Re: Compilation error libpcre.so.0: ELF file OS ABI invalid
In the last episode (Sep 28), vittorio said: > Context: Intel Core Duo, FreeBSD 6.2-p7, latest portsnap, gcc-4.2.2. > Updating the statistical program R, some packages fail to compile (I use > tarballs) because this cryptic error pops up > > * Installing *source* package 'MCMCpack' ... > checking for C++ compiler default output file name... a.out > checking whether the C++ compiler works... yes > checking whether we are cross compiling... no > checking for suffix of executables... > checking for suffix of object files... o > checking whether we are using the GNU C++ compiler... grep: error while > loading shared libraries: /usr/local/lib/libpcre.so.0: ELF file OS ABI invalid > yes > checking whether c++ accepts -g... grep: error while loading shared > libraries: /usr/local/lib/libpcre.so.0: ELF file OS ABI invalid > yes > > It seems that during the compilation a-linux-connected-lib is being looked > for, doesn't it? > > Anyway, what should I do? I'd check $PATH and maybe $LD_PRELOAD for things pointing into /compat/linux . Moving /usr/compat/linux/lib/libpcre.so.0 out of the way might change the error message and help you track down the cause. Running "ktrace -d" or "truss -f" on your port build might help too. Another possibility is that the freebsd-port version of /usr/local/lib/libpcre.so.0 has gotten damaged somehow. FreeBSD's grep doesn't link with pcre so it's sort of unlikely that that would be the cause, though. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: File size discrepancies
In the last episode (Sep 28), Modulok said: > I have a file hierarchy of about 18GiB which was copied from a UFS 2 > file system one hard disk to a UFS 2 file system on a gmirror array. > The size of the two hierarchies differ by 12 bytes according to du(1). > No errors were reported by cp(1) during the copy operation. du doesn't count in bytes; it counts in disk blocks or KB. A 12KB difference is probably just due to better packing of directories in your newly-created tree as compared to your orignal one. To figure out where the difference is, run "du -a" in both trees and diff the two outputs. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: url encoding a string with base system tools
In the last episode (Sep 29), Aryeh Friedman said: > I am creating a new port and part of the install procedure is the > install script needs to send some data to a web server in the form of > http://.?X where XXX is the url encoded plain text (can > include any ascii printable character) that needs to be sent. > > Since this for a port I want to do the plain to url encoded text with > tools *ONLY* found in the base freebsd install (no ports). How would > I do this? A google search for "url encode sed" returns some useful links, the best being http://www.mollerus.net/tom/blog/2007/06/cf_through_a_commandline_interface_part_2_programm.html although I see there are two entries for tab, one just having a space, so you'll definitely want to test it out. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: oflag option in GNU dd - equivalent in FreeBSD dd ?
In the last episode (Oct 22), Juri Mianovich said: > I am used to using this command in Linux, using GNU > dd: > > dd if=/blah of=/bleh oflag=append conv=notrunc > > The problem is, FreeBSD 'dd' does not understand the > "oflag" argument. > > Is there some equivalent in the FreeBSD 'dd' syntax > that I can use, or am I forced to install GNU utils ? Why not "cat /blah >> /bleh" ? dd is usually used on raw device nodes, and appending doesn't make sense there. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: oflag option in GNU dd - equivalent in FreeBSD dd ?
In the last episode (Oct 23), Oliver Fromme said: > Bahman M. <[EMAIL PROTECTED]> wrote: > > On 2007-10-22 Juri Mianovich wrote: > > > I am used to using this command in Linux, using GNU > > > dd: > > > > > > dd if=/blah of=/bleh oflag=append conv=notrunc > > > > > > The problem is, FreeBSD 'dd' does not understand the > > > "oflag" argument. > > > > > > Is there some equivalent in the FreeBSD 'dd' syntax > > > that I can use, or am I forced to install GNU utils ? > > Of course, the easiest way is to do this: > > $ dd if=/blah >> /bleh > > If you cannot do that, please explain why. If you know > your reason, there might be an alternative way to do it. > > > dd if=/blah of=/bleh conv=notrunc seek=`ls -s /bleh | cut -f1 -d ' ' -` > > > > I don't know if any simpler way is possible (anyone?). > > $ dd if=/blah of=/bleh conv=notrunc seek=$(stat -f%z /bleh) I don't think that will work, since seek's argument is in blocks. Even if you divide by 512 (or whatever you decide to set bs=), if the file you're appending do isn't a multiple of the blocksize, you'll end up chopping part of the end off. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Fetching sources for 6.2-Release including changes from Security Advisories
In the last episode (Oct 29), Tino Engel said: > Is there a proper entry in stable-supfile that gives me the > opportunity to cvsup the sources from 6.2-RELEASE including security > advisories? > > The tag > *default release=cvs tag=RELENG_6 > is actually not what I want, since I am not looking for 6.3-PRERELEASE You want RELENG_6_2 http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/cvs-tags.html -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: ncftpput & ncftpget
In the last episode (Oct 30), Bill Banks said:
> What port should I make to get ncftpput?
ftp/ncftp3
--
Dan Nelson
[EMAIL PROTECTED]
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: how many IPFW rules?
In the last episode (Oct 30), eBoundHost: Artur said: > Hello FreeBSD people! > > I have a smtp server under attack by what seems like a large botnet. My > inetd is choking under the load and not allowing real mail through. I've > successfully used tshark to find the offenders and put them into ipfw > firewall for port 25. > > So here is my question, I'm currently blocking 55,529 ip addresses and the > server seems pretty snappy, with no noticible load or lag. How many more > rulesets will I be able to handle before things start getting fuzzy? If you've created 55K separate rules and you're not seeing any slowdown, then you must have a fast machine :) Using an ipfw table should be even better, though. That lets you load any number of ip/netmask pairs into a tree-based lookup table and match all addresses using one ipfw rule. The ipfw manpage has examples. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: ipfw -- why need to let icmp out that I already let in?
In the last episode (Oct 31), Ivan Voras said: > [EMAIL PROTECTED] wrote: > > > add 10510 allow icmp from any to any out via oif() keep-state > > I don't think ICMP is stateful :) > > You need both in and out rules for ICMP because the logical responses > to packets can't be reliably connected into a single communication. I use "allow icmp from any to any icmptypes 0,3,11,12 in" those types being "echo reply", "destination unreachable", "time-to-live exceeded", and "IP header bad". -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: 7.0 BETA1 and cvsup
In the last episode (Oct 31), JD Bronson said: > Can someone tell me the correct tag to follow 7.0 and not 7.0 STABLE? > > I am presuming 7.0 has been frozen and I am only interested > in following the 7.0-standard (bug fixes only) and not 7.0-stable. > > I think this results in 7.0-stable? > > *default release=cvs tag=RELENG_7 There is no bug-fix-only 7.0 branch, since 7.0 hasn't been released yet. When it has, you will be able to use the RELENG_7_0 tag. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: sh script difficulties (running parallel functions)
In the last episode (Nov 01), David Naylor said:
> Hi,
>
> I am having a hard time getting (very complex script for me) to work. The
> basic idea is that this script runs a bunch of tarkets, many of which are
> time consuming but low on resources (such as downloading files). Now if I
> run the tarkets all at once (given some dependancy issues) it greatly speeds
> up the process (about 5 time speed increase). However I do not know how to
> do this using sh...
>
> Example
>
> #!/bin/sh
>
> worker1() {
> # Copy some files
> }
>
> worker2() {
> # Download some files
> }
>
> worker3() {
> # Do something else
> }
>
> . # and so on
>
> run_jobs() {
> worker1 &
> worker2 &
> worker3 &
> # !!! Somehow wait for over workers to finish before continuing !!!
> }
You can use the "wait" shell builtin to wait for all children.
> #Finished
>
> Furthermore, how can signals be handled such that the signals get
> accumulated and once all the other workers have finished the signals get
> passed on (appropriately)
You mean trapping ^C in the parent so that you can't stop the script
until all the workers are done? That's more complicated to do in a
shell script. Masking ^C completely is easy using the "trap" builtin ,
but catching it and then doing a "kill -INT $$" after all your workers
have completed is more complicated since the "wait" command will exit
when a signal is received, and I don't think it will tell you why it
exited (all children done, or got a signal).
--
Dan Nelson
[EMAIL PROTECTED]
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Best way to measure disk bandwidth usage
In the last episode (Nov 07), Andrea Venturoli said: > Doug Clements ha scritto: > >> gstat, iostat, and systat -vm are the tools I use for this. > > Thanks for the fast answer. > A couple more questions: > > gstat gives percentage values: what are these and how are they > computed? In the man page I can find no answer. Both gstat's and "iostat -x"'s %busy column is just an indicator of how often there was a pending request to the disks. If it's less than 100%, then you know the disk is idle some of the time. If it's 100%, however, that just means there's always at least one request on the queue. It doesn't mean the drive has hit any performance limits, though, since you can usually queue 32 or more commands to a single drive, and sequential requests complete a lot faster than random ones. A better indicator of overload is the command latency (ms/r in gstat or svc_t in iostat). > iostat is also expecially interesting, since it can run > non-interactively and I could poll it through cacti... However this > monitors only raw da devices. Is there a way to get gmirrors > monitored? I don't think any of the geom system hooks into devstat (which is where iostat gets its info). It'd sure be nice :) It would also allow you to query some stats via net-snmp's diskIO mib. > Finally this gives overall MB/s, which is very interesting, but I'd > also need to refer this to an end-of-scale value, in order to > understand if the disks are working to their fullest (and thus are > the bottleneck). Is this correct? Where could I desume such a value? > (I remember there was an utility... though I don't remember its > name). That really depends on your drives and the interfaces you use to connect them. For sata or sas drives, you could probably run each drive at its rated maximum speed (which will be different for each model; diskinfo -t can help here). For SCSI, FC, or drives hung off a raid controller, your bottleneck may be the speed of the shared interface or your pci bus, depending on how many drives you have connected. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Determine FreeBSD version of binary
In the last episode (Nov 08), John Smith said: > On Nov 8, 2007 6:59 PM, Yuri Pankov <[EMAIL PROTECTED]> wrote: > > May be not entirely correct, but close: > > > > ldd binary | grep libc.so > > Yes, that helps somewhat. At least I now know that it's FreeBSD 4.x. > And before I again forget something I forgot to mention earlier on: I > also have a file called 'kernel'. Could that somehow give somewhat > more detailed information about exactly which 4.x kernel it is, and > if so, how would I go about doing that ? Run "strings /kernel | tail" on it. There's also a better way to determine the FreeBSD version an executable was built for. As long as you didn't build world with -O2, the "file" command can print it. Note that you will need to run a 5.x or newer version of file, since even though 4.x puts the version in each binary, its file command doesn't print it. $ file /bin/ls /bin/ls: ELF 32-bit LSB executable, Intel 80386, version 1 (FreeBSD), for FreeBSD 7.0 (700052), dynamically linked (uses shared libs), FreeBSD-style, stripped $ file /mnt/oldsystem/bin/ls /mnt/oldsystem/bin/ls: ELF 32-bit LSB executable, Intel 80386, version 1, for FreeBSD 4.2, statically linked, stripped If you like building with -O2, apply the patch in PR 101590. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Determine FreeBSD version of binary
In the last episode (Nov 08), Erik Osterholm said: > On Thu, Nov 08, 2007 at 03:47:54PM -0600, Dan Nelson wrote: > > In the last episode (Nov 08), John Smith said: > > > On Nov 8, 2007 6:59 PM, Yuri Pankov <[EMAIL PROTECTED]> wrote: > > > > May be not entirely correct, but close: > > > > > > > > ldd binary | grep libc.so > > > > > > Yes, that helps somewhat. At least I now know that it's FreeBSD > > > 4.x. And before I again forget something I forgot to mention > > > earlier on: I also have a file called 'kernel'. Could that > > > somehow give somewhat more detailed information about exactly > > > which 4.x kernel it is, and if so, how would I go about doing > > > that ? > > > > Run "strings /kernel | tail" on it. > > > > There's also a better way to determine the FreeBSD version an > > executable was built for. As long as you didn't build world with > > -O2, the "file" command can print it. Note that you will need to > > run a 5.x or newer version of file, since even though 4.x puts the > > version in each binary, its file command doesn't print it. > > > > $ file /bin/ls > > /bin/ls: ELF 32-bit LSB executable, Intel 80386, version 1 (FreeBSD), for > > FreeBSD 7.0 (700052), dynamically linked (uses shared libs), FreeBSD-style, > > stripped > > $ file /mnt/oldsystem/bin/ls > > /mnt/oldsystem/bin/ls: ELF 32-bit LSB executable, Intel 80386, version 1, > > for FreeBSD 4.2, statically linked, stripped > > > > If you like building with -O2, apply the patch in PR 101590. > > Interesting. Does 6.2/amd64 build with -O2 by default? The default was switched from -O to -O2 in src/share/mk/sys.mk r1.81, which was before 6.x was branched, so none of the 6.x releases have versions in their executables. > $ file /bin/ls > /bin/ls: ELF 64-bit LSB executable, AMD x86-64, version 1 (FreeBSD), > dynamically linked (uses shared libs), stripped > > This was after a buildworld with no special options added, and > nothing affecting the kernel in make.conf. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: problems using ls with for_in (SH)
In the last episode (Nov 08), Sdvtaker said: > Im trying to get a file with all the md5 hashes of one directory. My > initial script was this: > > #!/bin/sh > for file in $(ls) > do >echo $file >md5 $file > done > > The problem is with the file names who contains "whitespaces" becouse > the for_in passed each word as one iteration and not the full > filename, I'd tried using -B in ls, but doesnt help. There's no need to call ls at all. The shell can expand wildcards just fine by itself: for file in * do echo $file md5 $file done but in your case, since md5 can take multiple filenames on its commandline and prints the filename in its output: md5 * will suffice. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: ps options
In the last episode (Nov 10), Chuck Robey said: > I have spent all the time I can stand, going over the ps man page, but I > can't see any option to get a hierarchical listing. I mean, where the > listings are sorted to where parents come before children, and the children > get indentation, so you can see at a glance what's running more easily. > It's a standard thing on many OSes, and I was sorta hoping it'd be > available on FreeBSD. Maybe under a different name? It's usually a separate command (ptree on Solaris for example). Try the sysutils/pstree port. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: can't think of program
In the last episode (Nov 11), Robert Huff said: > There is a program out there that will allow you to run a child > program, read what it sends to stdout, and feed things to stdin using > a script-like language. > I am blanking on the name of this program, My mind says it > should be "expect", but that's not right. Trust your mind :) It's expect. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: 7.0, make buildworld without contrib. old top binary in contrib.
In the last episode (Nov 14), peceka said: > Hi, > > can someone tell me if there is any possibility to do make buildworld > without compiling all contrib/ directory? I don't need for example bind9 > in base, ipfilter, lukemftp(d). > And why in /usr/src/contrib is very old top binary, 3.5beta12? On > http://www.sourceforge.net/projects/unixtop there is top-3.7beta2 where > ie. -c option, very helpful, works. FreeBSD's top recently got an "-a" option that seems to do the same thing. It might be nice if this were renamed to -c before 7.0 gets released, to make subsequent merges from unixtop easier. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: How to know PID responsible for network connection/listen?
In the last episode (Nov 15), Yuri said: > 'netstat -a' gives me the listing of network connection/listening > records. But there's no link to the process id that opened it. With > lots of processes this can be a significant problem to figure out who > opened which connection. Try /usr/bin/sockstat or the sysutils/lsof port. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: mbuf allocation on FreeBSD 6.x
In the last episode (Nov 21), Jay Aikat said:
> I recall being able to set kern.ipc.nmbufs in the older
> versions of FreeBSD. How do we set the same in 6.x - it does not seem
> to exist. Or maybe there is a different way to manage mbufs?
According to the mbuf manpage, as of FreeBSD 5.3 mbufs are just regular
kernel-malloced memory and have no hard limit apart from available
memory.
--
Dan Nelson
[EMAIL PROTECTED]
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: running port as non-root
In the last episode (Nov 30), rihad said:
> Hi, I'm trying to run the net/openradius port as non-root
> by first changing /usr/local/etc/rc.d/openradius:
>
> . /etc/rc.subr
>
> name="openradius"
> rcvar=`set_rcvar`
>
> load_rc_config ${name}
>
> : ${openradius_enable="NO"}
> : ${openradius_flags="-o /var/log/openradius.log"}
>
> -command=/usr/local/sbin/radiusd
> +command=/usr/local/bin/sudo
> +flags="-u radius /usr/local/sbin/radiusd"
It's much cleaner to simply set openradius_user=radius and let rc.subr
handle the userid switching using su.
> run_rc_command "$1"
>
> (Lines marked with -/+ were removed/added by me, respectively).
>
> Then I add the radius user, allow it to run without sudo's asking for
> the password, and finally start the thing up:
> # pw useradd radius -d /nonexistent -s /usr/sbin/nologin
> # visudo
> # ... do the necessary chown/chmod on openradius logs/dictionaries ...
> # /usr/local/etc/rc.d/openradius start
>
> All fine so far: everything starts up fine and runs. The problem starts
> here:
> # /usr/local/etc/rc.d/openradius stop
> openradius not running?
> # ps -auxww | fgrep radiusd
> radius 89300 0.0 0.0 3756 1588 ?? Is 10:56AM 0:00.00
> /usr/local/sbin/radiusd
> ...
>
> any built-in way to make "stop" work, without changing rc.d/ too much?
Problem is you set command=/usr/local/bin/sudo, so the stop sequence
isn't looking for radiusd at all.
For more info on rc scripts, see the rc.subr manpage and
http://www.freebsd.org/doc/en_US.ISO8859-1/articles/rc-scripting/index.html
--
Dan Nelson
[EMAIL PROTECTED]
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: cron pile up! Lot's of "cron: running job (cron)"
In the last episode (Dec 03), Support (Rudy) said:
> Below is part of the cron... Seems like any random cronjob can get
> clogged up... load varies from 0.2 to 1.0 on this dual-core box. I
> rebooted the box -- cron's continue to slowly pile up.
>
> One of the cronjobs that is 'stuck' is this one: /root/bin/raid-status.sh
> which can be found here:
> http://www.monkeybrains.net/~rudy/example/raid_status.html
>
> Forgot to mention, I am running:
> 6.2-STABLE FreeBSD 6.2-STABLE #3: Thu May 31 01:18:15 PDT 2007
>
> OH, ps shows this:
> 58383 ?? D 0:00.00 cron: running job (cron)
> 58384 ?? IVs0:00.00 cron: running job (cron)
In general, when troubleshhoting, "ps axlw" is a more useful command.
It adds among other columns, the MWCHAN one, which details exactly why
a process is stuck in the D state.
Anyway, cron does a fork and then a vfork creating a child and a
grandchild process. I'm sort of surprised at the amount of code
between vfork and exec in the grandchild in
/src/usr.sbin/cron/cron/do_command.c . Since process 3 is actually
using process 2's address space one must be extremely careful not to
modify static variables or change other global state that would affect
the parent once it resumes execution, and all the logging,
environment-setting, and user-context calls are certain to mess with
the parent's state, especially with nss modules in the mix. I'd
personally recompile cron with all vforks replaced with fork and see
what happens.
It couldn't hurt to update to a newer kernel version along the RELENG_6
branch as a test, I guess. Note that your uname will change to
6.3-PRERELEASE, but apart from causing lsof to complain, you should be
okay.
> /var/log/cron has this entry:
> Dec 3 20:16:00 pita /usr/sbin/cron[58384]: (root) CMD
> (/root/bin/raid-status.sh CRON)
>
> BUT there is no 'raid-status.sh' stuck in the "ps axw". Seems like the
> vfork set off the cronjob, it ran, but then cron didn't 'stop' executing.
> Any debuggin tips?
Can you tell if raid-status.sh ever ran? i.e. is process 2
stuck at the start of vfork or at the end.
BTW, here's a minimal example of the danger of putting code between
vfork and exec:
#include
#include
#include
int main(void)
{
int i = 1;
switch (vfork())
{
case -1:
err(1, "vfork failed");
break;
case 0:
/* child */
i = 2;
execl("/usr/bin/true", "true", NULL);
_exit(0);
break;
default:
break;
}
printf("in parent, i is %d\n", i);
return 0;
}
--
Dan Nelson
[EMAIL PROTECTED]
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Character 208 acts strangs in console, when moving mouse
In the last episode (Dec 05), Reko Turja said: >> When i display character 208 in my console and move my mouse, i see >> the strangest things. FreeBSD 6.2-RELEASE-p8, with default screenmap >> and console settings. >> >> http://www.xs4all.nl/~whendrik/download/mouse >> >> wget and cat that file in a console to see the effect. >> >> I tried it on my laptop and desktop PC with same kernel. Anyone else >> experience this problem/effect? I think it should be a pi character >> upside down, mostly used to draw tables in combination with other >> characters... > > Mouse cursor mapping artifact. In text mode on PC hardware, the mouse > pointer has to be mapped to a character in order to show "fancy" pointer. > Nothing to worry about. Someone with a lot of time could modify syscons to treat the 256 hardware VGA characters as a "window" onto a larger character space, and dynamically remap them as needed. That and a unicode VGA font (like at http://www.inp.nsk.su/~bolkhov/files/fonts/univga/ ) would allow a utf-8 console to display the 256 most common characters on the screen (252 if the mouse is onscreen) whatever they happen to be. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: FreeBSD NFS server & Linux clients
In the last episode (Dec 05), Ian said: > I'm trying to get some opensuse 10.3 clients to access nfs shares on > a FreeBSD 5.5 server. It works fine most of the time, but every so > often when using big programs like openoffice & firefox, the program > will stop responding for a couple of minutes. > > After doing some googling, I enabled lockd & statd enabled on the > FBSD server - before I did that, those apps wouldn't even start > without massive delays. I still get these occasional lockup while > using the apps however. Interestingly, other apps seem to be able to > access the nfs shares even while the 'big' apps have stopped > responding. > > Just wondering if anyone has any idea how to fix or even troubleshoot > this problem? strace the hanging linux process to see if it always happens on a particular file or syscall, and run tcpdump/wireshark to see what NFS requests are being done. Also check dmesg or /var/log/messages on the client to see if it's reporting something there. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: loading modules not in /boot/kernel from loader.conf
In the last episode (Dec 10), Aryeh M. Friedman said: > sysutils/fusefs-kmod installs to /usr/local/modules/fuse.ko and a > fuse_load="YES" will not work because of this. But, mv/cp/ln'ing it > to /boot/kernel gets nuked if you rebuild/install the kernel. How to > get around this? The loader also checks /boot/modules/ , so copy your stuff there. That's where the kqemu-kmod port puts kqemu.ko, for example. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: [5.x] building. 4.x binaries ?
In the last episode (Nov 08), Jacques Beigbeder said:
> Since FreeBSD 5.x include 4.x dynamic libraries, is there any way on
> a 5.x box to build (with gcc) binaries which will run on a 4.x box?
No, since you also need 4.x headers and 4.x static libraries. You can
do this if you install 4.x in a chroot, or you can set up a vmware or
bochs virtual machine running 4.x
--
Dan Nelson
[EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Tool for Unix similar to pingplotter (http://www.pingplotter.com)
In the last episode (Nov 08), Marc G. Fournier said:
>
> Does anyone know of a tool available under X similar to this?
Well, mtr can give you the traceroute in a window, and smokeping can
give you the historical reporting. I don't know of a tool that does
both at once.
--
Dan Nelson
[EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: NFS v2? possible?
In the last episode (Nov 09), Chad Leigh -- Shire.Net LLC said: > On Nov 9, 2003, at 4:28 AM, Kris Kennaway wrote: > >On Sun, Nov 09, 2003 at 01:33:23AM -0700, Chad Leigh -- Shire.Net LLC > >wrote: > >>I have a linux server that needs to mount my FBSD server's web > >>volume and the linux server only has v2 support right now (and I > >>cannot make my own kernel since this is a UML kernel distributed as > >>part of a product) and so I would like to run a v2 compatible nfs > >>server on my FBSD server if possible. Can I do this? How? Thanks > > > >Read the mount_nfs and/or nfsd manpages..the answer is right there. > > The mount_nfs is for mounting on FreeBSD. I am serving from FreeBSD > and mounting on Linux, so that does not apply. I read the nfsd man > page a few time before posting, and just did again, and I see nothing > in their about nfs versions except that the server listens on the > port as outlined in thge NFS v3 spec. Please enlighten me on what I > should read in the nfsd man page. You don't have to do anything actually. FreeBSD serves both NFSv2 and NFSv3 clients by default: $ rpcinfo -p | grep nfs 132 udp 2049 nfs 133 udp 2049 nfs 132 tcp 2049 nfs 133 tcp 2049 nfs You can disable NFSv3 with the -2 option to mountd, but if you have to do that, then your Linux distro is extremely broken and you should have your vendor fix it. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: file sizes
In the last episode (Nov 12), Bryan Cassidy said:
> Trying to figure out how to find out the size of a file.
ls -l myfile
--
Dan Nelson
[EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: HP JetDirect EX printer problems ...
In the last episode (Nov 12), David Bein said: > I have an older J2383 Jet Direct box which I am trying to setup for a > FreeBSD 4.9/x86 box. I have tried various things in /etc/printcap, > but still I get nothing reasonable. I know the box is working because > if I telnet to port 9100 on the Jet Direct and type in some lines and > then quit out of telnet, it prints them. So my problem is not the > hardware or network connectivity. > > I am confused about whether this should be "lp=@" with > appropriate filters or some variant of "rm= rp=TEXT". Either should work. I prefer the "rm" style since that works a bit better when you have multiple systems printing to the same printer. Make sure to telnet to the jetdirect and disable banners, though. > None of the software examples I have seen appear to reference port > 9100 and so I am confused as to how it knows what to connect to > except in the @ configuration. Neither of these > approaches worked. What happens? -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: quick question about turning the annoying beep off in X
In the last episode (Nov 13), twig les said: > Hey all, in my /etc/csh.cshrc I have > > xset b off > > which does what I want, which is to tell X to shut up without me > having to type the command in every X session. But it seems clumsy > as it puts up an error "xset: unable to open display """ when I ssh > in. I fully understand that this *should* be an error because > csh.cshrc is a shell init file, I just tried this in .xinitrc and it > didn't work. Putting it in > /usr/X11R6/bin/startx had predictably bad results. > > So anyone know the correct file for this command? I'm pretty sure .xinitrc is the one you want. Make sure you run the command before any session managers etc. I just tested it and it works for me. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: cc (make.conf and make args precedence) question
In the last episode (Nov 15), Ion-Mihai Tetcu said: > buh# grep 'CFLAGS' /etc/make.conf > CFLAGS= -O2 -pipe > > buh# grep -A1 '(BUILD_OPTIMIZED' > /usr/ports/databases/mysql40-server/Makefile > .if defined(BUILD_OPTIMIZED) > CFLAGS+=-O3 -fno-omit-frame-pointer > > > cc -c -I. -I../.././bdb/dist/../include -D_THREAD_SAFE -DDBUG_OFF -O2 > -pipe -march=k6-2 -O3 -fno-omit-frame-pointer -fno-gcse > ../.././bdb/dist/../btree/bt_reclaim.c > > Now what flag have precedence, -O2 or -O3 ? Later flags always override earlier ones, so -O3 wins. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: non-proportional fonts for X
In the last episode (Nov 17), Harald Schmalzbauer said: > perhaps someone achieved something useful from the search for a good > mail font. I don't like misc-fixed for mails (I use kmail) so I took > courier-new 9pt. But with that font the 1 and l are looking very > similar with less that 10pt. And 10pt is too big for me. So I > switched to Lucida typewriter. Readability at 9pt is quiet good but > I'd like to know your favourites. My preferred TTF fixed-width font is LetterGothic Line (letgothl.ttf), but it only looks good in Windows. Under X there is too much horizontal space between the letters for some reason. 2nd choice is Vera Bitstream Mono, which has very distinct 1/l 0/O pairs, but has thinner vertical stems, so it gets washed-out at small sizes. If you prefer bitmap fonts, Terminus (x11-fonts/terminus-font) is pretty good. I use the bold flavor as my xterm font. The version in ports has almost identical o and a, though. I need to send in a PR to have the port updated to the current one. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Ethereal 0.9.16?
In the last episode (Nov 20), [EMAIL PROTECTED] said: > Has anybody out there had success in getting Ethereal-0.9.16 > up&running? > > The ports collection for 4.9 still lists 0.9.14. Any clean way to > upgrade this to 0.9.16? Try just editing the version number in the Makefile, run "make makesum", then build the port as usual. If you're lucky it'll build without errors. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Question
In the last episode (Nov 20), VastNET said: > Hello! > > Do you know what's the reason of it? My machine is rebooting few times a day. > > If answer is YES, what should I do? > > savecore: reboot after panic: page fault > <118>savecore: reboot after panic: sbflush: cc 0 || mb 0xc1818500 || mbcnt 2304 > <118>Nov 20 17:21:10 gateway savecore: reboot after panic: sbflush: cc 0 || mb 0 > xc1818500 || mbcnt 2304 > <118>savecore: reboot after panic: sbflush: cc 0 || mb 0xc1818500 || mbcnt 2304 > <118>Nov 20 17:21:10 gateway savecore: reboot after panic: sbflush: cc 0 || mb 0 > xc1818500 || mbcnt 2304 Since if looks like you have crashdumps already enabled, follow the instructions at http://www.freebsd.org/doc/en_US.ISO8859-1/books/faq/advanced.html#KERNEL-PANIC-TROUBLESHOOTING , and let us see the stack trace that gdb prints. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Installed ports
In the last episode (Nov 21), Valerian Galeru said:
> Pls tell me where are all the installed ports? After i installed the port, where i
> can find the bin file for the port?
Try /usr/local/bin or /usr/X11R6/bin.
--
Dan Nelson
[EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Process substitution in bash - doesn't work
In the last episode (Nov 21), Herculano de Lima Einloft Neto said: > > Hello all, > > Whenever I try something like vi <(diff file1 file2) in bash I get > an empty file on /dev/fd/63.. is there a way to make it work? Or does > FreeBSD not support this? Do you have fdescfs mounted on /dev/fd ? -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Bit Torrent
In the last episode (Nov 21), Ralph said:
> Was wondering if anyone out there knows of a bittorrent port for
> freebsd? I am running a tracker now in windows and would love to do
> it in Freebsd.
ports/net/py-bittorrent
--
Dan Nelson
[EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: fsck error messages don't get logged?
In the last episode (Nov 23), Herculano de Lima Einloft Neto said: > Lowell Gilbert wrote: > > >You can't scroll the window back to the messages? > > I just found out about Scroll Lock, checking better.. was convinced I > had some sort of keyboard problem because Shft-PgUp never worked :). > Well, tough luck.. I had already booted again, so the messages are gone. > It would be nice to have them logged. They should go into /var/log/console. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: POLA violation?: snmp renumbering stuff
In the last episode (Nov 23), paul beard said:
> For some reason, my locally installed snmp daemons decided to
> renumber the elements in the hrStorageTable, meaning all the attached
> disks were being either misreported or just plain dropped from my
> graphs (paulbeard.no-ip.org/mrtg/blue/index.html). Not that the new
> numbering doesn't make sense but I didn't know this was going to
> happen.
>
> How to discover and fix it? snmptable is my friend. As shown here,
> the memory used by the kernel is listed first, followed by the disks.
> The disks were numbered starting at 1 before . . . . .
I don't think snmp tables have any defined order. I don't even know if
the index for a particular resource is guaranteed to be stable across
filesystem dismount/remounts. Something like this should work:
snmptable -Cf : blue hrStorageTable | grep :/var: | awk -F : '{print $4 * $5}'
I use something similar in a script to graph disk usage in mrtg. It
sould be really nice if snmptable had a built-in flag to print a
particular cell from a table, though.
--
Dan Nelson
[EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Fwd: [bn@vastnet.co.uk: Re: Question]
- Forwarded message from VastNET <[EMAIL PROTECTED]> - Date: Sun, 23 Nov 2003 20:47:14 - From: VastNET <[EMAIL PROTECTED]> To: Dan Nelson <[EMAIL PROTECTED]> Subject: Re: Question X-Mailer: Microsoft Outlook Express 6.00.2720.3000 GNU gdb 4.18 (FreeBSD) Copyright 1998 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "i386-unknown-freebsd"...Deprecated bfd_read called at /usr/src/gnu/usr.bin/binutils/gdb/../../../../contrib/gdb/gdb/dbxread.c line 2627 in elfstab_build_psymtabs Deprecated bfd_read called at /usr/src/gnu/usr.bin/binutils/gdb/../../../../contrib/gdb/gdb/dbxread.c line 933 in fill_symbuf IdlePTD at phsyical address 0x0074 initial pcb at physical address 0x00371000 panic messages: --- dmesg: kvm_read: invalid address (c0365a98) --- #0 0xc034d3e0 in sysctl___compat () (kgdb) where #0 0xc034d3e0 in sysctl___compat () #1 0xc0389600 in sysctl__kern_children () #2 0xc03104fc in __set_sysctl_set_sym_sysctl___vfs_nfs_nfsv3_commit_on_close () cannot read proc at 0 (kgdb) > In the last episode (Nov 20), VastNET said: > > Hello! > > > > Do you know what's the reason of it? My machine is rebooting few times a day. > > > > If answer is YES, what should I do? > > > > savecore: reboot after panic: page fault > > <118>savecore: reboot after panic: sbflush: cc 0 || mb 0xc1818500 || mbcnt 2304 > > <118>Nov 20 17:21:10 gateway savecore: reboot after panic: sbflush: cc 0 || mb 0 > > xc1818500 || mbcnt 2304 > > <118>savecore: reboot after panic: sbflush: cc 0 || mb 0xc1818500 || mbcnt 2304 > > <118>Nov 20 17:21:10 gateway savecore: reboot after panic: sbflush: cc 0 || mb 0 > > xc1818500 || mbcnt 2304 > > Since if looks like you have crashdumps already enabled, follow the instructions at > http://www.freebsd.org/doc/en_US.ISO8859-1/books/faq/advanced.html#KERNEL-PA NIC-TROUBLESHOOTING > , and let us see the stack trace that gdb prints. > > -- > Dan Nelson > [EMAIL PROTECTED] - End forwarded message - -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: IPFILTER rules with shell symbloic substitution
In the last episode (Nov 26), fbsd_user said:
> Looking for way to code IPFILTER rules with shell symbolic
> substitution. I know how to do it in the rules, but how to get ipf
> pgm to exec in shell? Any body have example of how to setup this?
/etc/rc.firewall has lots of examples using ipfw; the concepts should
work just as well with ipf.
--
Dan Nelson
[EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: unneeded modules
In the last episode (Nov 28), Florin Betivoiu said: > I would like to know if there is any way to not compile in the kernel > drivers I don't need. For example kldstat -v shows that lots of > miibus modules are loaded when i just need pnaphy, or lots of agp > modules when I just need intel. Please excuse my poor english, as I > am not a native speaker. Thanks in advance for any suggestions, > pointers. In your kernel config file, you can use MODULES_OVERRIDE to build only the modules you need. Modules are defined in the /sys/modules directory, though, and it looks like there is a single "mii" module that includes support for all chipsets. Same for the agp device, so you have to build all of them or none. The sound driver is a bit more modular, and you can choose to build only the driver for a single chipset. See /sys/confi/NOTES for a description of the syntax for MODULES_OVERRIDE. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: unneeded modules
In the last episode (Nov 28), Florin Betivoiu said: > Ok. With this config file: http://tinfed.dap.ro/PONY I compiled a > kernel that has 2.1MB in size. Probably I shouldn't compare it with > others, but is it not too big? That's pretty small. My kernels are between 3 and 5MB, but I don't use modules at all (I compile everything into the kernel), and I build lots of drivers so I can swap SCSI and network cards. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: which version of automake/autoconf?
In the last episode (Nov 28), Aaron Walker said: > I was getting ready to install the automake and autoconf ports, when I > noticed there's 4 different versions of each. How do I know which ones > to install? Install the version you need :) The numbered ports will all coexist. The unnumbered port is the "default" one, which is not necessarily the best. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: very intresting question about groups
In the last episode (Dec 02), Konstantin Gritsenko said: > ./sys/sys/syslimits.h: > #define NGROUPS_MAX16 > /* max supplemental group id's */ > > how to explain this limitation? > and what can be if i increase this value and make "make World" I think NFS access may break, either failing completely with "bad RPC call" errors, or truncating the group list to the first 16 entries when accessing remote files. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: weird mouse reaction and movement 5.1
In the last episode (Dec 02), charles pelletier said: > So I just installed, for the first time, 5.1 on a new machine. My > problem is this: In Sysinstall, when i select my mouse (ps/2, MS 3 > button optical) then enable and test the port, all of a sudden I see > an early part of the installation wizard (the partitioning part) and > i can't escape out of it. It is as if the mouse scrolls through a > history of what i've done page by page until it reaches that part. I > can ctrl-c out of sysinstall and it is as if nothing ever happened. > > Is this a strange event to anyone else? Is there a way to fix it w/ > flags, etc? It was suggested to me by a friend that I kill moused > before I start up sysinstall but it isn't running in the first place. I've never seen that before, but you can probably just skip the mouse configuration part, and then set it up after the install is done (for your mouse, just adding moused_enable="YES" to /etc/rc.conf should suffice, since ps/2+autosense is the default). -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: growfs says 'we are not growing' ?!?!
In the last episode (Dec 04), Rishi Chopra said: > Trying to enlarge a partition (/usr) on a slice (da0s1) with plenty of > space: > > # fdisk > > The data for partition 1 is: > sysid 165 (0xa5), (FreeBSD/NetBSD/386BSD) >start 63, size 1171861362 (572197 Meg), flag 80 (active) > beg: cyl 0/ head 1/ sector 1 > end: cyl 1023/ head 254/ sector 63 > > # df -h: > > Filesystem Size Used Avail Capacity Mounted on > /dev/da0s1a 193M 80M 98M 45% / > /dev/da0s1d 193M 5.9M172M 3%/var > /dev/da0s1e 1.9G 670M 1.1G 37% /usr > > But when I try: > > # umount /usr > # growfs /dev/da0s1e You need to extend your e slice (with disklabel -e da0s1) first. You've got three "container" objects: Fdisk slices, BSD partitions, and filesystem. You've grown the slice, but you need to also expand the partition before growfs can resize the filesystem. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: growfs says 'we are not growing' ?!?!
In the last episode (Dec 04), Rishi Chopra said:
> Any idea why my partitions (da0s1a, da0s1d, da0s1e) don't have valid
> disk labels?
>
> # disklabel -A da0s1d | more
> disklabel: /dev/da0s1d: no valid label found
> # disklabel -A da0s1e | more
> disklabel: /dev/da0s1e: no valid label found
>
> I created the partitions and mount points when installing the os
> using the standard GUI.
You don't run disklabel on partitions; you run it on slices. Most
disks will have a single disklabel, which creates all the partitions
needed. Let me see if I can do an ascii-art drawing of a resize
operation:
Original system:
1 |__DISK__|
2 |MBR|__Slice 1_|__Slice 2|
3 |_a_|___b___|__d___|___e__|
4 |_/_| |_/var_|_/usr_|
Row 1 is the physical layout. Row 2 is composed of the MBR and fdisk
slices. Row 3 is BSD disklabels, and row 4 is filesystems. In this
example, we'll say da0s1 is a non-unix filesystem and ignore it.
da0s2b is swap, which always uses the entire space given to it, which
is why I didn't extend it to row 4. The disklabel header for Slice 2
is actually embedded in the first block of the first partition which is
why there's no |DL| on row 3 to match the |MBR| for slices on row 2.
UFS filesystems skip the first 8k or 64k in a partition to leave room
for the disklabel header and boot blocks.
Grow RAID array:
_
|DISK_|
|MBR|__Slice 1_|__Slice 2|
|_a_|___b___|__d___|___e__|
|_/_| |_/var_|_/usr_|
Run fdisk on da0 and grow slice 2:
_
|DISK_|
|MBR|__Slice 1_|___Slice 2|
|_a_|___b___|__d___|___e__|
|_/_| |_/var_|_/usr_|
Run disklabel on da0s2 and grow partition e:
_
|DISK_|
|MBR|__Slice 1_|___Slice 2|
|_a_|___b___|__d___|___e___|
|_/_| |_/var_|_/usr__|
Run growfs on da0s2e and grow the filesystem:
_
|DISK_|
|MBR|__Slice 1_|___Slice 2|
|_a_|___b___|__d___|___e___|
|_/_| |_/var_|_/usr__|
--
Dan Nelson
[EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: the proces that can be monitored
In the last episode (Dec 05), M.D. DeWar said: > Hello, > I am trying to get net-smnp setup on my system. > I'm at the point were it asks for process to monitor but am not sure what > ones I can and what they are are called. > In the example says httpd for http so I know that one > but not sure what it is for > memory > disk space > etc > a google search got me nothing useful (but I may not be typing it in right > cuz i am not sure what I am looking for) I think you can just skip that question, unless you want snmpd to monitor the process and send a trap if it stops. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Java VM problems
In the last episode (Dec 06), Marc Huffnagle said: > That solved it, thanks. I'll pass this along to the people at > freebsd-java, it seems like something that they should have in their > FAQ. You should have gotten this message when you installed the port: $ cat /usr/ports/java/linux-sun-jdk14/pkg-message == Warning: This JDK is currently unstable. You are advised to use either the FreeBSD jdk14 port or one of the jdk13 ports instead. This Java VM will attempt to obtain some system information by accessing files in linux's procfs. You must install the Linux emulation procfs filesystem for this to work correctly. The JVM will exhibit various problems otherwise. This can be accomplished by adding the following line to your /etc/fstab file: linprocfs /compat/linux/proc linprocfs rw 0 0 and then, as root, executing the commands: kldload linprocfs mount /compat/linux/proc ====== -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Problem with Disk/Partition???
In the last episode (Dec 07), Gerard Samuel said: > Running 4.9, and when running fsck on a partition craps out with -> > hivemind# fsck /files > ** /dev/da0s1h (NO WRITE) > > Any help would be grateful. > Thanks > > hivemind# df -Hi > FilesystemSize Used Avail Capacity iused ifree %iused Mounted on > /dev/da0s1h 808M 243M 500M33% 49133 149777 25% /files Dismount the filesystem first. But if the system let you mount it, you don't need to run fsck anyway. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: avcheck-0.9.tgz
In the last episode (Dec 08), RSB said: > 20031108 Las Vegas Nevada 89102 USA > [EMAIL PROTECTED] > > Subj: avcheck-0.9.tgz > > I ran a WINDOWS Mcafee VirusScan of the FreeBSD install disk (v 4.7) > included w/ the book 'SAMS Teach Yourself FreeBSD in 24 Hours' and it > identified a file on the CD, avcheck-0.9.tgz' as a virus. > > Can you provide any other information regarding this file? VirusScan should also have provided a description of the virus, which would have told you it was simply a test file for virus-scanner products. http://us.mcafee.com/virusInfo/default.asp?id=description&virus_k=98616 -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: a technical how to
In the last episode (Dec 08), Kevin D. Kinsey, DaleCo, S.P. said: > Harald Schmalzbauer wrote: > >On Tuesday 09 December 2003 02:51, homeyra g wrote: > >>So, I hope this is the right address for this type of question. If > >>not would you please forward this and/or let me know the correct > >>address. > >> > >>Thanks, > >> > >>Here is the question: How to truncate a file from the begining to a > >>certain point in the file? If you're writing a script, use the /usr/bin/truncate command. If you're writing a C program, use the truncate() function. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: archiving and restoring disk images...
In the last episode (Dec 09), [EMAIL PROTECTED] said: > I dumped a disk image of my root partition like this: > > dd if=/dev/da0s1a | gzip -9 -c > where.ever.gz > > Does this include the boot sector? In other words, if I need to > restore this disk image I can do the reverse and the boot sector and > root partition will be back to that state. How should I go about > this, if what I'm saying is valid? Can I do this in single user mode > or should I boot from CD and go from there? I haven't needed to do > this yet, but I want to know how. If dumping the root partition like > this is not a good means of backing up, please say so. What would you > recommend? I tar most everything else. The first BSD partition includes the disklabel partition and the boot blocks for that slice, but will not contain the MBR or fdisk partition table. Something like "dd if=/dev/da0 of=mbr count=1" would do that. To restore the disklabel partition you'll need to dd your where.ever dump to /dev/da0s1, since on a new disk there won't be an 'a' partition to write to. After the dd is done, the 'a' partition should appear. I usually either use dump or tar to back up my root partition, and use sysinstall or a bootable CD image to fdisk and disklabel new media before restoring. That way I can change the partition sizes if my new disk is larger/smaller than the original. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Adding SCSI Scanner
In the last episode (Dec 11), Dr. Lyman Hazelton said: > I didn't have my SCSI scanner (UMAX Astra 1200S) powered up when I > loaded FreeBSD 5.1. I turned it on and re-booted the system, and it > certainly sees it (as seen in the dmesg report), but it simply says > it's part of "pass4:"... in other words, no driver associated with > it. Is there some nice way of getting the system to use it, other > than reloading the entire system from scratch? The system has been > up a few days and I've loaded a lot of other good stuff and got it > running just the way I want it to, so reloading and reconfiguring all > of it would be a lotta work. pass4 should be all the SANE port needs, actually. There are no kernel drivers for scanners. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: dd of mounted filesystem
In the last episode (Dec 11), Matthew Seaman said: > On Thu, Dec 11, 2003 at 02:54:12PM -0500, Dru wrote: > > Can anyone describe or point me to resources explaining why it is > > dangerous to dd a filesystem while it is mounted? Is it still > > considered to be dangerous if the system is first dropped down to > > single-user mode? > > Remember that dd(1) traverses the block device sequentially, but that > most FS accesses are random, so any particular change can span either > side of dd(1)'s offset. Also that dd'ing from the block device > bypasses the usual machinery for doing file IO -- machinery that is > designed under the premise that it will have sole control over what > gets read or written where and when. On current you can get around the consistency problem by dd'ing a snapshot of the filesystem, just like dump's -L flag does. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: dd of mounted filesystem
In the last episode (Dec 12), Dru said: > On Thu, 11 Dec 2003, Dan Nelson wrote: > > In the last episode (Dec 11), Matthew Seaman said: > > > Remember that dd(1) traverses the block device sequentially, but > > > that most FS accesses are random, so any particular change can > > > span either side of dd(1)'s offset. Also that dd'ing from the > > > block device bypasses the usual machinery for doing file IO -- > > > machinery that is designed under the premise that it will have > > > sole control over what gets read or written where and when. > > > > On current you can get around the consistency problem by dd'ing a > > snapshot of the filesystem, just like dump's -L flag does. > > You mean, run "makesnap_ffs" first? I've been meaning to play with > that one, I'll have to try it out. I don't think that's a standard FreeBSD command; what I was thinking of was something like: mount -u -o snapshot /usr/.snap/snap1 /usr dd if=/usr/.snap/snap1 of=blah bs=64k rm /usr/.snap/snap1 -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: swap space
In the last episode (Dec 12), M.D. DeWar said: > Thanks, Read the man. so heres a print out. some makes sense. others > . don't like what is up referring to ? > still tho i don't see what is eating up 373M of swap space ? well, mysqld is probably consuming most of it, since it's almost completely swapped out (319MB process size, but only 3M in memory). Those two perl processes are the real problem, though. Are you sure that's mrtg? My mrtg processes don't take more than 9MB when they're running. > Last pid: 80656; load averages: 3.21, 3.23, 3.16 > 59 processes: 4 running, 55 sleeping > CPU states: 75.9% user, 0.0% nice, 19.0% system, 5.2% interrupt, 0.0% idle > Mem: 179M Active, 19M Inact, 38M Wired, 8212K Cache, 35M Buf, 1744K Free > Swap: 512M Total, 373M Used, 139M Free, 72% Inuse, 424K In, 316K Out > > PID USERNAME PRI NICE SIZERES STATETIME WCPUCPU COMMAND > 342 mysql 2 0 318M 3092K poll21:49 0.05% 0.05% mysqld > 13927 root -22 0 173M 91564K swread 32:31 14.01% 14.01% perl > 73396 root 51 0 151M 80728K RUN 13:46 23.73% 23.73% perl > 39818 snortman4 0 43844K 2940K bpf 3:06 0.00% 0.00% snort > 39189 nobody 18 0 24216K 0K lockf0:25 0.00% 0.00% httpd -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: tar, mt backup BLOCKSIZES, variable & fixed question Seagate SDT224000N DDS-3 DAT
In the last episode (Dec 12), Jonathan Wright said: > up" if there is a delay in the host computer's data access. However, > I am getting inconsistent information on the variable blocksize > setting in mt and don't know if I should use it or not. One site I > found said that variable blocksizes allow the drive to write blocks > at a size that it thinks will be best for the data speed at hand, > while another site said that it allows the drive to use the blocksize > being sent to it from the backup application (making 10,240 byte > blocks in the above tar example). Who is right, or have I got this > all mess up? :) variable blocksize means write data to tape using whatever size chunks the user write()s. I think only ancient QIC tapes actually need you to set a fixed blocksize via mt. > Also is there a preferred blocksize for DDS-3 DAT drives? I use 64k for everything. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: How to find out which drive isdown?
In the last episode (Dec 14), Joachim Dagerot said: > I have seven disks in my cabinett, three of them builds a vinum > RAID-5 plex. > > One of the disks in the RAID is down according to "vinum list". > > I could use some honts on how to determine which physical disk is > named what in my system, is there a good path to follow by disconnect > one, write down whats left and then re-connect it and connect another > one? "Vunum ld" will list the disks and their status. If you're not sure which physical disk is "da1", you'll probably have to look at the jumper settings on each drive. Another option is to run something like "camcontrol stop da1; camcontrol start da1" or "dd if=/dev/da1 of=/dev/null", and see which disk's activity LED lights up. http://www.vinumvm.org/vinum/replacing-drive.html has some instructions on swapping out a failed drive. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: NIS authontication problem.
In the last episode (Dec 14), Hossein said: > In our department we are going to use a 5.1 Stable FreeBSD, and > it must run NIS client to authonticate the users through a Linux NIS > server. > The ypbind works well and when I do "ypcat passwd" I get the > enteries in the passwd of the NIS server. I added the correct lines > to passwd.master and group according to the handbook. But no user can > log in and in the /var/log/auth.log it apears that the password is > not corect. If you edited passwd.master directly, you may need to run "pwd_mkdb" to rebuild the spwd.db and pwd.db database files that the system uses. The "vipw" command does this automatically. Does "finger somenisuser" on the client print the right info? -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: ID says no such user, PW says already exists ?!?!
In the last episode (Dec 15), Rishi Chopra said: > This is gonna sound crazy, but: I got my FreeBSD box up and running > and decided to add a user account for myself (avoid the perils of > working as root?) I added a named account 'rchopra' and everything > was honky-dorey. > > The following series of events occured and have left me confused and > unable to use or re-add the 'rchopra' account: > > 1) I manually added myself to the 'wheel' account, in order to allow > 'su' to root. Rather than using 'pw', I was under the impression > just editing the entry in '/etc/group' would work; my entry looked > like this: > > wheel:*:0:root:rchopra The userlist is comma-delimited: wheel:*:0:root,rchopra > 2) This action seemed to cause a problem; I attempted to login and su, > and I was not able to login successfully (my password would not work). > I attempted to run 'id rchopra' at the command prompt, which returns an > error: > > # id rchopra > 'id: rchopra: no such user' > > Strangely enough, the '/home/rchopra' directory still exists. > > 3) OK, no problem right? Just run 'adduser' and create the account > again... Unfortunately, after entering all the information for the user > account 'rchopra' the system says: > > pw: user 'rchopra' already exists > adduser: ERROR: There was an error adding user (rchopra) That's a little weird. Is there an 'rchopra' user when you run vipw? Messing with /etc/group shouldn't have affected your ability to log in. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: CPU type
In the last episode (Dec 16), flux said:
> How do I know what type of CPU (Model, Mhz, etc...) do I have
> without rebooting my FreeBSD box? Is there any command to show
> this kind of information?
> Thank you.
/var/run/dmesg.boot should hold your last bootup log, which will have
CPU info in it. Or you could install and run cpuid or x86info from
ports.
--
Dan Nelson
[EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: process accounting
In the last episode (Dec 16), fbsd_user said: > In the /stand/sysinstall/ under post configure is see option for > process accounting. > What is this? > How can I find info about what this is and how to use it? Each process that gets run will be logged into /var/account/acct. You can use the 'lastcomm' command to list the records, and the 'sa' command to summarize them. See the lastcomm and sa manpages for more info. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Increase space for /var/db/
In the last episode (Dec 17), samy lancher said: > Hello, I have a 4.5 FreeBSD server with mysql database. only 250MB > was allocated to /var/db. Our database is increasing so fast that now > i need to expand the memory for /var/db. I guess one way to do is to > install a new hard drive and then mount it to /var/db. But i would > like to know if there is any other way to solve that problem. Is > making a soft link a good idea. i have 70GB in /usr/ so i would like > to map /var/db/ to some location in /usr/. Please let me know if > anyone has any suggestions. That's what I do. $ ls -l /var/db/pkg lrwxr-xr-x 1 root wheel 20 Apr 14 2003 /var/db/pkg@ -> ../../usr/var_db_pkg -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Increase space for /var/db/
In the last episode (Dec 17), samy lancher said:
> thanks for the quick response.
>
> Is this a soft link?, what happens to the existing data in /var/db. I
> do not want to loose that data.
I did this:
$ cd /var/db
$ mv pkg /usr/var_db_pkg
$ ln -s ../../usr/var_db_pkg pkg
--
Dan Nelson
[EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Running dual CPUs
In the last episode (Dec 24), Gerard Samuel said: > I acquired a pair of Slot 1 CPUs that is supposedly a matched pair. > I've heard, that they must be of the same stepping, and Im not sure > of what Im looking for. CPU specs are at -> > http://processorfinder.intel.com/scripts/ > details.asp?sSpec=SL3XL&ProcFam=25&PkgType=ALL&SysBusSpd=ALL&CorSpd=5775 > Just trying to be careful before I fry the thing, so any pointers are > welcome. Thanks Most of the time, stepping doesn't matter. You can look at Intel's processor errata pages to see exactly what got fixed in each stepping, and determine whether the bugs would interfere with SMP or not. I've run dual-CPU systems with different steppings with no problems. I haven't heard of anyone damaging their CPUs or motherboard with mismatched CPUs. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: group members can't modify files even though 775 set
In the last episode (Dec 24), Uwe Doering said: > Seamus Abshere wrote: > >But this happens when I am logged in as sabshere: > > > >>$echo "whatev" > index.html > >>-bash: index.html: cannot overwrite existing file > > I think this has nothing to do with file permissions. Instead, the > 'noclobber' option is probably enabled in your bash settings, so the > shell refuses to overwrite existing files. Find the directive 'set -o > noclobber' in your shell's startup files and comment it out, or if this > behaviour is the default, try to override it with 'set +o noclobber'. Or use >! , which overrides noclobber for just that one time. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: mkisofs options question
In the last episode (Dec 24), Charles Howse said: > I have some directories that I need to include in a .iso file. I've > followed the online Handbook section 12.5.2, and read man mkisofs. > > The directories each have the file CUSTOM in them. > That causes mkisofs to error: > > [EMAIL PROTECTED] ~]# mkisofs -U -R -o /tmp/cdimg.iso /disk2/curly /disk2/larry > Warning: creating filesystem that does not conform to ISO-9660. > Using CUSTOM000 for /CUSTOM (CUSTOM) > mkisofs: Error: '/disk2/larry/CUSTOM' and '/disk2/curly/CUSTOM' have the same Rock > Ridge name 'CUSTOM'. > mkisofs: Error: '/disk2/larry/CUSTOM' and '/disk2/curly/CUSTOM' have the same Rock > Ridge name 'CUSTOM'. > mkisofs: Unable to sort directory > > What are the proper options for mkisofs so that I can have identical > filenames in different directories? They are in the same directory. Remember that when you pass mkisofs a list of directories, their contents are combined to create the CD. you can't have duplicates, though. /disk2/larry/CUSTOM and /disk2/curly/CUSTOM both map to "/CUSTOM" on the cdrom. If you really wanted the cdrom to contain two subdirectories named "/larry" and "/curly", try passing just "/disk2" to mkisofs, or if there are other directories in disk2, create a /disk2/cdrom, move larry and curly into there, and run "mkisofs /disk2/cdrom". -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: ghostview missing libc.so.4?
In the last episode (Dec 27), Chip Wiegand said: > Mine has libc.so.5. No wonder the app is failing, it specifically > wants .4 and is too dumb to work with anything newer. I'm getting > real frustrated with this again, I'd like to move away from MS but > it's not easy to do when the apps I need don't even work > properly. Somehow you managed to install a FreeBSD 4 package onto a FreeBSD 5 system. pkg_delete it, then run "pkg_add -r ghostview" to fetch the correct package from ftp.freebsd.org, or cd into /usr/ports/print/ghostview and run "make install" if you want to build it yourself. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Is it true?...
In the last episode (Dec 30), Nigel Weeks said: > If I have 'MATH_EMULATE' in my kernel, will it: > > A. Detect that the processor has a hardware FPU, and use that instead > B. Use the software one, ignoring the hardware FPU? > > (Hopefully the latter - a performance jump will be nice...) The former. Hardware FPU is much faster than software emulation. MATH_EMULATE is only necessary for CPUs with no hardware FP at all (386sx or possibly embedded 486-class CPUs). -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: file system
In the last episode (Dec 30), Larry Johnson said: > The file systems used in Linux are limited to 32,000 files or > subdirectories within any given directory. Does anybody know what the > limit is in FreeBSD? Can't find any info about this anywhere. I don't think there's really a limit on the number of files; I quite easily created 90k files in a directory with a shell script. The number of subdirectories in one directory is 32766, since the link count for a directory is stored as a signed 16-bit integer, and the parent directory will end up with 32767 links (one for itself, and 32766 to .. in each child directory). This is also trivial to verify with a shell script. -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Tape backup solution? [OT]
In the last episode (Dec 31), Eric F Crist said: > I have a question that's slightly off-topic, but not. I install > high-end surveillance equipment for CCTV and such. I have a rather > large client in Minneapolis who's using Dedicated Micros digital > video recorders. The particular model we're using has a 500 GB hdd, > but this client would like to archive images to tape for longer > storage. As of now, we're only getting about 2 months of recording > time. For off-site viewing, this unit can off-load images to a SCSI > cd recorder. Does anyone suggest a tape backup device that would be > SCSI and external, with a fairly high-capacity? I'm thinking around > 50 GB? I can't find a good web page to refer you to, but here's a quick summary of what's available. Capacity and transfer rate are native; if your data is 2:1 compressible, double both columns. Drive Capacity Xfer rate (GB) (MB/Sec) DLT 406 sDLT110-300 11-36 LTO 100 15 LTO2200 30 AIT3100 12 SAIT1 500 30 -- Dan Nelson [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: svgalib-1.4.3_5 is only for i386, while you are running amd64.
In the last episode (Sep 11), Dino Vliet said:
> Hi folks,
>
> On my adm64 dualcore system running freebsd 7 I want to install
> /usr/ports/multmedia/kino. However the port stops because of this:
>
> ==> Verifying install for gs in /usr/ports/print/ghostscript8
> NOTE: Default paper size will be A4.
> ===> Found saved configuration for ghostscript8-8.62_5
> ===> Extracting for ghostscript8-8.62_5
> => MD5 Checksum OK for ghostscript/ghostscript-8.62.tar.bz2.
> => SHA256 Checksum OK for ghostscript/ghostscript-8.62.tar.bz2.
> => MD5 Checksum OK for ghostscript/epag-3.09.tar.gz.
> => SHA256 Checksum OK for ghostscript/epag-3.09.tar.gz.
> /usr/bin/tar -C /usr/ports/print/ghostscript8/work/ghostscript-8.62 -xzf
> /usr/ports/distfiles/ghostscript/epag-3.09.tar.gz
> /bin/ln -s -f epag-3.09
> /usr/ports/print/ghostscript8/work/ghostscript-8.62/epag
> install -o root -g wheel -m 444
> /usr/ports/print/ghostscript8/files/epag.contrib.mak
> /usr/ports/print/ghostscript8/work/ghostscript-8.62/src/epag.contrib.mak
> ===> Patching for ghostscript8-8.62_5
> ===> Applying FreeBSD patches for ghostscript8-8.62_5
> /usr/bin/sed -i.bak -e 's|%%DATADIR%%|/usr/local/share/ghostscript|g ;
> s|%%GS_VERSION%%|8.62|g'
> /usr/ports/print/ghostscript8/work/ghostscript-8.62/contrib/japanese/dmp_site.ps
> ===> ghostscript8-8.62_5 depends on executable: gmake - found
> ===> ghostscript8-8.62_5 depends on file:
> /usr/local/libdata/pkgconfig/xt.pc - found
> ===> ghostscript8-8.62_5 depends on file:
> /usr/local/libdata/pkgconfig/xext.pc - found
> ===> ghostscript8-8.62_5 depends on shared library: jpeg.9 - found
> ===> ghostscript8-8.62_5 depends on shared library: png.5 - found
> ===> ghostscript8-8.62_5 depends on shared library: vga.1 - not found
> ===> Verifying install for vga.1 in /usr/ports/graphics/svgalib
> ===> svgalib-1.4.3_5 is only for i386, while you are running amd64.
Try cd'ing to /usr/ports/print/ghostscript8, then run "make config" and
make sure the two SVGALIB options are unselected. They should be about
15 lines from the top:
[ ] GS_lvga256 D: SVGAlib, 256-color VGA modes
[ ] GS_vgalibD: SVGAlib, 16-color VGA modes
Then rebuild ghostscript and you should be okay.
--
Dan Nelson
[EMAIL PROTECTED]
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: InstallJammer on Fbsd 7.0 ?
In the last episode (Sep 12), Len Conrad said: > >> Anybody got this running on FreeBSD 7.0? > >> > >> http://www.installjammer.com/ > > > >Installing misc/compat6x would be a good start. It would get you > >that library, at least. > > > >To be honest, I wouldn't trust an install package that I didn't have > >the source for... > > the source is here, I'm trying with 1.2.6 > > http://downloads.installjammer.com/installkit/src/installkit-1.2.6.tar.gz > > ./configure > gmake > > fails with > > rm -f libcraplzma0.1.so > ld -Bshareable -x -o libcraplzma0.1.so crapLzma.o CRC.o Alloc.o String.o > StringConvert.o OutBuffer.o > FileStreams.o StreamUtils.o LZInWindow.o LZMAEncoder.o RangeCoderBit.o > C_FileIO.o -lsupc++ -static-libgcc > -L/usr/local/src/installkitsrc/installkit-1.2.6/FreeBSD-7-x86/lib > -L/usr/local/src/installkitsrc/installkit-1.2.6/FreeBSD-7-x86/lib -ltclstub8.4 > ld: unrecognized -a option `tic-libgcc' > *** Error code 1 FreeBSD only has a static libgcc, which explains why -static-libgcc is not recognized. You can probably edit the Makefile and remove the flag. > Stop in > /usr/local/src/installkitsrc/installkit-1.2.6/FreeBSD-7-x86/build/craplzma. > rm -f libcraplzma0.1.so > ld -Bshareable -x -o libcraplzma0.1.so crapLzma.o CRC.o Alloc.o String.o > StringConvert.o OutBuffer.o FileStreams.o StreamUtils.o LZInWindow.o > LZMAEncoder.o RangeCoderBit.o C_FileIO.o -lsupc++ -static-libgcc > -L/usr/local/src/installkitsrc/installkit-1.2.6/FreeBSD-7-x86/lib > -L/usr/local/src/installkitsrc/installkit-1.2.6/FreeBSD-7-x86/lib -ltclstub8.4 > ld: unrecognized -a option `tic-libgcc' > *** Error code 1 > > Stop in > /usr/local/src/installkitsrc/installkit-1.2.6/FreeBSD-7-x86/build/craplzma. > gmake: *** [craplzma] Error 1 > > advice was "Makefile, and in the TARGETS list, remove the "craplzma"", > > done but: > > Install miniarc.tcl > /usr/local/src/installkitsrc/installkit-1.2.6/FreeBSD-7-x86/lib/miniarc0.1/miniarc.tcl > Install pkgIndex.tcl > /usr/local/src/installkitsrc/installkit-1.2.6/FreeBSD-7-x86/lib/miniarc0.1 > Installing header files in > /usr/local/src/installkitsrc/installkit-1.2.6/FreeBSD-7-x86/include > "Makefile.shared", line 2: Need an operator > "Makefile.shared", line 3: Need an operator > "Makefile.shared", line 4: Need an operator > "Makefile.shared", line 59: Need an operator > make: fatal errors encountered -- cannot continue > "Makefile.shared", line 2: Need an operator > "Makefile.shared", line 3: Need an operator > "Makefile.shared", line 4: Need an operator > "Makefile.shared", line 59: Need an operator > make: fatal errors encountered -- cannot continue Sounds like the Makefiles require gmake, and you ran plain make instead of gmake this time. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: CVS log
In the last episode (Sep 14), Walker said:
> Is there a CVS log that is web accessible and allows me to search for
> all changes between two releases (for example, version 7.0 and the
> upcoming 7.1)?
http://ftp.cz.freebsd.org/pub/FreeBSD-cvs/main/CVSROOT/commitlogs/ (and
some of the other regional FreeBSD sites) has all the commit entries.
7.0 was released in late February, so if you look at all the files
between then and now, only looking at things committed to the RELENG_7
branch, you'll have your changes.
If you install the subversion port, the command
svn log -v -r '{2008-2-27}:HEAD' svn://svn.freebsd.org/base/stable/7/
will print all commits to the RELENG_7 branch between then and now.
--
Dan Nelson
[EMAIL PROTECTED]
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: PAM_AUTHTOK_RECOVER_ERR
In the last episode (Sep 19), Berk Gulenler said: > I'm trying to install pam_ldap 1.84 to FreeBSD version 6.3 AMD64. But > I'm getting this error message from gmake compiler. I think the > problem is an undeclared function. About this any help would be > appreciated. > > Related log: > > gcc -DHAVE_CONFIG_H -I/usr/local/include -DLDAP_REFERRALS -DLDAP_DEPRECATED > -DPIC -D_REENTRANT -g -O2 -Wall -fPIC -c pam_ldap.c > pam_ldap.c:3569: error: `PAM_AUTHTOK_RECOVER_ERR' undeclared (first use in > this function) > pam_ldap.c:3569: error: (Each undeclared identifier is reported only once > pam_ldap.c:3569: error: for each function it appears in.) Are you building the security/pam_ldap port? That should build with no errors. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: PAM_AUTHTOK_RECOVER_ERR
In the last episode (Sep 19), Berk Gulenler said: > Dan Nelson wrote: > > In the last episode (Sep 19), Berk Gulenler said: >> I'm trying to install pam_ldap 1.84 to FreeBSD version 6.3 AMD64. But >> I'm getting this error message from gmake compiler. I think the >> problem is an undeclared function. About this any help would be >> appreciated. >> >> Related log: >> >> gcc -DHAVE_CONFIG_H -I/usr/local/include -DLDAP_REFERRALS >> -DLDAP_DEPRECATED -DPIC -D_REENTRANT -g -O2 -Wall -fPIC -c pam_ldap.c >> pam_ldap.c:3569: error: `PAM_AUTHTOK_RECOVER_ERR' undeclared (first use in >> this function) >> pam_ldap.c:3569: error: (Each undeclared identifier is reported only once >> pam_ldap.c:3569: error: for each function it appears in.) >> >> Are you building the security/pam_ldap port? That should build with no >> errors. > >No I'm not trying to build it from port tree because the ports >says my openldap version is not compatible with pam_ldap (a new >version is required). But I have to use an older version of >openldap so, I'm trying to compile pam_ldap for my self. Thanks. Try editing /etc/make.conf and put WANT_OPENLDAP_VER=23 and see if the port builds. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: FreeBSD Port of mrtg Not Happy when Trying to Start
In the last episode (Sep 23), Martin McCormick said: > When installing mrtg, I left the ipv6 box unchecked as we have none, > and turned on the snmpV3 support which made no difference. Still the > same error: > > Can't locate SNMP_util.pm in @INC (@INC contains: /usr/local/bin/../lib/mrtg2 > /usr/local/bin /usr/local/lib/perl5/5.8.8/BSDPAN > /usr/local/lib/perl5/site_perl/5.8.8/mach > /usr/local/lib/perl5/site_perl/5.8.8 /usr/local/lib/perl5/site_perl > /usr/local/lib/perl5/5.8.8/mach /usr/local/lib/perl5/5.8.8 .) at > /usr/local/bin/mrtg line 292. That file is supposed to be installed by the p5-SNMP_Session port, which is marked as a dependency by mrtg. Do you maybe have an old or bad install of that port? $ locate /SNMP_util.pm /usr/local/lib/perl5/site_perl/5.8.8/SNMP_util.pm $ pkg_which /usr/local/lib/perl5/site_perl/5.8.8/SNMP_util.pm p5-SNMP_Session-1.12 $ pkg_info -rR mrtg-2.16.2,1 Information for mrtg-2.16.2,1: Depends on: Dependency: perl-5.8.8_1 Dependency: p5-SNMP_Session-1.12 Dependency: png-1.2.32 Dependency: jpeg-6b_7 Dependency: pkg-config-0.23_1 Dependency: freetype2-2.3.7 Dependency: gd-2.0.35,1 Dependency: p5-PathTools-3.2701 Dependency: p5-Pod-Parser-1.35_2 $ pkg_info -L p5-SNMP_Session-1.12 Information for p5-SNMP_Session-1.12: Files: /usr/local/lib/perl5/site_perl/5.8.8/BER.pm /usr/local/lib/perl5/site_perl/5.8.8/SNMP_Session.pm /usr/local/lib/perl5/site_perl/5.8.8/SNMP_util.pm /usr/local/lib/perl5/site_perl/5.8.8/mach/auto/SNMP_Session/.packlist /usr/local/share/doc/SNMP_Session/README /usr/local/share/doc/SNMP_Session/README.SNMP_util /usr/local/share/doc/SNMP_Session/index.html -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Netprint perl script from Handbook doesn't work
In the last episode (Sep 24), Andy Kosela said:
> The netprint perl script provided in the Handbook (9.4.3.2) is not
> working.. or am I missing something:
>
> plotinus:~> cat new.txt | lp.sh
> Can't contact 10.10.21.12: Address family not supported by protocol family at
> /usr/local/libexec/netprint line 21.
Can you telnet to that ip address ("telnet 10.10.21.12 9100", or
whatever port you're using)?
> plotinus:> cat /usr/local/libexec/netprint
> #!/usr/bin/perl
> #
> # netprint - Text filter for printer attached to network
> # Installed in /usr/local/libexec/netprint
> #
> $#ARGV eq 1 || die "Usage: $0 ";
>
> $printer_host = $ARGV[0];
> $printer_port = $ARGV[1];
>
> require 'sys/socket.ph';
>
> ($ignore, $ignore, $protocol) = getprotobyname('tcp');
> ($ignore, $ignore, $ignore, $ignore, $address)
>= gethostbyname($printer_host);
>
> $sockaddr = pack('S n a4 x8', &AF_INET, $printer_port, $address);
>
> socket(PRINTER, &PF_INET, &SOCK_STREAM, $protocol)
>|| die "Can't create TCP/IP stream socket: $!";
> connect(PRINTER, $sockaddr) || die "Can't contact $printer_host: $!";
> while () { print PRINTER; }
> exit 0;
Wow. That's a really complicated way to say
#! /bin/sh
nc $1 $2
--
Dan Nelson
[EMAIL PROTECTED]
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: AW: ethernet statistics
In the last episode (Sep 25), Vonarburg, David said: > netstat -i is ok for a user to get system statictics. I'd like get > exactly this information but as a function call inside a "C" language > application netstat -i still digs into kernel memory to get those stats, I think, so you can't directly access those numbers as a regular user. You can shell out and parse the output of "netstat -i", or do some snmp queries to your local net-snmp daemon. Checking the dev.em.0.stats sysctl node is another option, and gives you some more hardware status counters than netstat, but not all drivers support it (em does so you're okay). -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: neon28 fails, conflicts with neon26
In the last episode (Sep 27), Eduardo Cerejo said: > FreeBSD 7 stable > > ports/UPDATING mentions this: > > subversion now uses neon-0.28.x (www/neon28) port, and automatic > portupgrade will fail because www/neon28 will conflict with installed > www/neon26 port. > > You should upgrade neon library before subversion with the command: > > # portupgrade -o www/neon28 neon26 > > I did exactly that and It still fails: > > ** Command failed [exit code 1]: /usr/bin/script -qa /tmp/portupgrade.44225.0 > env UPGRADE_TOOL=portupgrade UPGRADE_PORT=neon26-0.26.4_1 > UPGRADE_PORT_VER=0.26.4_1 make reinstall > ---> Restoring the old version > pkg_add: package 'neon26-0.26.4_1' conflicts with neon28-0.28.2_1 > pkg_add: -f specified; proceeding anyway > ** Fix the installation problem and try again. > [Updating the pkgdb in /var/db/pkg ... - 1002 packages > found (-0 +1) . done] > ** Listing the failed packages (-:ignored / *:skipped / !:failed) > ! www/neon28 (neon26-0.26.4_1) (install error) The last line you pasted indicates an install error of some sort, but you didn't list the original error. It'll be above the "** Command failed [exit code 1]: ..." line, possibly some lines up if the neon build requires gmake and recursed into a bunch of subdirectories before failing. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: sshguard upgrade error
In the last episode (Sep 27), Tim Kellers said:
> I have autoconf 2.62 installed. Is there a workaround?
>
> > ===> Building for sshguard-1.2
> > Making all in src
> > cd .. && /bin/sh
> > /usr/ports/security/sshguard/work/sshguard-1.2/missing --run autoheader
> > aclocal.m4:14: error: this file was generated for autoconf 2.61.
> > You have another version of autoconf. If you want to use that,
> > you should regenerate the build system entirely.
> > aclocal.m4:14: the top level
> > autom4te-2.62: /usr/local/bin/gm4 failed with exit status: 63
> > autoheader-2.62: '/usr/local/bin/autom4te-2.62' failed with exit
> > status: 63
> > *** Error code 1
It looks like the original author didn't prepare their source tree
correctly before creating the tarball (end-users should never need to
run autoheader), and the port maintainer didn't specify any autotools
dependencies to compensate.
Try this patch:
diff -u -r1.10 Makefile
--- Makefile23 Sep 2008 18:58:35 - 1.10
+++ Makefile27 Sep 2008 17:13:38 -
@@ -21,6 +21,7 @@
MANCOMPRESSED= no
USE_BZIP2= yes
MAKE_ARGS+=ACLOCAL="${TRUE}" AUTOCONF="${TRUE}" AUTOMAKE="${TRUE}"
+USE_AUTOTOOLS= aclocal:110 autoheader:262
HAS_CONFIGURE= yes
# sublist will be set afterward
--
Dan Nelson
[EMAIL PROTECTED]
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Definition of off64_t
In the last episode (Sep 30), hibablu said: > I am trying to port an application written on Linux to FreeBSD. > During compile, I am getting an error saying that off64_t is not > defined. Which header file do I need to include to get the definition > for off64_t ? There is no need for an off64_t on FreeBSD. The program should use off_t instead, and on Linux, they should add the compiler flags "-D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64" so that off_t is 64 bits on Linux as well. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Process memory inspection
In the last episode (Oct 01), Ivan Voras said: > Hi, > > The top utility has SIZE and RES, but doesn't have what part of SIZE > is sysv shared memory. Is there something that can print out in > detail how a process uses / allocates its memory (I'm specifically > interested in sysvshm but there's also the stack & mmap)? You can get detailed process memory info from /proc//map , or in 7.1 and later, "procstat -v". I don't know how easy it is to identify which block is shared memory, though. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: More vbox and build troubles.
In the last episode (Oct 02), Desmond Chapman said:
>
> * Checking libpng *
> compiling the following source file:
> #include
> #include
> extern "C" int main(void)
> {
> printf("found version %s", PNG_LIBPNG_VER_STRING);
> #if PNG_LIBPNG_VER >= 10205
> printf(", OK.\n");
> return 0;
> #else
> printf(", expected version 1.2.5 or higher\n");
> return 1;
> #endif
> }
> using the following command line:
> g++ -O -Wall -o .tmp_out .tmp_src.cc "-lpng "
> .tmp_src.cc:2:17: error: png.h: No such file or directory
> .tmp_src.cc: In function 'int main()':
> .tmp_src.cc:5: error: 'PNG_LIBPNG_VER_STRING' was not declared in this scope
>
>
> I need the headers for this library- libpng-1.2.32. Are they in ports
> as part of another package or do I need to build them? If so, how?
Assuming you have the png port installed, add -I/usr/local/include
-L/usr/local/lib to your compile line.
--
Dan Nelson
[EMAIL PROTECTED]
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: How To Get libm.so.4?
In the last episode (Oct 09), Drew Tomlinson said: > I am attempting to install a web analysis tool named "Urchin" from > Google. Installation instructions are here: > > https://secure.urchin.com/helpwiki/en/Urchin_Installation_Guide_(FreeBSD_and_Linux) > > Urchin claims to run on FBSD 6.2+ which I took to mean version 6.2 or > greater. Since this is a brand new install, I installed FBSD > 7.1-PRERELEASE, assuming the actual release would not be too far off. > > I'm following the install procedures which has me run a ./install.sh > script. This script fails, complaining about a missing libm.so.4. > I've Googled and found some reference that this has to do with > installing compatibility libraries for FBSD 4. Thus I've added > "COMPAT4X= yes" and rebuilt and installed my world. However I still > do not have this file. Major FreeBSD version bumps have historically resulted in most of the base system's shared library versions getting bumped as well. libc.so.4 would have been for FreeBSD 4, but libm doesn't get bumped as often. libm.so.4 corresponds to FreeBSD 6, and you can get it by installing the misc/compat6x port. FreeBSD 7 and later use symbol versioning (which allows different versions of the the same function to exist in a single library), so library version bumps theoretically are a thing of the past. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: FreeBSD RELENG_7 scsi and usb, usb disks take precedence
In the last episode (Oct 16), Eric Masson said: > "Svein Skogen (List Mail Account)" <[EMAIL PROTECTED]> writes: > > With all due respect, if sysinstall isn't able to write those labels, > > this is hardly the solution to the problem... ;) > > Sysinstall, even if it has proven to be useful, is outdated in many > respects, it doesn't cope really well with GEOM or ZFS for example. > > Using GEOM labels ensure that fstab will always be consistent, so > *this is* a solution. > > Now, you can try to patch the kernel to make it probe devices in a > predefined & sorted order, but I guess it will be much more difficult > ;) No patching needed. You can wire down the unit number of your scsi bus and drive with boot hints as described in the scsi(4) manpage. Wire the adaptec card down as scbus0, and wire the device at scbus0.0 down as da0. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Why does adding /usr/lib32 to LD_LIBRARY_PATH break 64-bit binaries?
In the last episode (Oct 23), Alexander Sack said: > On Thu, Oct 23, 2008 at 10:09 PM, Alexander Kabaev <[EMAIL PROTECTED]> wrote: > > LD_LIBRARY_PATH is for native 64bit rtld. If you want a specific > > path added for use by 32-bit ld-elf.so.1 only, use > > LD_32_LIBRARY_PATH. > > > > Said that, your problem is likely caused by the fact that there is > > no /lib32, only /usr/lib32. So if 64-bit library lives in /lib, > > your LD_LIBRARY_PATH will cause loader to find its 32-bit > > equivalent in /usr/lib32 first. > > > > Try LD_LIBRARY_PATH=/lib:/usr/lib:/usr/lib32:/usr/lib64 for better > > results. > > Yes I figured that out on my own but my question still exists, why > isn't /usr/lib similar in format to /usr/lib32 though with respect to > major numbers? Ever since the switch from static to dynamic-linked /bin and /sbin, some shared libraries are needed during the boot process. Those libraries live in /lib, and since there are no 32-bit binaries required to boot a 64-bit system, there is no need for a /lib32. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: MySQL not restarted after portupgrade
In the last episode (Nov 03), Andrew Berry said:
> For some reason, after running portupgrade databases/mysql51-server,
> MySQL is brought down, but not restarted. It's not a huge issue, as I
> just have to remember to run the rc.d script after upgrading, but I
> was wondering if anyone else had seen this? portupgrade seems to work
> fine with other daemons such as Apache and Courier IMAP.
It's in the pkg-list. The first line is "@stopdaemon mysql-server", so
any time you upgrade the port, it kills mysql. Enough ports do this,
and it's so annoying, that I recommend just disabling it completely:
--- bsd.port.mk 5 Sep 2008 19:41:43 - 1.604
+++ bsd.port.mk 8 Oct 2008 06:21:55 -
@@ -1588,10 +1588,9 @@
DATADIR=${DATADIR} DOCSDIR=${DOCSDIR}
EXAMPLESDIR=${EXAMPLESDIR} \
WWWDIR=${WWWDIR} ETCDIR=${ETCDIR}
-PLIST_REINPLACE+= dirrmtry stopdaemon rmtry
+PLIST_REINPLACE+= dirrmtry rmtry
[EMAIL PROTECTED] \(.*\)[EMAIL PROTECTED] rmdir %D/\1 2>/dev/null || true!
[EMAIL PROTECTED] \(.*\)[EMAIL PROTECTED] rm -f %D/\1 2>/dev/null || true!
[EMAIL PROTECTED] \(.*\)[EMAIL PROTECTED] %D/etc/rc.d/\1${RC_SUBR_SUFFIX}
forcestop 2>/dev/null || true!
# kludge to strip trailing whitespace from CFLAGS;
# sub-configure will not # survive double space
Some ports run the rc.d script themselves on uninstall (sysutils/hal,
for exmaple), so you have to edit the pkg-plist file and remove the
unexec line from those manually.
--
Dan Nelson
[EMAIL PROTECTED]
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: mtree
In the last episode (Nov 10), Garcia, Tony said: > Another developer received approval to test mtree for our project. > He has since left and no one knows anything about this application. > We are looking at mtree as a way to provide auditing of machines for > permissions, ownership and date changes as well as performing cksum > on each file. Is there any way you can point me to documentation > that gives me a high and low level of what mtree can do. I've tried > compiling the version that was downloaded, but it fails because it > needs other files which are not present (like .h files). > > I'd appreciate any help you can provide. The google returns are far > too numerous to make heads or tails from. I also have checked the > freebsd info but I can't find any documentation. Thank you. Manpage: http://www.freebsd.org/cgi/man.cgi?query=mtree Here's a blog entry that explains how to use it as a file verification tool: http://blogs.techrepublic.com.com/security/?p=283 "Use mtree for filesystem integrity auditing" -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Help needed: Which gcc version supports xmemalign
In the last episode (Nov 18), Vijayalakshmi BN said: > I'm working on Solaris 9 and using gcc version 2.9.5. I'm facing > issues with respect to memory misalignment. I'm getting SIGBUS > errors. Through various websites, I found that xmemalign can be used > while compiling to handle memory alignment issues. Bu the version I'm > using doesn't support xmemalign option. Even gcc version 3.4.3 > doesn't support. It says: gcc: language memalign=1i not recognized > Can you let me know which version of gcc supports xmemalign option on > solaris? Also where can I download the gcc version from? -xmemalign=ab is a Sun Studio compiler flag, not a GCC flag. If you don't already have it, you can download it free from http://developers.sun.com/sunstudio/ . If for some reason you must use a GCC frontend, you can try GCC for Sparc Systems ( http://cooltools.sunsource.net/gcc/ ), which supports all the regular GCC flags plus many Sun Studio ones, including xmemalign: http://cooltools.sunsource.net/gcc/flags.html But as Mel said, this has nothing to do with FreeBSD, so this post doesn't exist. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Anybody familiar with "publib"??
In the last episode (Nov 29), Gary Kline said: > I found a neat function in publib that should do what I want, but > adding either > > #imclude // as per man publib > > OR > > #include "/usr/local/include/publib.h" > > fails. Yes, I am adding "-lpub" to the enc of gcc. Still bombs. > Anybody know why? Prev'ly when I've used the publib functions, I've > had to move/copy a slew of them into my private build. Be nice if > this just-worked! Try adding "-I /usr/local/include" to your compile line, and "-L /usr/local/lib" to your link line. Also, providing the error message instead of just saying "it fails" is helpful. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Why process memory starts so high up in virtual space with FreeBSD malloc?
In the last episode (Dec 01), Yuri said: > Giorgos Keramidas wrote: > > The FreeBSD malloc(3) implementation can use either mmap() or > > sbrk() to obtain memory from the system. It does not 'waste a high > > percentage of memory' but it simply maps only high addresses (with > > an unmapped 'hole' in lower addresses). > > But the hole it leaves with MALLOC_OPTIONS='dM' is way larger than > the one left by 'Dm' option. Usually malloc will keep allocating > addresses higher than this initial value and will never come back and > fill some parts of this gap. Therefore "wasting" this space. Have you actually verified that space is wasted? Note that the default for malloc is "DM", which lets malloc use both break- and mmap-based memory. Depending on the libc version, one or the other will be preferred, but both will be used if necessary. Also, unless you have tuned your kernel, you may only be able to allocate 512MB of memory via break() (the kern.maxdsize tunable lets you adjust this). Here's what I get with a simple test program on a month-old 7.1-PRE (before the preference got switched to mmap): $ MALLOC_OPTIONS=Dm ./malloctest Malloced 535822336 bytes. First: 0x808, Last: 0x27e8 $ MALLOC_OPTIONS=dM ./malloctest Malloced 2542796800 bytes. First: 0x2820, Last: 0xbfa0 $ MALLOC_OPTIONS=DM ./malloctest Malloced 3078619136 bytes. First: 0x808, Last: 0xbfa0 $ ./malloctest Malloced 3078619136 bytes. First: 0x808, Last: 0xbfa0 So using only break(), I can allocate 511 MB. Using only mmap(), I can allocate 2.36 GB. Using both (the default) I can allocate 2.86 GB. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Why process memory starts so high up in virtual space with FreeBSD malloc?
In the last episode (Dec 01), Dan Nelson said:
> Here's what I get with a simple test program on a month-old 7.1-PRE
Gah. silly mailing-list attachment stripper.
#include
#include
int main(void)
{
size_t malloced = 0;
size_t chunksize = 1024*1024;
void *first = NULL;
void *last = NULL;
void *ptr;
while ((ptr = malloc(chunksize)) != NULL) {
if (first == NULL)
first = ptr;
last = ptr;
malloced += chunksize;
}
printf("Malloced %zu bytes. First: %p, Last: %p\n", malloced, first, last);
exit(0);
}
--
Dan Nelson
[EMAIL PROTECTED]
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Reversing a ZFS mistake
In the last episode (Dec 02), Reinis Ivanovs said: > It seems I've made a mistake using ZFS, and now my /usr/local/ is > empty. I wanted to create a snapshot of a directory inside of it, so > I ran "zfs create tank/usr/local" and "zfs create tank/usr/local/www" > as I had seen in the guides I'd been using. That worked, but the > filesystems created were empty. As I found out later, doing what I > did on Solaris would have created the filesystems but not mounted > them, but on FreeBSD they were mounted automatically, and the > previous contents hidden. The question now is, how do I get my files > back? The system is crippled without /usr/local/ and I can't unmount > or destroy it, because it says that the device is busy. Any help > would be appreciated. Solaris should have automatically mounted them too, unless you had "zfs set" canmount=noauto or mountpoint=legacy on an upper filesystem. If you intend to copy/move the existing contents into these new filesystems, you can just umount them and manually mount them somewhere else ( mount -t zfs tank/usr/local /tmp/local ) while you do the copy, then remount them in their final locations. umount -f should let you force-dismount them even if processes have open filehandles on them. If it doesn't, run "fstat -f /usr/local" and kill any processes that show up, then try umounting again. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Device naming on scbus using isp
In the last episode (Sep 22), Brent Bloxam said: > Brent Bloxam wrote: > > I'm wondering about how device names are assigned on scbus, specifically > > when using the isp driver. It seems to me that there's potential when > > an HBA has access to multiple LUNs that on boot the scbus will have > > entries in /dev scrambled compared to the previous run (thus messing up > > mounts). My experience so far has been that da0 will be assigned to the > > first target scanned, da1 to the second, etc. Is this generally > > something countered with device.hints? If a LUN were to go away, but a > > device hint pointing to the target:unit remained, would that cause any > > issues on boot? > > Thought I'd follow up with a bit of information I've determined about > this, despite the lack of response from anyone on list. Maybe someone > will find it useful :) > > I can only speak for this applying to use of isp(4) with scbus(4). > > Devices that operate in target mode appear to isp(4) and are assigned a > target ID starting at 0. The order in which they appear depends on their > fcid or what's known to isp(4) as PortID. This order is ascending, so the > lower fcid takes precedence. isp(4) will then check the target to see if > any LUNs are available to it. If not, the target disappears -- and here's > the important thing to note -- but its target ID does not go away. > > Say you have 5 devices with the following fcids, 4 in target mode: > > 0x00 - target > 0x01 - target > 0x02 - another server with an HBA > 0xF0 - target with LUN > 0xF1 - target with LUN > > isp(4) is loaded at boot, and the following occurs: > > 0x00 appears, is assigned target 0, and disappears because there are no LUNs > 0x01 appears, is assigned target 1, and disappears because there are no LUNs > 0x02 appears and simply disappears because it is not a target > 0xF0 appears, is assigned target 2, and is assigned to da0 > 0xF1 appears, is assigned target 3, and is assigned to da1 > > You can see because of this example that maintaining device names using > /boot/device.hints is impossible if targets in the fabric change. If > 0x00 were to disappear, the target IDs would change and render > /boot/device.hints invalid, or worse, the wrong LUN could be given the > wrong device name. > > Ideally, there would be a way to assign target IDs by fcid, but that > does not exist presently. If you're mounting UFS filesystems, you can label them and mount them by label (see the tunefs and glabel manpages for more info). ZFS should find its pool devices automatically, but you can always manually label devices with glabel and refer to the label instead of the da## name. -- Dan Nelson dnel...@allantgroup.com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: lacie external hard drive supported?
In the last episode (Sep 26), Chris Whitehouse said: > Hi all > > Does anyone have experience of a LaCie Hard Disk, Design by Neil Poulton > eSATA, FireWire 400 & Hi-Speed USB 2.0 1TB ? > > http://www.lacie.com/uk/products/product.htm?pid=11064 > > Are all the interfaces supported? The data sheet says it is supported > under Linux. Does that mean it is safe to assume it is under FreeBSD? In general, you can assume that any SATA/SAS/firewire/usb hard drive will be supported. The only thing that might not be supported is RAID array management for the more expensive external units (you'll still be able to acess the data though). -- Dan Nelson dnel...@allantgroup.com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Swap and memory optimization
In the last episode (Oct 01), Bill Moran said: > bsd wrote: > > I have a FBSD 6.4p7 box that I use as a mail server - 1Go RAM - RAID1 > > Works quite well. > > > > As I plan to put 100 more mail accounts soon on the server I was > > wondering if the memory & swap was ok on the server considering these > > figures: > > > > last pid: 18956; load averages: 0.04, 0.11, 0.05 up 19+08:36:23 > > 09:53:38 > > 125 processes: 1 running, 124 sleeping > > CPU: 0.0% user, 0.0% nice, 1.5% system, 0.4% interrupt, 98.1% idle > > Mem: 499M Active, 70M Inact, 362M Wired, 41M Cache, 111M Buf, 20M Free > > Swap: 2000M Total, 160M Used, 1840M Free, 8% Inuse > > > > Though It looks good to me - the server swaps a bit (between 8 to 14%) > > and there is not much memory left. > > Looks like the server would run more smoothly with a bit more RAM. At > least an additional 256M, I would think, but considering the price of RAM, > you might as well just up it to 2G. The amount of used swap is much less important than whether you are actively swapping (if there are In/Out values on the Swap line in top, or if "vmstat 1" shows nonzero values in the pi/po columns). 160MB of used swap is fine if it's just unused daemons (getty, idle webserver, etc). More memory can never hurt, but it doesn't seem like it's urgently needed here. -- Dan Nelson dnel...@allantgroup.com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
