - the 1st thing I need to do is to configure the keyboard, as I am not in the US. This is needed for an install, but also for using it as a live system. And the keyboard configuration dialog is only a part of the installation procedure. Which is why this is the very first screen of the installer? If my memory is good, it was in the first screen of the install dialog, not before the choice of installation / live system - the partition tool is too simple/rudimentary, compared to the old sysinstall dialog. I always want to have a total control of the partitions e.g. to have a proper alignement. So one must use the shell escape or the live system, which is a regression. The alignment is done to match the disk stripe size automatically, and the partition editor has many, many more features than the sysinstall one. Is there something in particular you wanted? I don't use any stripe (only plain UFS), and the many, many features where too well hidden for my old brain. - extracting the tarballs lead to (cryptic) errors: I discovered the hard way that I needed to execute a newfs. This is what the directions at the top of the partitioning shell say. As I not clearly understood these directions, I skipped to the live system for doing the gpart work. - I followed a succession of screens asking me to do the usual configuration steps (hostname, clock, network - IPv4 only ?? -, users) and at the end I get back a screen asking me if a wanted to do the steps I had done just before... The network configuration also allows IPv6 in newer versions -- that snapshot is 2 months out of date. The final screen says at the top that is there to modify earlier choices. Can you suggest a clearer wording? Clear wording is certainly a plus. - booting the installed system, I found that the hostname disappeared, the keyboard was not configured, nor the network, and so on This is inexplicable. This has worked perfectly for everyone else -- it's possible you made a mistake in the partitioning, but I can't imagine how it would have caused this. Are you able to reproduce the problem? My system is now running, and I don't have any other system to play with. - during the whole process the screen was scrambled by the occurence of a number of LORs displayed on top of the dialogs/messages of the installer. The actual 9.0 CDs will not have WITNESS enabled. It would be nice if the LORs in question were actually fixed, however. A good installer cannot suppose that there will not be any kernel message during its use, some of them will be benign. Furthermore the installer (and the whole make release process) has not for sole use the installation (and creation) of official releases. I started building my own releases at 2.2.X time.. - the file system of the installer/live system seems to be too small, leading to a number of system full messages as soon a few files are written to it. The live system is designed more as a fixit medium. What were you trying to do with it? I first copied the dmesg to be able to retrieve it on another system (was thinking that /var was a memory file system), then I saw the system full at different steps of the install. Referring to a thread I found recently a propos the documentation on the install media, I also want to say that a proper installer must be able to do its work without any Internet connectivity. There exist systems which are not connected, and networks without
Re: em problem in virtualbox since the weekend
I have same problems with em and ahci. Now in VirtialBox I temporarily set net iface to PCNet-PCI II (Am79C970A). It works with if_le driver. VirtualBox 4.1, recent FreeBSD 9. -- View this message in context: http://freebsd.1045724.n5.nabble.com/em-problem-in-virtualbox-since-the-weekend-tp4614403p4630323.html Sent from the freebsd-current mailing list archive at Nabble.com. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
On Mon, 25 Jul 2011, Andriy Gapon wrote: on 25/07/2011 07:47 Warren Block said the following: 2. The options don't always really apply. Create when ad0 is highlighted leads the user to think they can create a new device, like ad1. But it will really create another partition. Delete on ad0 deletes all the partitions, not ad0. No warning, either. Are you sure about this one? I have never expected that any installer would be able to create or delete hardware (a hard disk) in my computer. Device as in entry in /dev. It's a little blurrier than that. With no partitioning scheme, Create makes one, having the user select the type. After that, it creates new partitions. Having messed with this editor more, I can make it work and see the intent of the user interface. I wish I could suggest a good way to make it more clear, but can't quite get my brain around it right now. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
FreeBSD 9 + ZFS + MPS
I'm seeing some odd behavior on FreeBSD 9 with ZFS and an MPS controller. For two or three weeks, this configuration was working like a champ. It's a Bacula storage daemon server and was getting considerable usage - our backup set is in the multiple-terabytes-per-week of data being written to the disk. However, during the last two weeks something has broken. Starting on about July 15th or so we started to loose the data zpool which is connected via the mps driver. The disks are an external array of 32 drives, etc 2TB. When I did a zpool status, I saw that some of the drives had dropped out of the array. Rebooting the server brought them back until the next heavy write time. On July 22nd I did a make world to see if there were any kernel updates that may have fixed the problem, but any updates that may have happened did not seem to help. I'm including the last entry in my dmesg.yesterday file below. Have I stumbled upon a bug? lock order reversal: 1st 0xfe0019cf6db8 zfs (zfs) @ /usr/src/sys/kern/vfs_mount.c:829 2nd 0xfe0019cf69f8 devfs (devfs) @ /usr/src/sys/kern/vfs_subr.c:2134 KDB: stack backtrace: db_trace_self_wrapper() at db_trace_self_wrapper+0x2a kdb_backtrace() at kdb_backtrace+0x37 _witness_debugger() at _witness_debugger+0x2e witness_checkorder() at witness_checkorder+0x807 __lockmgr_args() at __lockmgr_args+0xd42 vop_stdlock() at vop_stdlock+0x39 VOP_LOCK1_APV() at VOP_LOCK1_APV+0x9b _vn_lock() at _vn_lock+0x47 vget() at vget+0x7b devfs_allocv() at devfs_allocv+0x13f devfs_root() at devfs_root+0x4d vfs_donmount() at vfs_donmount+0x988 nmount() at nmount+0x63 syscallenter() at syscallenter+0x1aa syscall() at syscall+0x4c Xfast_syscall() at Xfast_syscall+0xdd --- syscall (378, FreeBSD ELF64, nmount), rip = 0x800ab4dfc, rsp = 0x7fffccc8, rbp = 0x801009048 --- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Tim Gustafsont...@soe.ucsc.edu Baskin School of Engineering 831-459-5354 UC Santa Cruz Baskin Engineering 317B -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Heavy I/O blocks FreeBSD box for several seconds
On Mon, Jul 25, 2011 at 01:00:27PM +0300, Andriy Gapon wrote: on 12/07/2011 11:05 Andriy Gapon said the following: I think that the best thing you can further provide (as objective evidence for the problem at hand) is ktr(4) traces for at least KTR_SCHED mask. Perhaps you even already have them from your previous sessions with Jeff. P.S. This is not a promise to actually debug this issue based on the traces :-) So do you have an opportunity to provide this kind of information? Actually I would like KTR_SCHED|KTR_INTR|KTR_PROC|KTR_SYSC mask. Also, sysctl kern.sched output would be useful too. This is for the ULE case, of course. I won't have time until next week to investigate. -- Steve ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: FreeBSD 9 + ZFS + MPS
On 7/25/11 11:32 AM, Tim Gustafson wrote: I'm seeing some odd behavior on FreeBSD 9 with ZFS and an MPS controller. For two or three weeks, this configuration was working like a champ. It's a Bacula storage daemon server and was getting considerable usage - our backup set is in the multiple-terabytes-per-week of data being written to the disk. However, during the last two weeks something has broken. Starting on about July 15th or so we started to loose the data zpool which is connected via the mps driver. The disks are an external array of 32 drives, etc 2TB. When I did a zpool status, I saw that some of the drives had dropped out of the array. Rebooting the server brought them back until the next heavy write time. On July 22nd I did a make world to see if there were any kernel updates that may have fixed the problem, but any updates that may have happened did not seem to help. I'm including the last entry in my dmesg.yesterday file below. Have I stumbled upon a bug? lock order reversal: 1st 0xfe0019cf6db8 zfs (zfs) @ /usr/src/sys/kern/vfs_mount.c:829 2nd 0xfe0019cf69f8 devfs (devfs) @ /usr/src/sys/kern/vfs_subr.c:2134 KDB: stack backtrace: db_trace_self_wrapper() at db_trace_self_wrapper+0x2a kdb_backtrace() at kdb_backtrace+0x37 _witness_debugger() at _witness_debugger+0x2e witness_checkorder() at witness_checkorder+0x807 __lockmgr_args() at __lockmgr_args+0xd42 vop_stdlock() at vop_stdlock+0x39 VOP_LOCK1_APV() at VOP_LOCK1_APV+0x9b _vn_lock() at _vn_lock+0x47 vget() at vget+0x7b devfs_allocv() at devfs_allocv+0x13f devfs_root() at devfs_root+0x4d vfs_donmount() at vfs_donmount+0x988 nmount() at nmount+0x63 syscallenter() at syscallenter+0x1aa syscall() at syscall+0x4c Xfast_syscall() at Xfast_syscall+0xdd --- syscall (378, FreeBSD ELF64, nmount), rip = 0x800ab4dfc, rsp = 0x7fffccc8, rbp = 0x801009048 --- I've seen similar messages on a ZFS-based system, when rebooting, and when unmounting a USB memory stick (curiously, with an msdosfs filesystem, though my hard drive contains ZFS) ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
Quoting Warren Block wbl...@wonkity.com: On Mon, 25 Jul 2011, Andriy Gapon wrote: on 25/07/2011 07:47 Warren Block said the following: 2. The options don't always really apply. Create when ad0 is highlighted leads the user to think they can create a new device, like ad1. But it will really create another partition. Delete on ad0 deletes all the partitions, not ad0. No warning, either. Are you sure about this one? I have never expected that any installer would be able to create or delete hardware (a hard disk) in my computer. Device as in entry in /dev. It's a little blurrier than that. With no partitioning scheme, Create makes one, having the user select the type. After that, it creates new partitions. Having messed with this editor more, I can make it work and see the intent of the user interface. I wish I could suggest a good way to make it more clear, but can't quite get my brain around it right now. That makes two of us right now. I gave up, accepted the automatic partition and everything else went as expected, I suppose. The disk results are: # df Filesystem 1K-blocksUsed Avail Capacity Mounted on /dev/ada0p2 941441086 2150880 863974920 0%/ devfs 1 1 0 100%/dev In my world from the beginning of commercial unix, I have never had a one partition disk. I'm not sure if it is that bad with today's, controllers, drives, drivers, etc. I hope someone chimes in with a I see no major problems with gpt. My major problem was editing the automatic swap that was set at 4G and the menu would not let me change the 4G. The experienced option would not accept a blank value as swap even though there was message that said it would. I feel like a real idiot and am beginning to believe that it might be true. The rest of the install was brain dead. It was possibly a bit simpler than the previous. Less decisions ;) I had the idea the following were available in the new installer. 1. Raid configuration 2. ZFS 3. Regular everyday simple disk partitioning as before. I wasn't able to find any functional option except the one mentioned above. Now, I have to accept this single partition or upgrade sources to date, build a release and reinstall but I don't know if the problem has been fixed. I'll probably give it a try. It isn't that much of a deal. Thanks, ed ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
Nathan Whitehorn wrote: On 07/24/11 18:03, Ron McDowell wrote: Nathan Whitehorn wrote: On 07/24/11 16:29, eculp wrote: I have been hearing about a new installer but I obviously have not payed enough attention, I am afraid. I started running freebsd at 2.0 and never really had a problem with understanding the installation program. There is always a first time, I guess. ftp://ftp.freebsd.org/pub/FreeBSD/snapshots/201105/ When booting I seem to get a screen that makes me remember installer screens of the 1980s. (They were not exactly intuitive.) I somehow got the idea that the new installer was graphic. Maybe something like PCBsd that is not bad at all. I use it on all our employees computers. Actually, after seeing this, I would love to have the old installer back. Is their an option for that? Does this new ASCII installer have a how to with a bit of information on the flow of the installation. Thanks, Can you please describe what you didn't like about it, and what you would prefer be changed? Reminiscent of the 1980s is not really helpful, especially given that the new installer in fact looks very much like sysinstall, which you seemed to like. -Nathan I'll have to agree with the original poster. I have no problem with the look and feel of the new installer, but when functionality that WAS there is now gone, that's a problem. My two, make that three, biggest gripes are: 1) no back button/selection/mechanism on each screen. Rebooting because I fat-fingered something on the previous screen is, well, unacceptable. This is why almost all screens have a cancel button. You can also restart the installer by control-C at any time without rebooting. Providing an actual back button is quite tricky and not necessarily always well defined in behavior, since the installed system will then be in an inconsistent state at which previous steps cannot necessarily be repeated. For those steps where that is not true, they can be reentered from the menu at the end in case of fat-fingering. 22 screens require user input in a basic install [on my box, taking the default choices]. Only 7 have a 'cancel' button, which puts you back one screen, most likely to a screen you can't escape from. ctl-c to restart is about a half-step up from rebooting. How about a note at the start stating that you will be able to make changes later before committing to the install? 2) no minimal install. Most of my installs are single- or few-task servers where I need a base os and a couple ports. I'm not sure what you mean by this. You can install just a kernel and the base system by deselecting the ports tree, games, and docs when you select which system components to install. I see now that this is outside of the install program's scope...in 9.0 the number of distributions has been shrunk, so the install program has to work with what's available to it. 3) I see no post-install uses on the new one. Sysinstall could be used on an up-and-running system to do everything from adding a user to changing a nameserver and more. This is deliberate. This particular feature of sysinstall made it almost unmaintainable, especially as those features slowly bitrotted. We have very good system configuration utilities already vi-ing /etc/resolv.conf is a configuration utility? -- there is no need to duplicate them in the installer, especially when it makes maintaining and improving that installer more difficult. Fire up sysinstall on a system. Hit 'C'. Where else can a beginner go to find all this good stuff in one spot? I agree it doesn't need to be part of the install program, but it does need to be part of the OS. -Nathan One new thing I noticed is the new install does not eject the CD at the end before rebooting. I've seen systems where 'eject' didn't do anything...but it never caused a problem either. Please consider adding that. -- Ron McDowell San Antonio TX ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
On Mon, Jul 25, 2011 at 11:46 AM, eculp ec...@encontacto.net wrote: That makes two of us right now. I gave up, accepted the automatic partition and everything else went as expected, I suppose. The disk results are: # df Filesystem 1K-blocksUsed Avail Capacity Mounted on /dev/ada0p2 941441086 2150880 863974920 0%/ devfs 1 1 0 100%/dev In my world from the beginning of commercial unix, I have never had a one partition disk. I'm not sure if it is that bad with today's, controllers, drives, drivers, etc. I hope someone chimes in with a I see no major problems with gpt. My major problem was editing the automatic swap that was set at 4G and the menu would not let me change the 4G. The experienced option would not accept a blank value as swap even though there was message that said it would. I feel like a real idiot and am beginning to believe that it might be true. The rest of the install was brain dead. It was possibly a bit simpler than the previous. Less decisions ;) I had the idea the following were available in the new installer. 1. Raid configuration 2. ZFS 3. Regular everyday simple disk partitioning as before. I wasn't able to find any functional option except the one mentioned above. Now, I have to accept this single partition or upgrade sources to date, build a release and reinstall but I don't know if the problem has been fixed. I'll probably give it a try. It isn't that much of a deal. Hopefully I add something of value to this thread, but as a workaround you can use a PCBSD image and installer to install/partion plain vanilla FreeBSD with the options you mentioned earlier in a graphical enviroment. -- Adam Vande More ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: em problem in virtualbox since the weekend
On Monday, July 25, 2011 6:46:44 am timp wrote: I have same problems with em and ahci. Now in VirtialBox I temporarily set net iface to PCNet-PCI II (Am79C970A). It works with if_le driver. VirtualBox 4.1, recent FreeBSD 9. So are you having problems with the latest FreeBSD 9? Can you capture a verbose dmesg if so? -- John Baldwin ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: No disks usable on a P5NE MB
On Sunday, July 24, 2011 9:48:02 am Baptiste Daroussin wrote: Hi, Trying to upgrade one of my box from 8-stable to 9-current lead be to some important problems. I'm have tried both from sources (svn buildworld etc.) and from memdisk provided by allbsd.org. The motherboard is ASUS P5N-E SLI ACPI BIOS Revision 0901 more informations here : http://people.freebsd.org/~bapt/P5N-E.dmidecode.txt and http://people.freebsd.org/~bapt/P5N-E.pciconv.txt Can you get a verbose dmesg from 8-stable? -- John Baldwin ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
Quoting Adam Vande More amvandem...@gmail.com: On Mon, Jul 25, 2011 at 11:46 AM, eculp ec...@encontacto.net wrote: That makes two of us right now. I gave up, accepted the automatic partition and everything else went as expected, I suppose. The disk results are: # df Filesystem 1K-blocksUsed Avail Capacity Mounted on /dev/ada0p2 941441086 2150880 863974920 0%/ devfs 1 1 0 100%/dev In my world from the beginning of commercial unix, I have never had a one partition disk. I'm not sure if it is that bad with today's, controllers, drives, drivers, etc. I hope someone chimes in with a I see no major problems with gpt. My major problem was editing the automatic swap that was set at 4G and the menu would not let me change the 4G. The experienced option would not accept a blank value as swap even though there was message that said it would. I feel like a real idiot and am beginning to believe that it might be true. The rest of the install was brain dead. It was possibly a bit simpler than the previous. Less decisions ;) I had the idea the following were available in the new installer. 1. Raid configuration 2. ZFS 3. Regular everyday simple disk partitioning as before. I wasn't able to find any functional option except the one mentioned above. Now, I have to accept this single partition or upgrade sources to date, build a release and reinstall but I don't know if the problem has been fixed. I'll probably give it a try. It isn't that much of a deal. Hopefully I add something of value to this thread, but as a workaround you can use a PCBSD image and installer to install/partion plain vanilla FreeBSD with the options you mentioned earlier in a graphical enviroment. Thanks Adam, I suppose that I could just pull it into this machine and execute it as if it were sysinstall and reconfigure from the same box without having to update, make release, etc. I'm going to give that a try. Have a great day. ed -- Adam Vande More ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
On Sun, Jul 24, 2011 at 11:51 PM, Bruce Cran br...@cran.org.uk wrote: On 25/07/2011 06:01, Freddie Cash wrote: Thank goodness. The worst thing about sysinstall was that it tried to be a Swiss Army knife doing everything, yet not doing any one thing well. It made a royal mess of rc.conf if you tried to use it to configure a system. Usually the first time someone mentions they use it for post-install configuration, the recommendation is to stop doing that! An os installer should do just that: install the os and nothing else. I tend to disagree with this. For people unfamiliar with FreeBSD using it as a systems administration tool can be really useful, at least until they understand where all the various configuration files are and how they work. Having recently switched to opensuse from Ubuntu I know I find the YaST tool incredibly useful, and probably wouldn't have continued using SuSE if it hadn't been there. Its installer mode is one of the better installers I've come across, and lets you fine-tune the configuration. The difference is that YaST was designed from the get-go to be both a system management tool and a software installation tool and a system installation tool. Sysinstall was not, and sysinstall used as a post-install management tool the past couple of years has caused more issues for newbies than it's solved. If nothing else happened to sysinstall but all the post-install crud was removed from it, it would be improved a thousand-fold. Since no one has stepped up to fix the issues with the post-install management facets of sysinstall, it's only natural to remove those bits. And, since no one wants to create a new TUI management tool, there's no reason to burden the bsdinstall devs with it. Let's make an installation tool. Later, we can worry about a TUI management tool, if it's really needed. -- Freddie Cash fjwc...@gmail.com ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: No disks usable on a P5NE MB
On Mon, 25 Jul 2011 13:09:04 -0400, John Baldwin wrote: On Sunday, July 24, 2011 9:48:02 am Baptiste Daroussin wrote: Hi, Trying to upgrade one of my box from 8-stable to 9-current lead be to some important problems. I'm have tried both from sources (svn buildworld etc.) and from memdisk provided by allbsd.org. The motherboard is ASUS P5N-E SLI ACPI BIOS Revision 0901 more informations here : http://people.freebsd.org/~bapt/P5N-E.dmidecode.txt and http://people.freebsd.org/~bapt/P5N-E.pciconv.txt Can you get a verbose dmesg from 8-stable? A boot -v from current memdisk is full of: (aprobe0:ata0:0:1:0): ATAPI_IDENTIFY. ACB: a1 00 00 00 00 40 00 00 00 00 00 00 (aprobe0:ata0:0:1:0): CAM status: Command timeout here is the dmesg from boot -v on 8-stable: http://people.freebsd.org/~bapt/P5NE-dmesg-8-stable.txt the xpt_config message disapear by removing sbp from the kernel regards, Bapt ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: No disks usable on a P5NE MB
On Monday, July 25, 2011 2:42:59 pm Baptiste Daroussin wrote: On Mon, 25 Jul 2011 13:09:04 -0400, John Baldwin wrote: On Sunday, July 24, 2011 9:48:02 am Baptiste Daroussin wrote: Hi, Trying to upgrade one of my box from 8-stable to 9-current lead be to some important problems. I'm have tried both from sources (svn buildworld etc.) and from memdisk provided by allbsd.org. The motherboard is ASUS P5N-E SLI ACPI BIOS Revision 0901 more informations here : http://people.freebsd.org/~bapt/P5N-E.dmidecode.txt and http://people.freebsd.org/~bapt/P5N-E.pciconv.txt Can you get a verbose dmesg from 8-stable? A boot -v from current memdisk is full of: (aprobe0:ata0:0:1:0): ATAPI_IDENTIFY. ACB: a1 00 00 00 00 40 00 00 00 00 00 00 (aprobe0:ata0:0:1:0): CAM status: Command timeout here is the dmesg from boot -v on 8-stable: http://people.freebsd.org/~bapt/P5NE-dmesg-8-stable.txt the xpt_config message disapear by removing sbp from the kernel It seems that you have an ATAPI floppy-drive device that 8 doesn't like: afd0: setting PIO3 device_attach: afd0 attach returned 6 Have you tried enabling ATA_CAM on 8 as a test, or removing the ATAPI floppy- drive as a test? -- John Baldwin ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: No disks usable on a P5NE MB
On Mon, 25 Jul 2011 15:19:42 -0400, John Baldwin wrote: On Monday, July 25, 2011 2:42:59 pm Baptiste Daroussin wrote: On Mon, 25 Jul 2011 13:09:04 -0400, John Baldwin wrote: On Sunday, July 24, 2011 9:48:02 am Baptiste Daroussin wrote: Hi, Trying to upgrade one of my box from 8-stable to 9-current lead be to some important problems. I'm have tried both from sources (svn buildworld etc.) and from memdisk provided by allbsd.org. The motherboard is ASUS P5N-E SLI ACPI BIOS Revision 0901 more informations here : http://people.freebsd.org/~bapt/P5N-E.dmidecode.txt and http://people.freebsd.org/~bapt/P5N-E.pciconv.txt Can you get a verbose dmesg from 8-stable? A boot -v from current memdisk is full of: (aprobe0:ata0:0:1:0): ATAPI_IDENTIFY. ACB: a1 00 00 00 00 40 00 00 00 00 00 00 (aprobe0:ata0:0:1:0): CAM status: Command timeout here is the dmesg from boot -v on 8-stable: http://people.freebsd.org/~bapt/P5NE-dmesg-8-stable.txt the xpt_config message disapear by removing sbp from the kernel It seems that you have an ATAPI floppy-drive device that 8 doesn't like: afd0: setting PIO3 device_attach: afd0 attach returned 6 Have you tried enabling ATA_CAM on 8 as a test, or removing the ATAPI floppy- drive as a test? kldload atapicam is happy with it: $ camcontrol devlist SONY DVD RW DRU-820A 1.0bat scbus0 target 0 lun 0 (cd0,pass0) IOMEGA ZIP 250 41.S at scbus0 target 1 lun 0 (da0,pass1) Anyway I removed it and restart booting on the memstick but it fails the same way as before. Each time it fails booting on current, if I reset to reboot 8-stable it fails finding the disks. if I turn off and on again I am able to boot the 8-stable again... really strange. bapt ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: em problem in virtualbox since the weekend
Hmm, it was a few days ago. Something like Unable to allocate bus resource: memory Today I rebuilt latest kernel world and now em and ahci works! 2011/7/25 John Baldwin j...@freebsd.org On Monday, July 25, 2011 6:46:44 am timp wrote: I have same problems with em and ahci. Now in VirtialBox I temporarily set net iface to PCNet-PCI II (Am79C970A). It works with if_le driver. VirtualBox 4.1, recent FreeBSD 9. So are you having problems with the latest FreeBSD 9? Can you capture a verbose dmesg if so? -- John Baldwin ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: No disks usable on a P5NE MB
On Mon, 25 Jul 2011 19:37:36 +, Baptiste Daroussin wrote: On Mon, 25 Jul 2011 15:19:42 -0400, John Baldwin wrote: On Monday, July 25, 2011 2:42:59 pm Baptiste Daroussin wrote: On Mon, 25 Jul 2011 13:09:04 -0400, John Baldwin wrote: On Sunday, July 24, 2011 9:48:02 am Baptiste Daroussin wrote: Hi, Trying to upgrade one of my box from 8-stable to 9-current lead be to some important problems. I'm have tried both from sources (svn buildworld etc.) and from memdisk provided by allbsd.org. The motherboard is ASUS P5N-E SLI ACPI BIOS Revision 0901 more informations here : http://people.freebsd.org/~bapt/P5N-E.dmidecode.txt and http://people.freebsd.org/~bapt/P5N-E.pciconv.txt Can you get a verbose dmesg from 8-stable? A boot -v from current memdisk is full of: (aprobe0:ata0:0:1:0): ATAPI_IDENTIFY. ACB: a1 00 00 00 00 40 00 00 00 00 00 00 (aprobe0:ata0:0:1:0): CAM status: Command timeout here is the dmesg from boot -v on 8-stable: http://people.freebsd.org/~bapt/P5NE-dmesg-8-stable.txt the xpt_config message disapear by removing sbp from the kernel It seems that you have an ATAPI floppy-drive device that 8 doesn't like: afd0: setting PIO3 device_attach: afd0 attach returned 6 Have you tried enabling ATA_CAM on 8 as a test, or removing the ATAPI floppy- drive as a test? kldload atapicam is happy with it: $ camcontrol devlist SONY DVD RW DRU-820A 1.0bat scbus0 target 0 lun 0 (cd0,pass0) IOMEGA ZIP 250 41.S at scbus0 target 1 lun 0 (da0,pass1) Anyway I removed it and restart booting on the memstick but it fails the same way as before. Each time it fails booting on current, if I reset to reboot 8-stable it fails finding the disks. if I turn off and on again I am able to boot the 8-stable again... really strange. bapt ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org Even after having disabled the disk or even remove them, the boot still fail from both cdrom and memstick. BApt ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Heavy I/O blocks FreeBSD box for several seconds
Steve Kargl wrote: On Mon, Jul 25, 2011 at 01:00:27PM +0300, Andriy Gapon wrote: on 12/07/2011 11:05 Andriy Gapon said the following: I think that the best thing you can further provide (as objective evidence for the problem at hand) is ktr(4) traces for at least KTR_SCHED mask. Perhaps you even already have them from your previous sessions with Jeff. P.S. This is not a promise to actually debug this issue based on the traces :-) So do you have an opportunity to provide this kind of information? Actually I would like KTR_SCHED|KTR_INTR|KTR_PROC|KTR_SYSC mask. Also, sysctl kern.sched output would be useful too. This is for the ULE case, of course. I won't have time until next week to investigate. hrs sent me this panic. I'm wondering if it might be relevant to this? spin lock 0x80cb52c0 (sched lock 1) held by 0xff0012c7f8c0 (tid 100317) too long panic: spin lock held too long cpuid = 0 KDB: stack backtrace: db_trace_self_wrapper() at db_trace_self_wrapper+0x2a kdb_backtrace() at kdb_backtrace+0x37 panic() at panic+0x187 _mtx_lock_spin_failed() at _mtx_lock_spin_failed+0x39 _mtx_lock_spin() at _mtx_lock_spin+0x9e sched_add() at sched_add+0x117 setrunnable() at setrunnable+0x78 sleepq_signal() at sleepq_signal+0x7a cv_signal() at cv_signal+0x3b xprt_active() at xprt_active+0xe3 svc_vc_soupcall() at svc_vc_soupcall+0xc sowakeup() at sowakeup+0x69 tcp_do_segment() at tcp_do_segment+0x2cbd tcp_input() at tcp_input+0xcdd ip_input() at ip_input+0xac netisr_dispatch_src() at netisr_dispatch_src+0x7e ether_demux() at ether_demux+0x14d ether_input() at ether_input+0x17d em_rxeof() at em_rxeof+0x1ca em_handle_que() at em_handle_que+0x5b taskqueue_run_locked() at taskqueue_run_locked+0x85 taskqueue_thread_loop() at taskqueue_thread_loop+0x4e fork_exit() at fork_exit+0x11f fork_trampoline() at fork_trampoline+0xe --- trap 0, rip = 0, rsp = 0xff8000160d00, rbp = 0 --- KDB: enter: panic [thread pid 0 tid 100033 ] Stopped at kdb_enter+0x3b: movq$0,0x6b4e62(%rip) db ps ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: No disks usable on a P5NE MB
On Mon, 25 Jul 2011 15:19:42 -0400, John Baldwin wrote: On Monday, July 25, 2011 2:42:59 pm Baptiste Daroussin wrote: On Mon, 25 Jul 2011 13:09:04 -0400, John Baldwin wrote: On Sunday, July 24, 2011 9:48:02 am Baptiste Daroussin wrote: Hi, Trying to upgrade one of my box from 8-stable to 9-current lead be to some important problems. I'm have tried both from sources (svn buildworld etc.) and from memdisk provided by allbsd.org. The motherboard is ASUS P5N-E SLI ACPI BIOS Revision 0901 more informations here : http://people.freebsd.org/~bapt/P5N-E.dmidecode.txt and http://people.freebsd.org/~bapt/P5N-E.pciconv.txt Can you get a verbose dmesg from 8-stable? A boot -v from current memdisk is full of: (aprobe0:ata0:0:1:0): ATAPI_IDENTIFY. ACB: a1 00 00 00 00 40 00 00 00 00 00 00 (aprobe0:ata0:0:1:0): CAM status: Command timeout here is the dmesg from boot -v on 8-stable: http://people.freebsd.org/~bapt/P5NE-dmesg-8-stable.txt the xpt_config message disapear by removing sbp from the kernel It seems that you have an ATAPI floppy-drive device that 8 doesn't like: afd0: setting PIO3 device_attach: afd0 attach returned 6 Have you tried enabling ATA_CAM on 8 as a test, or removing the ATAPI floppy- drive as a test? Ok I'm dumb, I tested atapicam instead of ata_cam. I rebuild 8-stable kernel with ATA_CAM and it works (with lots of warnings concerning the dvdrw). Here is the 8-stable dmesg with boot -v: http://people.freebsd.org/~bapt/P5NE-dmesg-8-stable-ata_cam.txt Bapt ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: No disks usable on a P5NE MB
On Mon, 25 Jul 2011 21:49:44 +, Baptiste Daroussin wrote: On Mon, 25 Jul 2011 15:19:42 -0400, John Baldwin wrote: On Monday, July 25, 2011 2:42:59 pm Baptiste Daroussin wrote: On Mon, 25 Jul 2011 13:09:04 -0400, John Baldwin wrote: On Sunday, July 24, 2011 9:48:02 am Baptiste Daroussin wrote: Hi, Trying to upgrade one of my box from 8-stable to 9-current lead be to some important problems. I'm have tried both from sources (svn buildworld etc.) and from memdisk provided by allbsd.org. The motherboard is ASUS P5N-E SLI ACPI BIOS Revision 0901 more informations here : http://people.freebsd.org/~bapt/P5N-E.dmidecode.txt and http://people.freebsd.org/~bapt/P5N-E.pciconv.txt Can you get a verbose dmesg from 8-stable? A boot -v from current memdisk is full of: (aprobe0:ata0:0:1:0): ATAPI_IDENTIFY. ACB: a1 00 00 00 00 40 00 00 00 00 00 00 (aprobe0:ata0:0:1:0): CAM status: Command timeout here is the dmesg from boot -v on 8-stable: http://people.freebsd.org/~bapt/P5NE-dmesg-8-stable.txt the xpt_config message disapear by removing sbp from the kernel It seems that you have an ATAPI floppy-drive device that 8 doesn't like: afd0: setting PIO3 device_attach: afd0 attach returned 6 Have you tried enabling ATA_CAM on 8 as a test, or removing the ATAPI floppy- drive as a test? Ok I'm dumb, I tested atapicam instead of ata_cam. I rebuild 8-stable kernel with ATA_CAM and it works (with lots of warnings concerning the dvdrw). Here is the 8-stable dmesg with boot -v: http://people.freebsd.org/~bapt/P5NE-dmesg-8-stable-ata_cam.txt Bapt ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org New informations it doesn't work either on -CURRENT without ATA_CAM. regards, Bapt ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
Hi Freddie Cash! On Sun, 24 Jul 2011 22:01:44 -0700; Freddie Cash wrote about 'Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.': 3) I see no post-install uses on the new one. Sysinstall could be used on an up-and-running system to do everything from adding a user to changing a nameserver and more. Thank goodness. The worst thing about sysinstall was that it tried to be a Swiss Army knife doing everything, yet not doing any one thing well. It made a royal mess of rc.conf if you tried to use it to configure a system. Usually the first time someone mentions they use it for post-install configuration, the recommendation is to stop doing that! An os installer should do just that: install the os and nothing else. No. That's wrong. An installer should make a usable system. While using sysinstall for configuration multiple times made a mess, it is still needed to make configuration the _first time_ - and it really did, without any mess. You've got a working keyboard, TTY, network, users/passwords, etc. - before reboot. This is something which must be intuitive for a new user, even if it is used only one time in the system's life (at the installation). Cutting it - is a regression. -- WBR, Vadim Goncharov. ICQ#166852181 mailto:vadim_nucli...@mail.ru [Moderator of RU.ANTI-ECOLOGY][FreeBSD][http://antigreen.org][LJ:/nuclight] ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
Hi Freddie Cash! On Mon, 25 Jul 2011 10:36:59 -0700; Freddie Cash wrote about 'Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.': Thank goodness. The worst thing about sysinstall was that it tried to be a Swiss Army knife doing everything, yet not doing any one thing well. It made a royal mess of rc.conf if you tried to use it to configure a system. Usually the first time someone mentions they use it for post-install configuration, the recommendation is to stop doing that! An os installer should do just that: install the os and nothing else. I tend to disagree with this. For people unfamiliar with FreeBSD using it as a systems administration tool can be really useful, at least until they understand where all the various configuration files are and how they work. Having recently switched to opensuse from Ubuntu I know I find the YaST tool incredibly useful, and probably wouldn't have continued using SuSE if it hadn't been there. Its installer mode is one of the better installers I've come across, and lets you fine-tune the configuration. The difference is that YaST was designed from the get-go to be both a system management tool and a software installation tool and a system installation tool. Sysinstall was not, and sysinstall used as a post-install management tool the past couple of years has caused more issues for newbies than it's solved. If nothing else happened to sysinstall but all the post-install crud was removed from it, it would be improved a thousand-fold. Since no one has stepped up to fix the issues with the post-install management facets of sysinstall, it's only natural to remove those bits. The bad tool is better than absence of the tool. The sysinstall could at least something similar to YaST. The primary purpose is people unfamiliar with FreeBSD, of course. And for experiences - even YaST sucks in many aspects. And, since no one wants to create a new TUI management tool, there's no reason to burden the bsdinstall devs with it. Sure, no reason to burden with creation, but already existing couldbe adapted a little. E.g. disk partitioning was cutted to sade from sysinstall, the same could be done with parts of sysinstall, until something better is delivered. Let's make an installation tool. Later, we can worry about a TUI management tool, if it's really needed. The point is not a full-blown TUI tool like YaST but rather a regress in comparison with sysinstall. A something minimal must be present, not worse in features than something already existed. When later a userbase of FreeBSD will shrink due to installer issues, it will be much harder to regain it than to prevent it today. -- WBR, Vadim Goncharov. ICQ#166852181 mailto:vadim_nucli...@mail.ru [Moderator of RU.ANTI-ECOLOGY][FreeBSD][http://antigreen.org][LJ:/nuclight] ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Heavy I/O blocks FreeBSD box for several seconds
Do you have a corefile for this panic? Attilio 2011/7/25 Rick Macklem rmack...@uoguelph.ca: Steve Kargl wrote: On Mon, Jul 25, 2011 at 01:00:27PM +0300, Andriy Gapon wrote: on 12/07/2011 11:05 Andriy Gapon said the following: I think that the best thing you can further provide (as objective evidence for the problem at hand) is ktr(4) traces for at least KTR_SCHED mask. Perhaps you even already have them from your previous sessions with Jeff. P.S. This is not a promise to actually debug this issue based on the traces :-) So do you have an opportunity to provide this kind of information? Actually I would like KTR_SCHED|KTR_INTR|KTR_PROC|KTR_SYSC mask. Also, sysctl kern.sched output would be useful too. This is for the ULE case, of course. I won't have time until next week to investigate. hrs sent me this panic. I'm wondering if it might be relevant to this? spin lock 0x80cb52c0 (sched lock 1) held by 0xff0012c7f8c0 (tid 100317) too long panic: spin lock held too long cpuid = 0 KDB: stack backtrace: db_trace_self_wrapper() at db_trace_self_wrapper+0x2a kdb_backtrace() at kdb_backtrace+0x37 panic() at panic+0x187 _mtx_lock_spin_failed() at _mtx_lock_spin_failed+0x39 _mtx_lock_spin() at _mtx_lock_spin+0x9e sched_add() at sched_add+0x117 setrunnable() at setrunnable+0x78 sleepq_signal() at sleepq_signal+0x7a cv_signal() at cv_signal+0x3b xprt_active() at xprt_active+0xe3 svc_vc_soupcall() at svc_vc_soupcall+0xc sowakeup() at sowakeup+0x69 tcp_do_segment() at tcp_do_segment+0x2cbd tcp_input() at tcp_input+0xcdd ip_input() at ip_input+0xac netisr_dispatch_src() at netisr_dispatch_src+0x7e ether_demux() at ether_demux+0x14d ether_input() at ether_input+0x17d em_rxeof() at em_rxeof+0x1ca em_handle_que() at em_handle_que+0x5b taskqueue_run_locked() at taskqueue_run_locked+0x85 taskqueue_thread_loop() at taskqueue_thread_loop+0x4e fork_exit() at fork_exit+0x11f fork_trampoline() at fork_trampoline+0xe --- trap 0, rip = 0, rsp = 0xff8000160d00, rbp = 0 --- KDB: enter: panic [thread pid 0 tid 100033 ] Stopped at kdb_enter+0x3b: movq $0,0x6b4e62(%rip) db ps ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org -- Peace can only be achieved by understanding - A. Einstein ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: HEADSUP: Call for FreeBSD Status Reports - 2Q/2011
Hi Daniel Gerzo! On Wed, 29 Jun 2011 12:46:49 +0200; Daniel Gerzo dan...@freebsd.org wrote: Do not hesitate and write us a few lines; a short description about what you are working on, what your plans and goals are, or any other information that you consider interested is always welcome. This way we can inform our community about your great work! Check out the reports from the past to get some inspiration of what your submission should look like. [...] Note that the submissions are accepted from anyone involved within the FreeBSD community, you do not have to be a FreeBSD committer. Anything related to FreeBSD can be covered. Please email us the filled-in XML template which can be found at http://www.freebsd.org/news/status/report-sample.xml to mont...@freebsd.org, or alternatively use our web based form located at http://www.freebsd.org/cgi/monthly.cgi. A note for the future: you will tend to receive more submissions if you will make life a lot easier for submitter. The most natural way of using the aforementioned Web form - as, surprise, a Web form (BTW, the field to type in is too small and thus uncomfortable). That is, result of clicking Submit button must be immediately sending info to central base, requiring no further work from submitter to cut-n-paste the thing to e-mail. This is just frightening to everyone who is not a FreeBSD committer, with regard to needing to send this info to another e-mail which it suggests! (I was told that in IRC, and fro where has the Joe Random Contributor to know this?) We will have more docs when a contributor is not forced to efforts which are unneeded. Having to do such things which could be easily done on server-side looks just too unpolite for those who came in - and some of them will turn away from sending. P.S. This is general principle, not only for docs - forcing user to do something which could be already done by maintainer turns away from the system many of them. -- WBR, Vadim Goncharov. ICQ#166852181 mailto:vadim_nucli...@mail.ru [Moderator of RU.ANTI-ECOLOGY][FreeBSD][http://antigreen.org][LJ:/nuclight] ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
On 07/25/11 18:12, Vadim Goncharov wrote: Hi Freddie Cash! On Sun, 24 Jul 2011 22:01:44 -0700; Freddie Cash wrote about 'Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.': 3) I see no post-install uses on the new one. Sysinstall could be used on an up-and-running system to do everything from adding a user to changing a nameserver and more. Thank goodness. The worst thing about sysinstall was that it tried to be a Swiss Army knife doing everything, yet not doing any one thing well. It made a royal mess of rc.conf if you tried to use it to configure a system. Usually the first time someone mentions they use it for post-install configuration, the recommendation is to stop doing that! An os installer should do just that: install the os and nothing else. No. That's wrong. An installer should make a usable system. While using sysinstall for configuration multiple times made a mess, it is still needed to make configuration the _first time_ - and it really did, without any mess. You've got a working keyboard, TTY, network, users/passwords, etc. - before reboot. This is something which must be intuitive for a new user, even if it is used only one time in the system's life (at the installation). Cutting it - is a regression. That all works perfectly fine. The issue is whether it is useful for post-install configuration, which is something different entirely. -Nathan ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Trying to install current from a memory stick and then a DVD and got a new and strange installer.
On Jul 25, 2011, at 11:36 AM, Freddie Cash wrote: On Sun, Jul 24, 2011 at 11:51 PM, Bruce Cran br...@cran.org.uk wrote: On 25/07/2011 06:01, Freddie Cash wrote: Thank goodness. The worst thing about sysinstall was that it tried to be a Swiss Army knife doing everything, yet not doing any one thing well. It made a royal mess of rc.conf if you tried to use it to configure a system. Usually the first time someone mentions they use it for post-install configuration, the recommendation is to stop doing that! An os installer should do just that: install the os and nothing else. I tend to disagree with this. For people unfamiliar with FreeBSD using it as a systems administration tool can be really useful, at least until they understand where all the various configuration files are and how they work. Having recently switched to opensuse from Ubuntu I know I find the YaST tool incredibly useful, and probably wouldn't have continued using SuSE if it hadn't been there. Its installer mode is one of the better installers I've come across, and lets you fine-tune the configuration. The difference is that YaST was designed from the get-go to be both a system management tool and a software installation tool and a system installation tool. Sysinstall was not, and sysinstall used as a post-install management tool the past couple of years has caused more issues for newbies than it's solved. Um, no. Though sysinstall started life as a stop-gap until the real installer was written (which never happened), it quickly switched gears and strived to be both an installer and a configuration tool. It was designed to do both, and there are volumes of emails from the last... what... 15-18 years?... that will attest to this. The design flaw of sysinstall was that it didn't follow the model-view-controller design pattern, so over time it became harder and harder to maintain it, and it essentially rotted as the system evolved around it, despite many valiant efforts by many tireless developers. YaST did a much better job of following the MVC pattern, and it shows 10 years later. Scott ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
