Re: My unqualified host name
If you look for My unqualified host name unknown; sleeping for retry you will get a lot of possible answers; some suggesting to add your unqualified host name in /etc/hosts. That line is already in /etc/hosts, both with and without a trailing period. I still get the ~3 messages about a minute apart during boot. The first 4 lines of /etc/hosts look like: ::1localhost 127.0.0.1 localhost 192.168.200.61 foo 192.168.200.61 foo. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: My unqualified host name
nyana sm-mta[803]: My unqualified host name (nyana) unknown; sleeping for retry ... sendmail expects your machine to have working DNS and for the machine to have a valid FQDN. Either set that up, or add sendmail_enable=NONE to /etc/rc.conf to disable sendmail ... There is another approach, which is to ignore the message. After something like 3 repetitions, at something like a minute apart, it will give up on qualifying its name. Everything seems to work just fine thereafter until the next boot, when the entire sequence repeats. Respectfully, my gut reaction is this is, if not /bad/ practice, at least not /good/ practice. The requirements for geting sendmail to behave (at least in this regard) are not particularly onerous; If sendmail *will not work properly* without a valid FQDN, that alone is onerous. See below. why not just diagnose and fix the root problem? because I have no clue how to do it, without adopting settings that I don't want! Dunno about the OP, but my FreeBSD machines do not have nor need valid FQDNs because they sit behind a NAT firewall (and therefore do not have externally-identifiable IP addresses). I want hostname to simply return the unqualified host name (say, foo), not foo.com nor foo.uucp nor even foo.bogus. I don't need sendmail to handle anything but purely local traffic, such as the periodic reports to root, and it's just fine for it to identify itself simply as foo. We were able to do things like this back in the days of SunOS 4, so why should it be difficult to accomplish today? Indeed, why should it not be the default mode of operation when hostname returns an unqualified name? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: My unqualified host name
nyana sm-mta[803]: My unqualified host name (nyana) unknown; sleeping for retry ... sendmail expects your machine to have working DNS and for the machine to have a valid FQDN. Either set that up, or add sendmail_enable=NONE to /etc/rc.conf to disable sendmail ... There is another approach, which is to ignore the message. After something like 3 repetitions, at something like a minute apart, it will give up on qualifying its name. Everything seems to work just fine thereafter until the next boot, when the entire sequence repeats. This leads to the question of how to get sendmail -- or whatever -- into the state where it will eventually land after the 3-miunte delay, without the delay and the messages. It seems as if this ought not be all that difficult. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Realtek 8111C?
... I thought I'd try copying the driver from 7.1. The module (if_re.ko) appears to load successfully at boot time, however the NIC is still not shown in dmesg and unavailable as re0. I admit I don't know if this should even work (7.1-compiled module on 6.3 kernel) ... That would indeed not be expected to work. Even a 7.1 module on a 7.0 system would be hit-or-miss; across a major revision it's a wonder it didn't panic. You could try building the module from the 7.1 source on 6.3. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: server is crashing constantly
Jonathan Horne [EMAIL PROTECTED] wrote: I have a new web server [which] is crashing 2-3 times a day ... That is what im getting in the /var/log/messages. Sep 13 20:09:25 rps savecore: reboot after panic: page fault Sep 13 20:09:25 rps savecore: writing core to vmcore.0 ... Any ideas or recommendations about where to start looking to track this down would really be appreciated. It's likely to be hardware problems (yes, even on a new box). memtest and/or memtest86, in ports/sysutils. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Capturing tar output
redirect stderr with 2 operator Using the following command, # /usr/local/gtar/bin/tar -cvf - /home/hallja 2 /var/log/test.txt | /usr/local/bin/gpg --encrypt recipient | dd of=/dev/nsa0 obs=128k I receive an error meesage stating, Ambiguous output redirect. Wojtek correctly pointed out that there should be no space between the 2 and the , but I suspect the primary problem is that 2 is Bourne/Korn/Bash syntax and your root shell is most probably csh. Does it work any better if you first start /bin/sh: # sh sh will give you another # prompt, and then it should work: # /usr/local/gtar/bin/tar -cvf - /home/hallja 2 /var/log/test.txt | /usr/local/bin/gpg --encrypt recipient | dd of=/dev/nsa0 obs=128k After it finishes, and sh prompts again: # ^D The Ctrl-D will exit from sh, returning to the csh prompt. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Postfix issue
why are you not using your ISP to relay emails, using its mail gateway (which should have a static IP address)? ... I do not like the fact that a number of governments (including most european ones) now have the right to access all emails that pass through an ISP's server. They do not have the right to access private server systems unless they have a warrant. This *is* a valid concern, but it's not clear to me how it applies to messages that are being sent to public mailing lists where they will be as available to Big Brother as to anyone else. How about configuring your MTA to send anything going to a public list via your ISP, and send directly only messages that aren't going to be posted for the world to see? Another emerging issue is cable operators refusing to allow fixed IP address so they can receive revenue from reporting on user usage data. I seriously doubt that as a motivation. If anything, static IP assignments would make it *easier* to track per-customer usage. A more likely reason is that most residential users, even on cable or DSL, do not keep their router (or system, if they have only one and therefore don't use a router) on-line anywhere near 24-7. The ISP can serve several customers per IP address by using DHCP (so that customers occupy IP addresses only when on-line). ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: rss-glx screen saver compilation fails?
spirographx.o(.text+0x30f): In function `getAll': : undefined reference to `sincosf' Per Google, it's a gnu-ism: http://linux.die.net/man/3/sincosf void sincosf(float x, float *sin, float *cos); Several applications need sine and cosine of the same angle x. This function computes both at the same time, and stores the results via the given pointers. Probably a 10-liner by just calling sin() and cos() separately -- a bit more work to do it properly -- or just grab the gnu code if you don't need to be BSD-licensed. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: which gray is best for print?
So you're saying that the white on my [monster] CRT is not the same as on a future LCD Display? rats:) Not only that, but your monster CRT probably doesn't match a smaller CRT; and an old-ish CRT whose phosphors have aged (and whose focus may have gotten a bit fuzzy) probably doesn't match a new, sharp one. Different LCDs may not match each other either, esp. if they use different backlight technologies or if some of the backlights -- or faceplates -- are subject to color shifts with age. This is due to the nature that these devices use different color spaces (RGB, composed additively, CMY, composed negatively), and most of them even aren't calibrated ... I took all 5 quarters of physics, like most of us, but never got far into optics ... and there's more involved than physics and optics anyway, e.g. the neuropsychology of human visual perception. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD 7.0-RELEASE getting terrible throughput using sk0 adapter
Try disabling usb and firewire in BIOS. You may need to have a tech there do it for you. Your box has the sk NIC and usb sharing an irq. The NIC driver is MPSAFE but the usb stack is still under the GIANT lock. Disable usb and the NIC driver should perform better. Alternatively, to avoid involving the provider's tech support, could the OP get the same effect by building a kernel without USB? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Formatting dates to a specific pattern
I need to format the current date ... to the pattern m-d- ... date(1) seems to always put leading zeros. # date +%m-%d-%Y | sed 's/^0//g' 8-30-2008 Not quite. That fixes the month, but not the day: $ echo 02-04-2008 | sed 's/^0//g' 2-04-2008 (The g does nothing, because the ^ can match only at the beginning of a line.) This does both: $ echo 02-04-2008 | sed -e 's/^0//' -e 's/-0*/-/' 2-4-2008 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD, Ubuntu and Win XP on one system
I recommend installing FreeBSD first, then Windows and then Ubuntu ... Unless something has changed since the last time I was messing with this sort of thing, one hazard of installing a Linux last is that there may by then be no space left for the /boot partition, which has to be below cylinder 1024 to be accessible by BIOS. One might want to allocate what will become /boot as early in the process as possible. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Xerox Phaser 6110 printer
Does anybody have a Xerox Phaser 6110 printer working with FreeBSD? I've never had any trouble with my 6120, but I guess the crucial difference is the PostScript support in the 6120. The 6130 just works -- it internally supports lpr/lpd, not even needing CUPS -- but it, too, is PostScript. I'd be very cautious about any printer that doesn't support PostScript or at least PCL, even one from a first-rate supplier like Xerox. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: setting the other end's TCP segment size
1) create a static ARP entry, this will create an entry to the routing table i.e. arp -S IPADDR MACADDR 2) modify the mtu for that destination i.e. route change IPADDR -mtu MTU Seems to work fine :) One problem with this approach is that a hard-coded MAC address would break if the destination's MAC address changed :( but this can be scripted around by pinging the destination (to ensure that it's up, and get an arp entry the usual way), then reading the MAC address from the arp table. d=192.168.200.3 ping -c 1 $d \ arp -S $d ` arp -n $d | sed -e 's/^.* at //' -e 's/ on .*$//' ` \ route change $d -mtu 640 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: setting the other end's TCP segment size
Is there a simple way for a FreeBSD system to cause its peer to use a transmit segment size of, say, 640 bytes -- so that the peer will never try to send a packet larger than that? I'm trying to get around a network packet-size problem. In case it matters, the other end is SunOS 4.1.1 on a sun3, and I've been unable to find a way to limit its packet size directly. ... Each tcp conversation can have it's own size set along with a bunch of other params. Good point. The TCP_MAXSEG can reduce the maximum segment size for a single TCP connection to something smaller than the interface MTU :) That would be OK, provided I could somehow arrange for it to apply to all conversations with this particular destination (which is what the next item seems to do :) Just adding that MTU can be set per destination with the help of route(8) and the -mtu modifier. That would be better than setting the local mtu -- which has been causing other problems although it takes care of the original -- and it is a better match to the physical situation. (The culprit is neither the Sun nor the FreeBSD system, but the physical link between the Sun and the hub.) What I haven't been able to come up with is a way of making such a setting permanent. If I've communicated with the Sun recently enough, netstat -r -W reports a line like this (some spaces removed, for length, and I've no longer got xl0's mtu set low) Destination Gateway Flags Refs Use Mtu Netif Expire 192.168.200.3 08:00:20:00:a7:a6 UHLW 1 34 1500 xl0 1184 Now if I do # route change 192.168.200.3 -lock -mtu 640 the mtu column changes to 640 and it works fine, but only until the routing entry expires. Adding -static makes no difference -- the entry still expires and loses the mtu specification. I've been unable to come up with a route command that will *create* an entry like that (vs modifying an existing one), nor that will transform a transient entry into a permanent one. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
building only part of the world
How would I go about building, not the entire world, but only a small part of it? If I just cd to the desired subdirectory and type make -n -- intending to find out what it would try to do -- I get a warning about not having changed the object directory. I suppose I'm supposed to type something along the lines of make -n OBJ=something but what should I be setting OBJ to? An attempt to find(1) some of the expected output files, so as to discover where they are conventionally located, found nothing. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
undocumented tar --unlink switch
Around line 37 of /usr/src/usr.sbin/pkg_install/add/extract.c there's an invocation of /usr/bin/tar with a --unlink switch, which I don't see mentioned in the tar(1) manpage. Anyone happen to know what this does, or do I need to dig into the code? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: setting the other end's TCP segment size
You can edit `/etc/hostname.foo0' in the Sun too, and add something like: 192.168.1.10/24 mtu 640 [getting OT for FreeBSD] Are you sure that works as far back as SunOS 4.1.1? /etc/hostname.le0 currently consists of the single word pluto and it looks as if this causes /etc/rc.boot to do ifconfig le0 pluto netmask + -trailers up (where the name pluto resolves to 192.168.200.1 via /etc/hosts). If I were to change /etc/hostname.le0 to pluto mtu 640 I think the ifconfig command would become ifconfig le0 pluto mtu 640 netmask + -trailers up and it doesn't look as if ifconfig recognizes mtu as a keyword (at least while running -- granted I haven't tried actually editing /etc/hostname.le0 and rebooting): # ifconfig le0 pluto mtu 640 ifconfig: mtu: bad address ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: setting the other end's TCP segment size
Is there a simple way for a FreeBSD system to cause its peer to use a transmit segment size of, say, 640 bytes -- so that the peer will never try to send a packet larger than that? I'm trying to get around a network packet-size problem. In case it matters, the other end is SunOS 4.1.1 on a sun3, and I've been unable to find a way to limit its packet size directly. Setting the interface MTU should do it, i.e.: ifconfig re0 mtu 640 Not all interfaces support setting the MTU and some may have range restrictions though. In particular, this seems to work with my wlan0 interface, but not with my re0 interface ... That's certainly simple enough, and xl0 apparently supports the reduced mtu setting. It seems to be working just fine. Thanks! I'd thought of trying to set the sun's MTU, but hadn't been able to find a way to do it. It had never occurred to me that setting the *recipient's* MTU would limit the *sender's* packet size. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
setting the other end's TCP segment size
[TCP] splits traffic to 'segments' using its own logic ... Is there a simple way for a FreeBSD system to cause its peer to use a transmit segment size of, say, 640 bytes -- so that the peer will never try to send a packet larger than that? I'm trying to get around a network packet-size problem. In case it matters, the other end is SunOS 4.1.1 on a sun3, and I've been unable to find a way to limit its packet size directly. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: PACKAGESITE
Did you specify the -r flag? Without that, the PACKAGESITE environment variable is note used ... No, I didn't, because -- unless I am misunderstanding the description of the -r flag -- that will cause pkg_add to look *only* on the FTP site. I want it to use packages that have already been downloaded, and use the FTP site only when a needed package is not available locally. I'm trying to install an already-downloaded 10MB package which has quite a few dependencies, several of which were already fetched during a previous attempt. IOW I want the equivalent of specifying the current directory, followed by the FTP site, in PKG_PATH; but the colon in the URL messes that up by looking like a pathname separator. If I tried something like setenv PKG_PATH .:ftp://ftp.freebsd.org/pub/FreeBSD/ports/i386/packages-7-stable/All/ it would look first in the current directory, then in a subdirectory named ftp, and finally in a directory named //ftp.freebsd.org/... ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: PACKAGESITE
As far as I know, pkg_add will only fetch dependencies recursively from the Internet when used with -r but it will then ignore PKG_PATH. Seems what you are asking cannot be done this way ... I wonder if portinstall -P (or even -PP) might do what the OP wants? You are correct, according to the man page, portinstall -PP would be his best bet. Except that portinstall is part of portupgrade, which has its own boatload of dependencies. Is there any way to do this with, say, portmaster? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: PACKAGESITE
... portinstall is part of portupgrade, which has its own boatload of dependencies. You must be used to sailing in very small boats. From lurking on questions@ for a while, I have gotten the impression that ruby alone would pretty well fill up a Panamax :) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
PACKAGESITE
Can someone provide a correct example of setting PACKAGESITE so that pkg_add will find the 7-stable packages for i386? I have tried ftp://ftp.freebsd.org/pub/FreeBSD/ports/i386/packages-7-stable/Latest/ as shown in the handbook, and also: ftp://ftp.freebsd.org/pub/FreeBSD/ports/i386/packages-7-stable/All/ ftp://ftp.freebsd.org/pub/FreeBSD/ports/i386/packages-7-stable/ ftp://ftp.freebsd.org/pub/FreeBSD/ports/i386/packages-7-stable and all have failed. I get messages like pkg_add: could not find package expat-2.0.1 ! pkg_add: could not find package png-1.2.28 ! pkg_add: could not find package pkg-config-0.23_1 ! etc. Even specifying -v does not cause pkg_add to show exactly where it is looking (which might provide a clue as to what the correct setting would look like). The pkg_add manpage shows an example for PACKAGEROOT, but not for PACKAGESITE. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Data loss after power out - fsck: bad inode number to nextinode
What should I do?
In theory,
clri {special-file} 306176
should wipe the inode containing the bad pointer and allow fsck to
continue, perhaps recovering the files pointed to by that directory
into lost+found.
Definitely try this on a copy first if at all possible.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: anyone been crazy enough to mirror wikipedia?
In my house, we had an encyclopedia because I was in school ... it was useful for research papers. I suspect the usefulness would depend on what one's teachers meant by research, which tends to change with grade level. In elementary and middle school, certainly. In high school, maybe. In college, probably not. Postgraduate, almost certainly not; at that level one should be using primary sources (and likely know enough to be writing articles *for* an encyclopedia :) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: mount_nfs not accepting syntax specified by its usage complaint
option 'g' has been removed temporarily, but usage() hasn't been updated accordingly. Aha! Removing the -g 8 fixed it, and the 512-byte read and readdir restrictions seem to be working well (albeit slowly). That a temporary removal present in 7.0 has been there at least since 6.1 reminds me of an OS/360 PTF, which officially stood for Program Temporary Fix but was sometimes reputed to actually mean Permanent Temporary Fix. Is there any realistic prospect of the -g option being restored in the near future? If not, perhaps I should submit a patch to update the usage() and manpage. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
mount_nfs not accepting syntax specified by its usage complaint
How does this command: # mount_nfs -dis -g 8 -I 512 -R 3 -r 512 -w 512 solomon:/var/spool/uucp /solomon/uucp not comply with the resulting usage complaint? usage: mount_nfs [-234bcdiLlNPsTU] [-a maxreadahead] [-D deadthresh] [-g maxgroups] [-I readdirsize] [-o options] [-R retrycnt] [-r readsize] [-t timeout] [-w writesize] [-x retrans] rhost:path node And yes, I really do want to set the read, write, and readdir sizes to 512 bytes (to get around a network packet-size problem to which I can find no other solution). The server is up: # ping solomon PING solomon (192.168.200.3): 56 data bytes 64 bytes from 192.168.200.3: icmp_seq=0 ttl=255 time=2.807 ms 64 bytes from 192.168.200.3: icmp_seq=1 ttl=255 time=2.755 ms 64 bytes from 192.168.200.3: icmp_seq=2 ttl=255 time=2.555 ms ^C --- solomon ping statistics --- 3 packets transmitted, 3 packets received, 0% packet loss round-trip min/avg/max/stddev = 2.555/2.706/2.807/0.109 ms and the mount point does exist: # ls -ld /solomon/uucp drwxr-xr-x 2 root wheel 512 Jul 4 18:34 /solomon/uucp # uname -a FreeBSD fbsd70.uucp 7.0-RELEASE FreeBSD 7.0-RELEASE #0: Sun Feb 24 19:59:52 UTC 2008 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC i386 I get the same behavior, with a very slightly different usage message, from a 6.1 system. The server is an ancient sun3 running SunOS 4.1.1-U1, but it looks as if the mount attempt is not getting far enough for that to matter. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Windows Unix volunteers
... the changing of wallpaper is VERY window manager centric ... xsetroot(1) would not work for all? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: sendmail's outgoing IPs
i have 3 different links to ISP all are ADSL's so outgoing bandwidth is low, i would like to spread the load generated by outgoing mails. Pardon my lack of imagination, but how could anyone -- other than a spammer -- be generating enough outbound email traffic to *need* to load-balance it, and yet have little enough total Internet traffic to be using DSL rather than something oriented to commercial use (like a T1)? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD and User Security
How do you know that the bios has not been reflashed by a virus, trojan, or rootkit? For that matter, how do you know that the *original* bios was free of interesting non-essentials? It's been a few years since bios were delivered in socketed ROMs/EPROMs (readable by a standalone device, independently of their own operation) or since sources were typically published :) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD and User Security
It is my understanding that since 1995 all computers must have a hardware back door that permits undetectable access by the government to the computer. This capability can be implemented using System Monitor(Maintenance) Mode which is built into all x86 computers now. It would appear that, if you are connected to the internet, the government has access to your computer. if it were true, this system maintenance mode would have to access your network card in parallel with main OS without making conflicts A near-trivial exercise in virtualization, provided it knows what kind of card is in use and what addresses it occupies, which is rather easy if the card is in fact built onto the mainboard. Of course, it is also trivial to defeat it by using an add-in card instead of the one on the mainboard, esp. a card whose design did not exist when the bios was written. Cycles consumed by SMM might also explain why some PCs' clocks seem to run slower than real time ... ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Duplex printer advice
my own requirements list includes (color duplex printer scanner). I don't need it to be a laser, but I do need both color, multifunc, and duplex printing ... I begin to wonder if I could find one with the same specs ESCEPTING it was the cheaper technology of inkjet. ... please don't spend time trying to talk me out of features like color, or duplex, I like both too well ... If you want *good* color, look into Xerox. Seriously. They are not as costly as you may be expecting. My experience with inkjet is that it produces inferior results, and saves little or no money if you consider the cost of ink, esp. when you have to pitch a 3/4-full cartridge because it clogged up. BT, DT. And no, I don't work for Xerox. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
non-RAID SATA
I am looking for a cost-effective way to add a SATA drive to an existing 7.0 system whose on-board controller is PATA, and am not getting very far at all in identifying an inexpensive controller which would be expected to work well. (I'd prefer PCI, since the USB in this box is probably 1.0 and not capable of sustaining desirable data rates.) All mentions of SATA in the hardware guide and FAQ seem to be of RAID controllers, or else too generic to guide a choice of add-in cards. Does anyone have any experience with this that they would be inclined to share? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: time drift
FreeBSD 6.2 running on X86 hardware (FSC) shows a remarkable time drift running ntpdate every half hour shows that the system looses about 10-14 sec each time. 15 May 10:06:48 ntpdate[7200]: step time ... offset -13.799602 sec 15 May 10:36:48 ntpdate[7515]: step time ... offset -12.813941 sec 15 May 11:06:48 ntpdate[7879]: step time ... offset -13.651921 sec 15 May 11:36:50 ntpdate[8079]: step time ... offset -11.109298 sec 15 May 12:06:50 ntpdate[8289]: step time ... offset -11.836499 sec ... The 6.2 system is a production system, has a uptime of almost 300 days and I don't want to experiment a lot with acpi, battery or so. What would be your suspicion on the large time drift of the FreeBSD 6.2 system? With an uptime of nearly a year -- commendation to the power company -- and (I take it) a recently-developed problem, I'd be asking what might have changed shortly before the problem appeared. Is the system clock source by any chance the CMOS RTC? If so, I'd suspect that its battery may be dying. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Belkin F5D9050 ver 4000
I checked the FreeBSD 7.0-RELEASE Hardware Notes before I bought this USB Ethernet device. It should be supported by the rum driver, but it gets picked up by ugen instead. The Hardware Notes mentions version 3 and the box says ver 4000, but I think it's probably actually ver 4. ... What am I missing? It sure sounds as if you are missing a supported USB device :( Unfortunately, it is not at all uncommon for manufacturers to make significant internal changes to a product, without changing the name or the packaging. At least they changed the version label. Doing a descriptor dump, and posting the results to freebsd-usb@, might find someone who knows how to get that particular device to work. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Belkin F5D9050 ver 4000
Doing a descriptor dump, and posting the results to freebsd-usb@, might find someone who knows how to get that particular device to work. Ok, I'll bite. How do you do a descriptor dump? One way is to use sysutils/udesc_dump, from ports, as recommended here: http://lists.freebsd.org/pipermail/freebsd-usb/2008-January/004308.html ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: wine: notepad OK, others not
... very shortly after starting the actual install I got an error box: VISIO Setup ! Tried to create an invalid path using 'A:\' and 'clipart.vs_' and it locked up the display so that CtrlAltF1 would not switch to a text screen (although it did allow AltTab to bring up FVWM's window list). After clicking OK: Visio Setup i Setup failed. and it quit. Were there any messages printed in the terminal window? No. What version of Visio is this? 3.0. Long before M$ took it over, so it should be just a generic Win32 app with no secret M$ tricks. You might also want to try on a Linux system if you have that somewhere, just to see if it works there. If I had a Linux system set up, I'd have tried this there in the first place. Also, you should really take this to the [EMAIL PROTECTED] mailing list. The people there should know more about this than here on a FreeBSD mailing list. I suppose I can try it, but I wonder how much interest there will be on a wine list in supporting FreeBSD. At a minimum I suppose they'll want to know if it still breaks on the latest wine version, and I'll have no way to find out since the FreeBSD port doesn't support the latest wine version. In any case, it seems FreeBSD should not be allowing a port -- any port -- to lock out CtrlAltF1. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: wine: notepad OK, others not
Wine has its own simple version of Wordpad though. Just run wine wordpad. This version has its own notepad, but it doesn't appear to have wordpad. (There's no wordpad.exe that I can find, but there are two identical copies of notepad.exe -- one in .../windows and the other in .../windows/system32.) wine wordpad still works though. It's in /usr/local/lib/wine. Along with yet another notepad, this one twice the size of the ones in .../windows and .../windows/system32! The obvious followup is which one actually gets used if someone runs wine notepad, but I doubt it's worth looking into. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
wine: app won't install from floppy
What is the correct way of installing something into wine from floppy? The obvious approach: $ wine 'a:\setup.exe' did not work with a (fairly old version of) Visio: the option dialogs seemed to work properly, but very shortly after starting the actual install I got an error box: VISIO Setup ! Tried to create an invalid path using 'A:\' and 'clipart.vs_' It somehow locked out CtrlAltF1 so I could not switch to a text screen, but AltTab did bring up FVWM's window list (thus making it possible to transcribe the error message into an xterm that was not related to the wine session). After clicking OK I got an info box: Visio Setup i Setup failed. and it quit when I clicked OK there. Comparing the contents of .wine before/after the attempt, it seems the only change involves the font entries in the registry: $ diff -r wineBak .wine diff -r wineBak/system.reg .wine/system.reg 9082c9082 [Software\\Microsoft\\Windows NT\\CurrentVersion\\Fonts] 1208923638 --- [Software\\Microsoft\\Windows NT\\CurrentVersion\\Fonts] 1209616619 diff -r wineBak/user.reg .wine/user.reg 453c453 [Software\\Wine\\Fonts] 1208923638 --- [Software\\Wine\\Fonts] 1209616619 456c456 [Software\\Wine\\Fonts\\External Fonts] 1208923638 --- [Software\\Wine\\Fonts\\External Fonts] 1209616619 and of course I have no clue whether this makes any difference to anything. $ uname -a FreeBSD fbsd70.uucp 7.0-RELEASE FreeBSD 7.0-RELEASE #0: Sun Feb 24 19:59:52 UTC 2008 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC i386 $ wine --version wine-0.9.48 a: is symlinked to /fd in .wine/dosdevices and /dev/fd0 is mounted on /fd ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: wine: notepad OK, others not
You have to mount the floppy and then link a: to the mount point. So if you mount it under /mnt you'd need this: mount -t msdosfs /dev/fd0 /mnt ln -s /mnt ~/.wine/dosdevices/a: That got only a little bit farther. It did find the setup program, and the option dialogs seemed to work properly, but very shortly after starting the actual install I got an error box: VISIO Setup ! Tried to create an invalid path using 'A:\' and 'clipart.vs_' and it locked up the display so that CtrlAltF1 would not switch to a text screen (although it did allow AltTab to bring up FVWM's window list). After clicking OK: Visio Setup i Setup failed. and it quit. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: wine: notepad OK, others not
Wine has its own simple version of Wordpad though. Just run wine wordpad. This version has its own notepad, but it doesn't appear to have wordpad. (There's no wordpad.exe that I can find, but there are two identical copies of notepad.exe -- one in .../windows and the other in .../windows/system32.) I'm not entirely sure, but I think the :: link is only used for raw access to devices. Wine doesn't mount disks on its own. For floppies, which AFAIK are always formatted as FAT, I'd settle for having it use mtools so the disk wouldn't need to be mounted at all :) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: wine: notepad OK, others not
It's really easier to try to install an app under Wine ... With, it seems, at least two exceptions: * Some apps -- such as Wordpad and Write -- are packaged and installed with Windows, rather than on separate media. Are there instructions somewhere for installing such an app under wine? I'm certainly not finding it at all obvious. * Some add-on (separately installable) apps are packaged on multiple diskettes (or multiple CDs for that matter). Pre-mounting the first, and pointing wine at the mount point, seems likely to result in getting stuck partway through the install when it asks for the second disk. The version of Visio that I have is in the second category. The manpage describes a way of pointing wine to a device rather than to a mounted filesystem: The Unix device corresponding to a DOS drive can be specified the same way, except with '::' instead of ':'. So for the previous example, if the CDROM device is mounted from /dev/hdc, the corresponding symlink would be $WINEPREFIX/dosdevices/d:: - /dev/hdc. but, as reported elsewhere, wine could not find setup.exe on the Visio install diskette with dosdevices set up this way. ... You also might want to have a look at http://wiki.winehq.org/winetricks for a script that can install and setup various packages ... Unfortunately, I can't find Visio in its list of packages. Is there something else to try, or is installing an app like Visio beyond Wine's current capabilities? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: wine: notepad OK, others not
If I were you I'd just try to install Visio under Wine and see how it goes. That is, mount the install cd, check with winecfg that Wine can see the mount point as a drive D: or something and then run wine d:\\setup.exe ... It did not work at all. This version of Visio is old enough that it installs from floppies, rather than from CD :) and I've set it up in .wine/dosdevices according to the manpage: $ ls -la .wine/dosdevices total 4 drwxr-xr-x 2 perryh staff 512 Apr 21 00:17 . drwxr-xr-x 4 perryh staff 512 Apr 22 21:07 .. lrwxr-xr-x 1 perryh staff8 Apr 21 00:17 a:: - /dev/fd0 lrwxr-xr-x 1 perryh staff 10 Apr 19 16:39 c: - ../drive_c lrwxr-xr-x 1 perryh staff1 Apr 19 16:39 z: - / mdir can read the disk, and it does contain a setup.exe, but wine can't see it: $ wine a:setup.exe wine: cannot find 'a:setup.exe' $ wine 'a:\setup.exe' wine: cannot find 'a:\setup.exe' There doesn't seem to be a manpage for winecfg: $ man winecfg No manual entry for winecfg and when I tried to run it it was not at all obvious what to do. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: wine: notepad OK, others not
$ ls -la .wine/dosdevices total 4 drwxr-xr-x 2 perryh staff 512 Apr 21 00:17 . drwxr-xr-x 4 perryh staff 512 Apr 22 21:07 .. lrwxr-xr-x 1 perryh staff8 Apr 21 00:17 a:: - /dev/fd0 Is the second colon intentional Yes! That is exactly what the manpage says to do, so as to have wine use a *device* rather than a node in the Unix filesystem: $WINEPREFIX/dosdevices Directory containing the DOS device mappings. Each file in that directory is a symlink to the Unix device file implementing a given device. For instance, if COM1 is mapped to /dev/ttyS0 you'd have a symlink of the form $WINEPREFIX/dosdevices/com1 - /dev/ttyS0. DOS drives are also specified with symlinks; for instance if drive D: corresponds to the CDROM mounted at /mnt/cdrom, you'd have a symlink $WINEPREFIX/dosdevices/d: - /mnt/cdrom. The Unix device corresponding to a DOS drive can be specified the same way, except with '::' instead of ':'. So for the previous exam- ple, if the CDROM device is mounted from /dev/hdc, the corre- sponding symlink would be $WINEPREFIX/dosdevices/d:: - /dev/hdc. Presumably this method is provided so that wine can be given access to a removable device without a particular disk having to be mounted. It would be, at the least, inconvenient to have to mount and unmount a sequence of 5 floppies to do this installation. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: wine: notepad OK, others not
... If you want to run applications under Wine either install them under Wine or (with simple applications) copy them over from a Windows install into ~/.wine/drive_c. And specifically I'd populate ~/.wine/drive_c/windows/fonts from a real windows installation. Which raises the question: how does one figure out what-all pieces of a real windows installation should and should not be copied (or symlinked) into ~/.wine/drive_c? So far it looks as if some (but surely not all) .exe's and .dll's, and (all?) fonts, should be imported. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: wine: notepad OK, others not
I have installed wine-0.9.20 from ports, and there's a Win98 FAT32 slice mounted on /windoze deletia How do I fix this? I would start by upgrading and re-installing wine. You have 0.9.20; the current version is 0.9.55 and I believe there have been substantial improvements. When I updated my ports, the newer wine refused to install on 6.1, saying it wouldn't work properly on anything prior to 6.3 IIRC. Rather than risk breaking my primary system in an upgrade attempt, I installed 7.0-RELEASE and wine-0.9.48 (the version from the 7.0-RELEASE ports) on a different machine. It happened to have XP, so I'm now trying to run XP .exe's instead of win98 .exe's. Notepad and Write are, if anything, worse than before: now, if I just start typing without selecting a font, I get something that looks more or less like dingbats. They do seem to work if I explicitly select Courier. (The ultimate goal is to run Visio, not to do word processing, but I'm trying to start with something simple.) The problem with wordpad has not changed very much: $ wine /winxp/Program Files/Windows NT/Accessories/wordpad.exe err:module:import_dll Library MFC42u.DLL (which is needed by LZ:\\winxp\\Program Files\\Windows NT\\Accessories\\wordpad.exe) not found err:module:LdrInitializeThunk Main exe initialization for LZ:\\winxp\\Program Files\\Windows NT\\Accessories\\wordpad.exe failed, status c135 (Previously it was looking for MFC42.DLL instead of MFC42u.DLL.) OK, it doesn't know where to find the DLLs. Try making a symlink to a place which (per the manpage) is always searched: $ ls -l /winxp/WINDOWS/SYSTEM32/mfc42u.dll -rwxr-xr-x 1 root wheel 995384 Aug 23 2001 /winxp/WINDOWS/SYSTEM32/mfc42u.dll $ ln -s /winxp/WINDOWS/SYSTEM32/mfc42u.dll /usr/local/lib/wine $ wine /winxp/Program Files/Windows NT/Accessories/wordpad.exe err:module:import_dll Library MFC42u.DLL (which is needed by LZ:\\winxp\\Program Files\\Windows NT\\Accessories\\wordpad.exe) not found err:module:LdrInitializeThunk Main exe initialization for LZ:\\winxp\\Program Files\\Windows NT\\Accessories\\wordpad.exe failed, status c135 Maybe the search is case-sensitive (although Windows ordinarily isn't)? $ ln -s /winxp/WINDOWS/SYSTEM32/mfc42u.dll /usr/local/lib/wine/MFC42u.DLL $ wine /winxp/Program Files/Windows NT/Accessories/wordpad.exe err:module:import_dll Library MFC42u.DLL (which is needed by LZ:\\winxp\\Program Files\\Windows NT\\Accessories\\wordpad.exe) not found err:module:LdrInitializeThunk Main exe initialization for LZ:\\winxp\\Program Files\\Windows NT\\Accessories\\wordpad.exe failed, status c135 The symlinks in /usr/local/lib/wine *do* point to that DLL, and they *can* be followed successfully: $ ( cd /usr/local/lib/wine ; ls -lL mfc* MFC* ) -rwxr-xr-x 1 root wheel 995384 Aug 23 2001 MFC42u.DLL -rwxr-xr-x 1 root wheel 995384 Aug 23 2001 mfc42u.dll Now what? I can't imagine anyone would be able to do much with wine if problems finding DLLs were common. What am I doing wrong? $ wine --version wine-0.9.48 $ uname -a FreeBSD fbsd70.uucp 7.0-RELEASE FreeBSD 7.0-RELEASE #0: Sun Feb 24 19:59:52 UTC 2008 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC i386 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Adobe Flash Player Petition
Where do such articles go for FreeBSD? ... I can make a website on my server for it, but it seems to be a kind of overkill, for just such installing sequence.. If you have things working for Flash 9, enough people /will/ be interested this is not overkill. My suggestion: 1) create a web page; present the steps (with commentary is appropriate) in an easily readable format; _date the page_. 2) post the URL to questions@, ports@, and possibly www@ and multimedia@ in separate posts. Make the subject line something relevant - this is not about Adobe Flash Player Petition. and/or send a PR, including the content (not just the URL), to have it added to the Handbook or the FAQ. That way, even if the website goes away before a doc committer gets to it, it's archived in the PR database. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: pkg_add goofiness in 7.0
... extract: execute '/usr/local/bin/xmlcatmgr -sc /usr/local/share/sgml/catalog.ports add linuxdoc/catalog' xmlcatmgr: unbalanced arguments for `add' action pkg_add: command '/usr/local/bin/xmlcatmgr -sc /usr/local/share/sgml/catalog.ports add linuxdoc/catalog' failed ... That looks like a warning from xmlcatmgr which may or may not be important, but the package apparently added itself completely (no errors were reported by pkg_add). It may not make a lot of difference, but I am now wondering how to recognize an error reported by pkg_add since this: pkg_add: command '/usr/local/bin/xmlcatmgr -sc /usr/local/share/sgml/catalog.ports add linuxdoc/catalog' failed looks like one to me, but apparently it isn't. You should look into the xmlcatmgr documentation, or talk to the port maintainer, to find out what that warning means and if it is important. I didn't find anything pertinent in the (minimal) installed documentation. Just taking the xmlcatmgr message at face value, it looks as if some addition that linuxdoc intended to make in some catalog did not get done. Anything following that step in the postinstall script may also not have gotten done. My gut suspicion is that there is something wrong with the linuxdoc postinstall script -- or perhaps linuxdoc has an unstated dependency which I don't happen to have installed -- rather than something wrong with xmlcatmgr. PR time, I guess :( ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
pkg_add goofiness in 7.0
The first time I tried to add linuxdoc-1.1_1.tbz to a new-from-CD 7.0 installation, it complained about a missing dependency that was on the other CD. OK, I switched CDs and installed that, then switched back and retried linuxdoc-1.1_1.tbz, and it gave me some sort of error about an unbalanced add operation. (I didn't try to copy down all the details, figuring instead to retry with a script(1) active so as to capture them.) Upon that retry, it now tells me that the package is already installed, even though the prior attempt failed. What is going on? # uname -a FreeBSD fbsd70.uucp 7.0-RELEASE FreeBSD 7.0-RELEASE #0: Sun Feb 24 19:59:52 UTC 2008 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC i386 # ls -l linuxdoc* -r--r--r-- 1 root wheel 9156 Feb 24 08:18 linuxdoc-1.1_1.tbz # pkg_add -Kv linuxdoc* Requested space: 36624 bytes, free space: 774981632 bytes in /var/tmp/instmp.3DBbHN pkg_add: package 'linuxdoc-1.1_1' or its older version already installed pkg_add: 1 package addition(s) failed ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: pkg_add goofiness in 7.0
[trying to install linuxdoc-1.1_1.tbz] it gave me some sort of error about an unbalanced add operation. (I didn't try to copy down all the details, figuring instead to retry with a script(1) active so as to capture them.) Upon that retry, it now tells me that the package is already installed, even though the prior attempt failed. ... # ls -l linuxdoc* -r--r--r-- 1 root wheel 9156 Feb 24 08:18 linuxdoc-1.1_1.tbz # pkg_add -Kv linuxdoc* Requested space: 36624 bytes, free space: 774981632 bytes in /var/tmp/instmp.3DBbHN pkg_add: package 'linuxdoc-1.1_1' or its older version already installed pkg_add: 1 package addition(s) failed No idea, you'll have to recreate the failure and show us. Which failure are you referring to? The original one with the unbalanced add message, or the new one where it claims the package is already installed even though the previous installation reportedly failed? I can recreate the second one any number of times, but absent some specific suggestion it's not going to produce any more output than shown above. (I'm already specifying -v.) Short of wiping the drive and starting completely over, I have no idea how to go about reproducing the original failure without first fixing the newer one. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: pkg_add goofiness in 7.0
What you showed does not indicate a failure. If you are saying that the package wasn't actually installed completely, then pkg_delete it and retry. I am not saying that the package was installed incompletely, incorrectly, or something else because I don't know which of those applies. *The package installation itself* threw an error message, with no instructions for recovery, and left the package database corrupted (incorrectly showing the package as installed). pkg_delete was only partially successful, perhaps because it was unable to completely clean up the corruption. Now what? # pkg_delete -v linuxdoc-1.1_1 Trying to remove dependency on package 'xmlcatmgr-2.2' with 'textproc/xmlcatmgr' origin. Trying to remove dependency on package 'iso8879-1986_2' with 'textproc/iso8879' origin. Change working directory to /usr/local Delete file /usr/local/share/sgml/linuxdoc/README Execute '/usr/local/bin/xmlcatmgr -sc /usr/local/share/sgml/catalog.ports remove linuxdoc/catalog' xmlcatmgr: enabling compatibility mode; removing ALL matching entries xmlcatmgr: no matching entry for `linuxdoc/catalog' of any type pkg_delete: unexec command for '/usr/local/bin/xmlcatmgr -sc /usr/local/share/sgml/catalog.ports remove linuxdoc/catalog' failed Delete file /usr/local/share/sgml/linuxdoc/catalog Delete file /usr/local/share/sgml/linuxdoc/freebsd-1.0.dtd Delete file /usr/local/share/sgml/linuxdoc/freebsd-1.1.dtd Delete file /usr/local/share/sgml/linuxdoc/linuxdoc.dec Delete file /usr/local/share/sgml/linuxdoc/original.dtd Delete directory /usr/local/share/sgml/linuxdoc Change working directory to . pkg_delete: couldn't entirely delete package (perhaps the packing list is incorrectly specified?) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: pkg_add goofiness in 7.0
I believe this is the same as the error message I saw originally (when I had not specified -v, so it wasn't buried among a pile of other stuff): xmlcatmgr: unbalanced arguments for `add' action # pkg_add -Kv linuxdoc* Requested space: 36624 bytes, free space: 774946816 bytes in /var/tmp/instmp.CfA0bH Package 'linuxdoc-1.1_1' depends on 'xmlcatmgr-2.2' with 'textproc/xmlcatmgr' origin. - already installed. Package 'linuxdoc-1.1_1' depends on 'iso8879-1986_2' with 'textproc/iso8879' origin. - already installed. extract: Package name is linuxdoc-1.1_1 extract: CWD to /usr/local extract: /usr/local/share/sgml/linuxdoc/README extract: /usr/local/share/sgml/linuxdoc/catalog extract: execute '/usr/local/bin/xmlcatmgr -sc /usr/local/share/sgml/catalog.ports add linuxdoc/catalog' xmlcatmgr: unbalanced arguments for `add' action pkg_add: command '/usr/local/bin/xmlcatmgr -sc /usr/local/share/sgml/catalog.ports add linuxdoc/catalog' failed extract: /usr/local/share/sgml/linuxdoc/freebsd-1.0.dtd extract: /usr/local/share/sgml/linuxdoc/freebsd-1.1.dtd extract: /usr/local/share/sgml/linuxdoc/linuxdoc.dec extract: /usr/local/share/sgml/linuxdoc/original.dtd extract: CWD to . Running mtree for linuxdoc-1.1_1.. mtree -U -f +MTREE_DIRS -d -e -p /usr/local /dev/null Attempting to record package into /var/db/pkg/linuxdoc-1.1_1.. Trying to record dependency on package 'xmlcatmgr-2.2' with 'textproc/xmlcatmgr' origin. Trying to record dependency on package 'iso8879-1986_2' with 'textproc/iso8879' origin. Package linuxdoc-1.1_1 registered in /var/db/pkg/linuxdoc-1.1_1 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Mac osX drivers
I know I keep asking about drivers, but what about Mac drivers? I understand that Mac osX is based fairly well on BSD, so would the drivers be portable? Last I heard, MacOs X userland was based on FreeBSD but the MacOS X kernel was Mach. The part of a driver that deals with the hardware might be portable, but it doesn't seem so likely for the part that deals with the OS. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: List replies
There is a touching concern for newbies in all this, which is out of step with the somewhat edgy aspect of FreeBSD that most of you seem to embrace in other connections. And the bottom line is that most newbies end up elsewhere. If making FreeBSD more popular is a priority, there is long list of issues that are more vital than this detail that we are discussing in this thread. Given that the project is run almost entirely by volunteers, I suppose some disconnects are inevitable. One example: it's considered very important for questions@ to be newbie-friendly, to the point of accepting posts from non-subscribers, but it's *not* considered important to tune up sysinstall to prevent botches like installing a system that can't be booted because it has no kernel. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
dependencies in portmaster
I am trying to figure out what-all would happen if I were to install
a particular port. IOW I want to do something like
# portmaster {some set of options} name-of-port
and have it report something along the lines of
name-of-port vn #.## requires:
port status
--
dependency-1 OK
dependency-2 need vn 2.22, current 1.05
dependency-3 not installed
...
dependency-2 vn 2.22 requires:
port status
--
dependency-4 OK
dependency-5 need vn 5.03, current 4.57
...
dependency-3 vn #.## requires:
port status
--
dependency-6 not installed
...
I do not want it to actually build or install anything.
If I am understanding the portmaster manpage correctly this is close
to what -n would do, but I don't even want it to do 'make config' --
I just want a report of what would have to be added or upgraded in
order to install the port in question. (I imagine portmaster has
to already be collecting this sort of information internally, the
question is how to get it reported externally.)
BTW I am looking for a solution that does not involve portupgrade,
because I do not have portupgrade installed and before attempting
to install it I would want to see this sort of report regarding it.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: dependencies in portmaster
If you truly are just looking at the dependency list and do not wish to have make do anything, wouldn't this do the trick: http://portsmon.freebsd.org/portdependencytree.py Not quite, because it doesn't show which of the dependencies I have already got installed, and which of those would need to be updated. Secondly, I would have needed to know it existed :) I guess you would need to have an up to date ports tree for this to be accurate. Not a problem in this case. The whole point is to find out what-all I would be stuck with building in order to build one particular port after updating the tree a few days ago. Why not just rebuild everything? Because I'm not willing to attempt a rebuild of OpenOffice -- that was a collosal PITA the first time -- nor the xorg migration; I figure those are better accomplished by a clean install once the Mall's 7.0 CD set is ready. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: setting X11BASE
I'm finding it especially interesting that /etc/make.conf,
which to judge from its location is part of the base, depends
on a setting from something in the /usr/ports tree.
Well, actually it doesn't. What gives you this impression?
Paul Schmehl reported where LOCALBASE is set:
in /usr/ports/Mk/bsd.port.mk
Now I'm being told to add this:
X11BASE=${LOCALBASE}
to /etc/make.conf, so that /etc/make.conf needs LOCALBASE to be
set in order to set X11BASE correctly. Is that not a dependency?
You assume make(1)'s variable assignment is done on encounter base
at runtime. It isn't:
# echo LOCALBASE=/usr/local /tmp/foo.mk
# echo 'X11BASE=${LOCALBASE}' /etc/make.conf
# make -f /tmp/foo.mk -V X11BASE
/usr/local
# echo LOCALBASE=/tmp /tmp/foo.mk
# make -f /tmp/foo.mk -V X11BASE
/tmp
For your academic interest:
gzcat /usr/share/doc/psd/12.make/paper.ascii.gz|$PAGER
I know perfectly well how make works. The point is, if we have
X11BASE=${LOCALBASE}
in /etc/make.conf, X11BASE is going to be set *correctly* during
any particular execution of make only if LOCALBASE is set somewhere
in the makefiles that are processed during that execution of make.
If we run make under conditions that *don't* involve processing
/usr/ports/Mk/bsd.port.mk -- such as when building something
that isn't a port -- X11BASE is going to be *wrong* (unless a
definition gets provided somewhere else, as in your examples).
IOW adding this line to /etc/make.conf creates a dependency on
/usr/ports/Mk/bsd.port.mk, and that seems undesirable. Would
it not be better to put it somewhere under /usr/ports/Mk or
/usr/local/etc, rather than polluting the base with a ports-ism?
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: setting X11BASE
LOCALBASE is /usr/local unless you've changed it (but then you would already know what it was if you had.) You can find its value in /usr/ports/Mk/ grep LOCALBASE?= /usr/ports/Mk/* /usr/ports/Mk/bsd.port.mk:LOCALBASE?= /usr/local Aha! In case it matters, I have not upgraded to the modular Xorg, and would prefer not to go through all that. It ain't broke ... ... if you want to stay with the old system, you're probably going to need to put USE_NONDEFAULT_X11BASE?=/usr/X11R6 in your make.conf file to keep your ports from breaking in interesting ways ... Aha! again. Read /usr/ports/UPDATING carefully before proceeding. I did, but only as far back as the last time I updated, and I skipped entries which were identified as affecting ports I haven't installed or don't use ... including the modular xorg which I'm trying to avoid. (I figure it can wait until I do a clean install, on a different machine, using 7.0 when it comes out.) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: setting X11BASE
* What is the value of LOCALBASE? I'm not finding any
definition, or other reference, in /etc/make.conf.
Just set it to ${LOCALBASE} verbatim. Not what you think is the
value of the variable LOCALBASE but the word ${LOCALBASE}.
Academic interest :)
I'm finding it especially interesting that /etc/make.conf, which
to judge from its location is part of the base, depends on a setting
from something in the /usr/ports tree.
In case it matters, I have not upgraded to the modular Xorg, and
would prefer not to go through all that. It ain't broke ...
True, but your portstree is now 'broken', because support for how
it used to work is being phased out, like the whole X11BASE thing.
I think if you upgrade anything that depends on xorg, you'll find
dependencies being pulled in that are part of the modular xorg,
unless you really know what you're doing.
... which is why I am trying to install portmaster before doing
anything else. I may try pkg_tree also, if it doesn't lead into
a dependency maze of its own.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: setting X11BASE
* What is the value of LOCALBASE? I'm not finding any
definition, or other reference, in /etc/make.conf.
Just set it to ${LOCALBASE} verbatim. Not what you think is the
value of the variable LOCALBASE but the word ${LOCALBASE}.
Academic interest :)
I'm finding it especially interesting that /etc/make.conf,
which to judge from its location is part of the base, depends
on a setting from something in the /usr/ports tree.
Well, actually it doesn't. What gives you this impression?
Paul Schmehl reported where LOCALBASE is set: in /usr/ports/Mk/bsd.port.mk
Now I'm being told to add this:
X11BASE=${LOCALBASE}
to /etc/make.conf, so that /etc/make.conf needs LOCALBASE to be set
in order to set X11BASE correctly. Is that not a dependency?
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: DJ500 dead after = 16 years.
Guys, I need some input about what kind of _new_ printer to buy for my desktops. I'd like to hang the printer off my FBSD box; my Ubuntu platform is probably too far away. At least 3 meters. A few months ago I got a Samsung ML-2571N for well under $100 at Fry's. It is small, light, fast; has a built-in 10/100 network port, handles PostScript, and speaks native lpr What is lpr? see man lpr and its see also entries. Usually printers speak Post Script or PCL printer command language As I said, this one speaks PostScript. in which case you need a driver. Not in this case :) LPD, LPRng, and CUPS are different spooling systems. Did you attach the printer to a computer or is acting as a free standing printer server. It is freestanding. There is a lpr driver by Brother for Linux. Brother and Canon have binary blob drivers. Did you use that driver may be? I didn't install any drivers, just added it to printcap and hosts: lp|Samsung ML-2571N PostScript network printer:\ :sh:\ :rm=ml2571n:sd=/var/spool/output/ml2571n:lf=/var/log/lpd-errs: 192.168.200.201 ml2571n and it Just Works (TM). I suppose apps like OpenOffice might need to be told it is a PostScript printer. If I needed to do that, and didn't have anything more specific handy, I would just configure it as a LaserWriter NTX. I have yet to find a monochrome PostScript printer for which that does not work well enough. Does anyone know if those binary blobs can be useful for anything on FreeBSD. They appear to be wrappers for standard Ghost Script drivers. I noticed also that some drivers for older Brother printers are removed from Ghost Script 7.0 and 8.0. AFAIK GS is only needed if you want to print PostScript files on a printer that does not contain a PostScript interpreter. It's not needed when dealing with a PostScript printer. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD Linux distro
When defining the differences to my clients as to windows, Linux, and FreeBSD I use a 60's model VW beetle for windows ... Sheesh! What did VW do to you to deserve an insult like that? I still see the occasional beetle on the roads. I doubt that would be the case if they had to be rebooted a couple of times a day. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
setting X11BASE
After updating with portsnap, I am getting an insufficiently-helpful
error message:
On FreeBSD before 6.2 ports system unfortunately can not set
default X11BASE by itself so please help it a bit by setting
X11BASE=${LOCALBASE} in make.conf. On the other hand, if
you do wish to use non-default X11BASE, please set variable
USE_NONDEFAULT_X11BASE.
* Am I correct in *guessing* that make.conf refers to
/etc/make.conf?
* What is the value of LOCALBASE? I'm not finding any
definition, or other reference, in /etc/make.conf.
* How do I figure out whether I should set USE_NONDEFAULT_X11BASE?
* Why does it even need this? The port I am trying to install
ATM (portmaster, to get a handle on the dependency maze) has
nothing to do with X11.
In case it matters, I have not upgraded to the modular Xorg, and
would prefer not to go through all that. It ain't broke ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: DJ500 dead after = 16 years.
Guys, I need some input about what kind of _new_ printer to buy for my desktops. I'd like to hang the printer off my FBSD box; my Ubuntu platform is probably too far away. At least 3 meters. A few months ago I got a Samsung ML-2571N for well under $100 at Fry's. It is small, light, fast; has a built-in 10/100 network port, handles PostScript, and speaks native lpr (so you don't need to bother with CUPS). I am still on the original 1000-page starter cartridge. Replacements are rated 3000 sheets; I haven't priced them. That's black only. The cheapest color-capable networked PostScript printer I've found so far is the Xerox 6130N, for which I've been quoted $375 including $380 worth of cartridges (C, M, Y, K @ $95 each) -- Xerox seems to have some promotional pricing this month. IIRC the color cartridges are rated 1900 sheets and the black 2500. This one is also supposed to handle lpr natively. While I haven't got one (yet), I figure it is almost guaranteed to be good -- Xerox do not make junk. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Mounting a read-only md FS as read-only
does mount_ext2fs support readonly at all? The manpage implies that it does: The options are as follows: -o Options are specified with a -o flag followed by a comma sepa- rated string of options. See the mount(8) man page for possible options and their meanings. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
wine: notepad OK, others not
I have installed wine-0.9.20 from ports, and there's a Win98 FAT32 slice mounted on /windoze % grep -w windoze /etc/fstab /dev/ad0s1/windozemsdosfsro00 If I run Notepad, like this, it seems to work % wine /windoze/WIN98/NOTEPAD.EXE but if I then try to run Write: % wine /windoze/WIN98/WRITE.EXE the window title bar says Wordpad and several capabilities (like Save) don't work. Meanwhile the *real* Wordpad doesn't even start: % wine /windoze/PROGRA~1/ACCESS~1/WORDPAD.EXE err:module:import_dll Library MFC42.DLL (which is needed by LZ:\\windoze\\PROGRA~1\\ACCESS~1\\WORDPAD.EXE) not found err:module:LdrInitializeThunk Main exe initialization for LZ:\\windoze\\PROGRA~1\\ACCESS~1\\WORDPAD.EXE failed, status c135 but MFC42.DLL does exist, in what I think is the usual place: % find /windoze -name MFC42.DLL -ls 3536377 1948 -rwxr-xr-x 1 root wheel 995383 Apr 23 1999 /windoze/WIN98/SYSTEM/MFC42.DLL How do I fix this? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Mounting a read-only md FS as read-only
I'm trying to mount an ext2fs image (in a file), following sec. 17.13.2 of the Handbook and the mdconfig(8) manpage. Since I don't want to change the image, just examine it, I specified -o readonly to mdconfig and the equivalent to mount. Is there some reason why this should not work? The backing file, and the mountpoint, do exist. # ls -ld [filename] -rw-r--r-- 1 perryh perryh 104857600 Mar 16 2007 [filename] # mdconfig -a -t vnode -f [filename] -o readonly md0 # ls -ld /dev/md0 /[mountpoint] crw-r- 1 root operator1, 15 Nov 24 21:17 /dev/md0 drwxr-xr-x 2 perryh perryh512 Feb 17 21:57 /[mountpoint] # mount -r -o noexec -t ext2fs /dev/md0 /[mountpoint] mount_ext2fs: /dev/md0: Read-only file system # mount -o ro,noexec -t ext2fs /dev/md0 /[mountpoint] mount_ext2fs: /dev/md0: Read-only file system # /sbin/mount_ext2fs -o ro,noexec /dev/md0 /[mountpoint] mount_ext2fs: /dev/md0: Read-only file system # mdconfig -l -u 0 md0 vnode 100M /[mountpoint] # uname -a FreeBSD fbsd61 6.1-RELEASE FreeBSD 6.1-RELEASE #30: Mon Jan 1 23:01:34 PST 2007 [EMAIL PROTECTED]:/usr/src/sys/i386/compile/GENERIC i386 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: projectm questions
Linking CXX shared library libprojectM.so /usr/bin/ld: cannot find -lGLEW *** Error code 1 ... After a bit of poking around I found ln -s /usr/local/lib/libGLEW.a /usr/lib/libGLEW.a fixed it. I also had to do ln -s /usr/local/lib/libftgl.a /usr/lib/libftgl.a ln -s /usr/local/lib/libfreetype.a /usr/lib/libfreetype.a Well that's alright for a fix but Question 2: what do I have to do to get that to work automatically? Lose the symlinks, and instead figure out how to add -L/usr/local/lib to the link command line so that the linker looks for libs there as well as in /usr/lib. You might find the FreeBSD porter's handbook helpful. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: cron to attach a gz file
I know I can use mail -s logfile /var/log/httpd_access.log in cron to email the content of a log file to a particular email address but how do I make that log file a binary attachment (*.gz)? gzip -c /var/log/httpd_access.log | uuencode httpd_access.log.gz | mail -s logfile [EMAIL PROTECTED] If you want an actual MIME attachment, see /usr/ports/mail/nail ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Network configuration in FreeBSD
You need to set the default gateway in /etc/rc.conf. Without a default gateway, you will need to add a default route with the route command. Without a route your machine will only be able to ping itself. Unless something has changed dramatically -- and fairly recently -- a machine that knows its own IP address and netmask should be able to ping anything on the same subnet as itself (an interface being implicitly a route to any other IP address on the same subnet). ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: mutt??
Giorgos Keramidas [EMAIL PROTECTED] wrote: On 2007-12-27 11:02, Gary Kline [EMAIL PROTECTED] wrote: I've been trtryinng to rebuild everything on tao to get my i810 graphics working. Somehow, mutt bbroke. It seems to break with something undefined in perl5.8. Anybody know what this is: Undefined symbol __sbmaskrune ? That's odd. The mutt-devel port (which I am using to type and post this message) does not seem to depend on Perl: Last time I recall seeing a symbol name ending in rune it had something to do with handling charsets and/or locales. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD for Sony Playstation3?
... will FreeBSD.org consider porting FreeBSD to Sony Playstation3? ... NetBSD works like a charm on Sony Playstation2 http://www.netbsd.org/ports/playstation2/. and my guess will be that NetBSD 4.0 which is supposed to be released about the same time as FreeBSD 7.0 will work on Playstation3. The IBM Cell processor in the PS3 is unique beast, similar in many ways to a PPC970 but with enough subtle (and some not-so-subtle) differences that the port would likely need to be overseen by someone familiar with such undertakings. This is not to predict that NetBSD 4.0 will or won't support it -- that could reasonably be asked on a NetBSD list -- but be aware that it may turn out to be a bigger job than one might initially expect. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: K3b
BTW: I really gets annoyed when people say by the way it works in Linux, Windows, Solaris or whatever. What is that suppose to mean? When someone describes a problem getting a certain hardware setup to work as desired in FreeBSD, and reports that it works in some other OS, I would take it to mean that one need not suggest testing the hardware, since it is already known to work properly in the other environment. For someone who happens to be familiar with whatever other OS was mentioned, it may also serve as an example of the sort of operation that the poster was attempting to achieve in FreeBSD. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: hidden disk geometry on Compaq Presario V2000
Lorin Lund [EMAIL PROTECTED] wrote: I have a Compaq Presario Notebook in the V2000 series. I just replaced the hard drive because the original was getting disk errors. I have a WD Scorpio 120 GB. When I try to load FreeBSD I get an error message when I get to the partition the disk stage. It says my disk geometry is wrong. It says I need to use whatever numbers my BIOS uses. But my BIOS doesn't show the disk geometry numbers anywhere I can see. How can I proceed? How can I find out what disk geometry to use? One method, which I think may be mentioned in the Handbook, is to boot the Windows install CD (that presumably came with the Presario) and use its fdisk to create a small partition. You don't need to actually install Windows, just create a partition as if you were going to install it. Then boot the FreeBSD CD and sysinstall will figure out the geometry from the Windows master boot record. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: add application names to the drop-down menu??
Can anybodytell me how to add apps to the Gnome drop-down menu beneath the string Applications? (Upper-left-hand corner) Try using deskutils/alacarte ... x11-wm/wmconfig is another possibility. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD 6.2 on J7F4 locks up
I've now also experienced lock ups when building world in in single user mode, without using the nics. So the problem might not only be the nics. Cany anyone confirm this? ACPI, SATA, AC97 and USB were disabled when the lock up occurred. The canonical suspect would be flakey memory. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
How to set up a network-attached printer
Where would I find a specific method for setting up a Samsung ML-2571N network-attached PostScript printer in FreeBSD 6.1? I'm hoping for something less generic than what I've found in the handbook. It just works from MacOS X, as did the old LaserWriter IIf that the Samsung replaced, so I suppose one approach would be to use the Mac as a print server; but I would prefer to print from FreeBSD directly so that the Mac does not need to be up in order to print from the FreeBSD machine. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: shooting oneself in the foot with ldconfig -v
The previously configured directory list was fully populated, so effectively there should have been no change as the previously configured directories were untouched and I specified no additional pathnames. ... Are you saying that by specifying -v I no longer satisfied the no parameters are given clause and ended up in a default place in the logic? That wasn't actually what I was saying, but after checking the source code it turns out you are right and that is exactly what happens. ... IMHO a verbose switch shouldn't change behavior; it should just spam the console a lot. True. Current behavior sounds like, at best, a LOLA violation. Perhaps the OP would consider submitting a PR. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Hello!
From a pilot's point of view: FreeBSD is an F-4 Phantom. Mac is a P-38 Trainer. Windows is a DC-10. (with a hydraulic leak) Nah, a pig. See RFC 1925 and/or Oliver Fromme's .sig. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
OT: good vs evil (Re: vlc won't play region encoded DVDs)
... another story in the neverending race between good and evil. ;-) For some reason, that reminded me of a John Byrom quote (which is likely better known than its author): http://www.born-today.com/Today/d09-23.htm ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
X + WM != GUI? (Re: Convince me, please! - too much about GUI)
Wojciech Puchar [EMAIL PROTECTED] wrote: i don't use GUI. it takes a lot and gives nothing. i use both text and graphic (X) based apps and no gui. i use fvwm2 with my config, there are plenty of nice other wm's good for that. I am not following this. If (X.org + some WM) is not a GUI, how would you define * a GUI * X.org + some WM ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: what way to update named?
Are you sure your run the make with sufficient priviledges? I think this was it. I originally used sudo but second time I did it as su and it went very well. Thank you! If sudo to root does not give the same privs as su to root, I'd guess sudo is either buggy or not configured properly. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: non-interactive dump
Is there a way to tell dump to do it's working without it asking Is the new volume mounted and ready to go?: (yes or no) everytime it changes mount points? How else can it tell when you've swapped in new media? If it automatically continued it would just overwrite the previous segment. In principle, when dumping to a sequence of volumes on a removable device, it could watch for the device to become not-ready and then ready again. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: The worst error message in history belongs to... BIND9!
If one is going to require the installation of something that may not be part of a base system, that something might as well be bash :) Except that bash requires all the icky GNU utilities to build so you have to GNUify your system. And perl doesn't? It was GPL last I knew. The second you put in gmake, gmake requires iconv, readline and all the other nasty libraries, and from that point on if you build something you never know if it's going to link in to one of those libraries. ... This can cause major problems for commercial users. How? Last I heard, the *L*GPL only requires making the *library* source available (and that only if the library has been modified). It doesn't extend to the using application. I'd love for someone to modify the gmake port to have a variable you can set that would build all the GNUified dependency libraries, build and install gmake and statically link in all it's GNUified libraries, then remove all the GNUified libraries. Or, change all the gnu ports to install into something like /usr/local/gnu or /usr/local/gpl instead of straight into /usr/local. You'd still have the gnu libs when needed, but without having them included in normal search paths. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: The worst error message in history belongs to... BIND9!
This is actually just the difference between sh and bash ... differences in, say, arithmetic handling and loops can sometimes mean rewriting parts of shell scripts depending on whether it is going to run in BSD or Linux. That's a major argument for doing things in python or perl as they are consistent across all platforms ... If one is going to require the installation of something that may not be part of a base system, that something might as well be bash :) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: umount -f
1. If I use umount -f /dev/ad4s1a to forcefully umount a file system, does this jeopardize the integrity of said file system? Like...will it jerk the run out from under a process in the middle of a disk write, thus leaving a half written file, or will it wait until the write is complete? (I guess this would largely depend on the disk controller?) I don't believe there are any guarantees if your -f it. The filesystem will probably be OK, but I would expect files to get corrupt. Shouldn't happen, if it does it's a bug. umount -f should not corrupt the filesystem, IOW fsck of an FS immediately after umount -f should not report anything more serious than some homeless files (if the last link had been removed, but the inode had not been released because some process still had it open). However data corruption seems likely if it yanks the rug out from under a process that has, say, written part of a transaction to a database file. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: laser printer - which one?
I've chosen Samsung ML-2571N. It was pretty cheap for it's features: 400 Mhz CPU, 32MB, USB 2.0, parallel, ethernet, PS3, PCL6. I haven't tried it with BSD / Linux, but it prints slow and well under Windows. Slow? I got one of those recently, to replace an old LaserWriter IIf that seems to have died, and have been using it from a Mac via Ethernet. It was a drop-in replacement, and much faster. Someday I will get around to setting it up on FreeBSD :) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: OT: Re: The worst error message in history belongs to... BIND9!
Favourite worst written error message in history: Keyboard not found. Press F1 to continue. I have always loved this one!! Who made that up!? Someone at IBM. That's what the original IBM PC, PC-AT, and (presumably) PC-XT displayed if the keyboard was dead or not plugged in. It was probably a case of modular code: any problem in POST would display a message and return a fail status, and the generic code would append Press F1 to continue. and wait. Not a bad idea at all -- certainly better than blindly trying to boot the machine without giving the operator a chance to decide what to do about the problem -- but this particular combination does have a chicken- egg aspect :( ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: connecting user root with ssh
you are warned, do not allow SSH to your box with user root at all. ... Having root logon enabled remotely is just asking for trouble. The O.P. might be interested in knowing *why* allowing remote root login is considered unwise: * The name root is very well known. * If root can log in remotely, a cracker need only guess root's password to obtain root access. * If root cannot log in remotely, a cracker has to guess three things to obtain root access, instead of just one: + A valid username which is in the wheel group; + That user's password; + The root password. This at least doubles the difficulty of a brute-force attack: even if a suitable username were obvious, there would still be two passwords to be cracked. It can be made even tougher by having only one username (other than root) in the wheel group, choosing that name as if it were a password, and not allowing it to be externally known (e.g. never using it for mail). ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Dell Inspiron 1501 Express PCI slot
Fed Ex delivered a Wistron Neweb CM9 Mini-PCI wireless nic to me ... U, now I find that it won't fit in the PCI slot in my Dell Inspiron 1501 because Dell no longer uses standard sized PCI slots. They call them Express PCI slots. Does anyone know of an adapter that will mate the one to the other? In principle, it should be possible to build a PCI-Express card containing a PCI-Express to PCI bridge chip and one or more PCI slots. I have no idea whether such a device is commercially available. In practice, the combination might not fit in the box, and/or the BIOS might not be set up to enumerate a PCI bus downstream of the PCI-Express bus. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Is FreeBSD simple enough for Novices, Will FreeBSD accept Office 98 + Publisher?
OpenOffice in OSX still isn't that great either because there still isn't a native (Aqua) build. I suspect the NeoOffice folks would be surprised to hear that :) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
can't print to disk from Firefox
I got an error when trying to print *to disk* from Firefox/1.5.0.6: Printer Error There was a problem printing because the paper size you specified is not supported by your printer. This message makes no sense at all when printing to disk, since there's no way for it to know what kind of printer I'll eventually send the file to. I am using the only available printer selection, PostScript/default, and the first paper selection in the list, letterSize (8.5x11 inch); and that size is certainly available on the printer I intend to use (which is elsewhere, hence the need to print to disk). There is no default paper size selection. I really don't care what paper size it formats the page for -- that can be fixed later if necessary. It is the content, not the formatting, that is important. It eventually fixed itself after much fumbling and several retries, but I have no idea what fixed it and I suppose it will probably unfix itself at some inconvenient time in the future. How do I fix it properly and permanently? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Append only directory ? Is this possible with unix permissions ?
Gore Jarold [EMAIL PROTECTED] wrote: I have a user whose home directory I would like to make append only. ... As someone else suggested, ACLs are likely the strongest way of handling this. On the other hand, if all that is needed is a way to make it a little tougher for said user to shoot him/herself in the foot, set noclobber in csh or tcsh might help. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RESEND: Re: BSDstats report for Mar 1st, 2006
suggested adding a prompt to sysinstall asking if ppl wanted to participate, and the response I heard was that someone basically needed to submit a patch ... anyone here know enough about sysinstall to do so? If considering work on sysinstall to improve the stats, how about fixing some of the pitfalls that drive away prospective new users? (IOW increase the actual number of installations rather than just the fraction that get reported.) Surely it should be possible to avoid lethal messes like a newly- installed system that has no kernel, or a partitioning that will not accommodate the set of components selected for installation (resulting in failure when /usr fills up). And no, I am not suggesting any substantial change to the overall look and feel of the UI. That would be too much work for, at best, a cosmetic improvement. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
What is an implicit destination
I go this auto-response after replying to a message on questions@ a few hours ago: From: [EMAIL PROTECTED] Subject: Your message to freebsd-questions awaits moderator approval Date: Sun, 04 Mar 2007 05:43:56 + Your mail to 'freebsd-questions' with the subject Re: BSDstats report for Mar 1st, 2006 Is being held until the list moderator can review it for approval. The reason it is being held: Message has implicit destination Either the message will get posted to the list, or you will receive notification of the moderator's decision. If you would like to cancel this posting, please visit the following URL: snip PLEASE NOTE! If you would like to post freely to the list, please subscribe first. If you post from multiple addresses, you can subscribe each address and go into the options page and select 'no mail' for all but one address. This will allow you to post without delay in the future. but I *am* subscribed! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Outlook With FreeBSD IMAP
How do I use the Cram-MD5 passwords with Outlook? Or do I have to go plain text? Off-topic for FreeBSD-Questions but I don't believe Outlook supports CRAM-MD5 out of the box. *Not* off-topic, the context being how best to configure Outlook for use with FreeBSD IMAP. One hopes something more secure than plain-text passwords can be made to work. My answer is Don't use Outlook. For anything. Period. but the OP may be stuck with it for some reason. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: ssh to VMS - terminal problems
yes - VMS only knows about DEC-compatible terminals. None of the *BSD console emulators do well enough to be usable on VMS. xterm supports ANSI color, VT220 emulation and UTF-8 There's an faq at http://invisible-island.net/xterm/xterm.faq.html ftp://invisible-island.net/xterm/ Which is fine if one doesn't mind having to fire up X. Another possible approach would be to run ports/sysutils/screen, which should provide a decent VT100 over just about anything with a terminfo better than dumb or unknown :) So, what was pcvt driver designed for? I understood from the man pages that it is supposed to be compartible with DEC function keys? Generating the escape sequences for the DEC function keys is one thing. Handling all the escape sequences that VMS throws at it is another. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: GMail [and other free email] and these lists?
space on my ancient PII. wow, PII these days.. is anybody still running on 486? :D I am. :P As a small router, I does it's work just fine. :D I've got a 486 FreeBSD-based GNATbox firewall, temporarily in retirement until I get around to switching DSL ISPs, but my entry in the geriatric-hardware-still-in-use dept. is a Sun-3/50, with SunOS 4.1.1_U1, as UUCP-connected mailhost :) The Sun has a goofy MTU problem though -- anyone been around long enough to remember what magic is needed to get one of those to work properly when talking through a 10Base-T hub? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: backup solution
Then I will install freebsd on the first disk and will use the two spare IDE-disks on the same cable as a geom-mirror. ... Please be aware that the ATA implementation of the VIA EPIA chipset isn't the greatest, especially when both are active at the same time. I've seen drive performance drop towards 5MB/s for WDC600/WDC800/WDC1200-grade drives which normally run at 40MB/s if you do something on the other channel as well. Dunno about that chipset in particular; some will perform much better with concurrent operations split across the two channels than on the same channel. You might be better off to have root and one of the mirror disks on ide0 and the other mirror disk on ide1 with the DVD. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
