Interaction between mpd and ipfilter/ipnat
I have a FreeBSD firewall which does packet filtering and NAT. The internal address range is 172.16.64.0/24. The only filtering is incoming on the external NIC, fxp0. The machine also runs mpd for remote access. By pure chance I was tailing ipf.log when I connected an XP laptop to the mpd service, and immediately I saw these: Mar 16 16:57:41 inchgower ipmon[61]: 16:57:40.923619 fxp0 @0:2 b 172.16.64.168,137 -> 172.16.64.200,137 PR udp len 20 96 IN Mar 16 16:57:42 inchgower ipmon[61]: 16:57:42.425811 fxp0 @0:2 b 172.16.64.168,137 -> 172.16.64.200,137 PR udp len 20 96 IN 172.16.64.168 is the address given out by mpd to the laptop. 172.16.64.200 is the Active Directory Domain Controller. I'm confused as to why ipf is seeing these packets coming in on fxp0. Surely what comes in is the GRE packet to the external NIC's address, this is then decapsulated and the embedded packet routed on. Why does ipf even see it, let alone block it? I would expect the source interface to be ng0, not fxp0. From the laptop I can ping and connect to internal machines, so most packets are not being blocked in this way. tcpdump also sees the packets coming in on fxp0, but I'm not convinced they are. I guess I can only really tell if I get the switch to copy packets to another port and monitor from there. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
FBSD5 and FBSD6 aren't too keen on the SFU/AD NIS server
My current situation is that I have a Solaris based NIS server and various client machines, including FreeBSD. A production FreeBSD 5 machine and a test FreeBSD 6 machine work just fine with it, except for a small glitch on the FreeBSD 6 machine: >speyburn# ypwhich >panther.internal.local >speyburn# ypwhich -m >ypwhich: can't find the master of ``: reason: No such map in server's domain The FreeBSD 5 machine correctly lists all the maps. I want to move to a NIS server provided by Microsoft's Services for Unix running on an Active Directory domain controller. FreeBSD 6 seems to work OK, though the error message changes slightly: >speyburn# ypwhich >axiom.internal.local >speyburn# ypwhich -m >ypwhich: can't find the master of `: reason: No such map in server's domain (only one quote rather than two). But it actually functions fine: >speyburn# id jhatfield >uid=115(jhatfield) gid=100(Domain Users) groups=100(Domain Users), 0(wheel) I can log in with no problem. Unfortunately the FreeBSD 5 machine does not work fine at all. Firstly it can only find one map: >banff# ypwhich >axiom.internal.local >banff# ypwhich -m >ypservers axiom And it can't convert login names to UIDs: >banff# id jhatfield >id: jhatfield: no such user If anyone has experience of pointing FreeBSD at an SFU/AD NIS server I'd like to know if they had this problem and if so how it was solved. I really want to move to the AD NIS so if I have to I'll replace the machine with one running FreeBSD 6, but I'd prefer not to have to. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
NIS client differences between FBSD5 and FBSD6?
Still trying to migrate our NIS from an old Sparc to a Services For Unix/Active Directory setup. AMD won't play so I'm using text files for that. Now to move forward to actually logging in A test machine running FreeBSD 6 seems to work OK, though "ypwhich -m" behaves strangely: >speyburn# uname -v >FreeBSD 6.0-RELEASE #0: Thu Nov 17 12:42:17 GMT 2005 [EMAIL >PROTECTED]:/usr/obj/usr/src/sys/GENERIC >speyburn# ypwhich -m >ypwhich: can't find the master of `: reason: No such map in server's domain >speyburn# ypwhich -m ypservers >axiom >speyburn# id jhatfield >uid=115(jhatfield) gid=100(Domain Users) groups=100(Domain Users), 0(wheel) But a production machine running FreeBSD 5 does not: >banff# uname -v >FreeBSD 5.1-RELEASE-p16 #1: Wed Apr 7 15:14:39 BST 2004 [EMAIL >PROTECTED]:/usr/obj/usr/src/sys/BANFF >banff# ypwhich -m >ypservers axiom >banff# id jhatfield >id: jhatfield: no such user The only change I made is to the nisdomainname line in /etc/rc.conf, followed by a reboot. All the ypxxx commmands seem to work OK, yet usernames can't be looked up. Why would this be? I really don't want to replace this machine with a FBSD6 machine just to fix this, though I will if I have to. Jim ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
amd doesn't like NIS maps from a Windows NIS server
We've been using NIS-based automounter maps for ages, using a Solaris NIS server. Some of our machines use autofs type maps, and some use amd, notably the FreeBSD boxes. As part of a move to single sign-on I've implemented a NIS server using Microsoft's Services for Unix installed on an Active Directory domain controller, and (painfully) created the automounting maps. The autofs-using machines are OK, but amd just doesn't want to play. It seems to be OK about the top-level "master" map, but not the ones referenced by that. As an experiment I've mixed file-based and NIS maps. My rc.conf contains: >amd_enable="YES" >amd_map_program="cat /usr/local/etc/amd.master" >amd_flags="-x all -D info -l syslog:local7" /usr/local/etc/amd.master contains: >/home amd.home >/mp /usr/local/etc/amd.mp >/net/usr/local/etc/amd.net >/users /usr/local/etc/amd.users When the machine starts I get: >Jan 11 11:52:30 speyburn amd[408]: /mp: disabling nfs congestion window >Jan 11 11:52:30 speyburn amd[410]: /users: disabling nfs congestion window >Jan 11 11:52:30 speyburn amd[411]: /net: disabling nfs congestion window >Jan 11 11:52:30 speyburn amd[409]: /home: disabling nfs congestion window >Jan 11 11:52:30 speyburn amd[407]: first time load of map /usr/local/etc/amd.mp >succeeded >Jan 11 11:52:30 speyburn amd[407]: /usr/local/etc/amd.mp mounted fstype toplvl >o >n /mp >Jan 11 11:52:30 speyburn amd[407]: first time load of map >/usr/local/etc/amd.use >rs succeeded >Jan 11 11:52:30 speyburn amd[407]: /usr/local/etc/amd.users mounted fstype >toplv >l on /users >Jan 11 11:52:30 speyburn amd[407]: first time load of map >/usr/local/etc/amd.net > succeeded >Jan 11 11:52:30 speyburn amd[407]: /usr/local/etc/amd.net mounted fstype toplvl >on /net >Jan 11 11:52:30 speyburn amd[407]: No source data for map amd.home >Jan 11 11:52:30 speyburn amd[407]: amd.home mounted fstype toplvl on /home So the file-based ones work fine, but the single NIS map, amd.home, does not. And yet if I do a "ypcat -k amd.home" I get exactly the same as when I was talking to the Solaris NIS server - but it just doesn't work. If I replace it with a file with the same contents, it works fine. Aaargh! I've looked at the source of amd but my C experience is 15 years old now and I'm struggling a bit! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Correct configuration of pam_winbind.so for login using AD accounts
I'm using a newly-installed FBSD 6 system to experiment with Single Sign-On to an Active Directory network. Samba is installed, the machine is joined to the domain, winbind seems to work fine, wbinfo -u lets me enumerate users OK. I'm trying to work out how to edit the files in /etc/pam.d to get pam_winbind to let me log on to the console using an AD account. Most of the Samba docs seems to be Linux-specific and the sample pam files don't match the ones in the FBSD 6 system. What I did was to edit /etc/pam.d/login: add "auth sufficient pam_winbind.so" as the penultimate line of the auth section, and the same in the account section. If I try to log in as an AD user on the console I get this in /var/log/messages: >Nov 23 15:30:36 speyburn pam_winbind[1330]: user 'INTERNAL+jhatfield' granted >access >Nov 23 15:30:36 speyburn pam_winbind[1330]: user 'INTERNAL+jhatfield' granted >access >Nov 23 15:30:36 speyburn winbindd[1324]: [2005/11/23 15:30:36, 0] >rpc_client/cli_pipe.c:cli_rpc_open_noauth(1700) >Nov 23 15:30:36 speyburn winbindd[1324]: rpc_pipe_bind failed >Nov 23 15:30:37 speyburn winbindd[1324]: [2005/11/23 15:30:37, 0] >rpc_client/cli_pipe.c:cli_rpc_open_noauth(1700) >Nov 23 15:30:37 speyburn winbindd[1324]: rpc_pipe_bind failed >Nov 23 15:30:37 speyburn login[1331]: setlogin(INTERNAL+jhatfield): Invalid >argument - exiting So I'm close but not there yet. As an aside, I'm confused as to the difference between what pam_winbind offers and what nss_winbind offers - I would have thought either of them would be adequate to provide login access. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Failures after upgrading perl
On Wed, 9 Feb 2005 10:07:59 - , in local.freebsd.questions you wrote: >I have a 4.7-based system I use as a mail gateway. Yesterday I >did a portupgrade of perl from 5.6.1 to 5.6.2. Today I find that >I have no incoming mail, due to mimedefang no longer functioning: Bang my head against a wall and repeat ten times: "I will read /usr/ports/UPDATING *before* I do a portupgrade, not after!". ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Failures after upgrading perl
I have a 4.7-based system I use as a mail gateway. Yesterday I did a portupgrade of perl from 5.6.1 to 5.6.2. Today I find that I have no incoming mail, due to mimedefang no longer functioning: >Feb 9 09:56:39 highland mimedefang-multiplexor[91186]: Slave 0 stderr: Can't >locate MIME/Base64.pm in @INC (@INC contains: >/usr/local/lib/perl5/site_perl/5.6.1 /usr/local/lib/perl5/site_perl/5.6.2/mach >/usr/local/lib/perl5/site_perl/5.6.2 >/usr/local/lib/perl5/site_perl/5.005/i386-freebsd >/usr/local/lib/perl5/site_perl/5.005 /usr/local/lib/perl5/site_perl >/usr/local/lib/perl5/5.6.2/BSDPAN /usr/local/lib/perl5/5.6.2/mach >/usr/local/lib/perl5/5.6.2 .) at >/usr/local/lib/perl5/site_perl/5.6.1/MIME/Words.pm line 85. BEGIN >failed--compilation aborted at /usr/local/lib/perl5/site_per But pkg_info shows p5-MIME-Base64 as present! And it is, but not on any of the directories on the @INC path: >highland# find /usr/local/lib -name Base64.pm -print >/usr/local/lib/perl5/site_perl/5.6.1/mach/MIME/Base64.pm >/usr/local/lib/perl5/site_perl/5.6.1/MIME/Decoder/Base64.pm If I try a portupgrade -f it seems to get confused about perl versions: >highland# portupgrade -f p5-MIME-Base64 >---> Reinstalling 'p5-MIME-Base64-3.05' (converters/p5-MIME-Base64) >---> Building '/usr/ports/converters/p5-MIME-Base64' >===> Cleaning for perl-5.6.2_2 >===> Cleaning for p5-MIME-Base64-3.05 >===> Vulnerability check disabled >===> Extracting for p5-MIME-Base64-3.05 >=> Checksum OK for MIME-Base64-3.05.tar.gz. >===> p5-MIME-Base64-3.05 depends on file: /usr/local/bin/perl5.6.1 - not >found >===>Verifying install for /usr/local/bin/perl5.6.1 in /usr/ports/lang/perl5 >===> Vulnerability check disabled >===> Extracting for perl-5.6.2_2 >=> Checksum OK for perl-5.6.2.tar.gz. >=> Checksum OK for BSDPAN-5.6.2.tar.gz. It seems to think it's dependent on perl 5.6.1, which isn't present, then wants to install 5.6.2, which is already installed. Do I have to remove and reinstall all the perl addons due to the upgrade of the perl version? And what's the "mach" directory for? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: ipnat and "udp consistent translation" (Skype related)
On Wed, 29 Sep 2004 13:45:40 +0100, in local.freebsd.questions you wrote: >Using Skype on a machine behind a FreeBSD 4.x firewall using >ipf/ipnat, if I try a file transfer I get "your connection is relayed" >which suggests that there are problems using "UDP hole punching" to >get a direct connection. The Skype help page sends you to: > >http://bgp.lcs.mit.edu/~dga/view.cgi > >where ipnat gets a "no" in the "udp consistent translation" column. I think this info must be out of date, and that ipnat really does do "UDP consistent translation". This is a bit of output from ipnat -l: >MAP 172.16.64.1613432 <- -> 195.74.141.61 5132 [159.148.187.95 27452] >MAP 172.16.64.1613432 <- -> 195.74.141.61 5132 [70.48.222.77 51689] >MAP 172.16.64.1613432 <- -> 195.74.141.61 5132 [80.131.15.67 24122] >MAP 172.16.64.1613432 <- -> 195.74.141.61 5132 [67.8.167.204 52284] >MAP 172.16.64.1613432 <- -> 195.74.141.61 5132 [24.201.154.49 57657] >MAP 172.16.64.1613432 <- -> 195.74.141.61 5132 [82.36.75.76 41765] >MAP 172.16.64.1613432 <- -> 195.74.141.61 5132 [139.91.190.109 4709] >MAP 172.16.64.1613432 <- -> 195.74.141.61 5132 [65.93.139.84 56670] >MAP 172.16.64.1613432 <- -> 195.74.141.61 5132 [210.221.94.233 5387] The machine running Skype is 172.16.64.16 and Skype's Options->Connections property page shows 13432 as the port number. I presume 5132 is the translated port number. Looks pretty consistent to me. So it was a red herring and I'll have to look elsewhere. jim ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
ipnat and "udp consistent translation" (Skype related)
Using Skype on a machine behind a FreeBSD 4.x firewall using ipf/ipnat, if I try a file transfer I get "your connection is relayed" which suggests that there are problems using "UDP hole punching" to get a direct connection. The Skype help page sends you to: http://bgp.lcs.mit.edu/~dga/view.cgi where ipnat gets a "no" in the "udp consistent translation" column. I also ran the "natcheck" utility from here: http://midcom-p2p.sourceforge.net/ on the firewall box itself (ie no NAT) I get: >Request 20 of 20... > >TCP RESULTS: >TCP consistent translation: YES (GOOD for peer-to-peer) >TCP simultaneous open:YES (GOOD for peer-to-peer) >TCP loopback translation: YES (GOOD for peer-to-peer) >TCP unsolicited connections filtered: YES (GOOD for security) > >UDP RESULTS: >UDP consistent translation: YES (GOOD for peer-to-peer) >UDP loopback translation: YES (GOOD for peer-to-peer) >UDP unsolicited messages filtered:YES (GOOD for security) but on a machine inside I get: >Request 4 of 20... >Request 5 of 20... >checkloopback connect: Invalid argument which doesn't look good. Googling didn't find anything so I was wondering if anyone else had experienced this and if so what their resolution was. It would be a shame to have to switch to a different firewall when ipf/ipnat is so easy to use and works so well for everything else, but at the same time I don't like the idea of someone else having to relay the Skype traffic unnecessarily. jim ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
New-style PHP and Sybase DB-Lib
Just looking at the Makefile for lang/php4-extensions and I see in the list of options SYBASE_CT but no SYBASE_DB. I think my existing PHP setup was compiled with SYBASE_DB, certainly phpinfo() reveals in the Sybase section: > $Id: dblib.c,v 1.160.2.1 2004/04/04 09:07:04 freddy77 Exp $ I also recall I did try SYBASE_CT last time I installed but had no success. So now I'm concerned that if I upgrade PHP I will lose Sybase connectivity. If anyone has upgraded and think they previously used DB-Lib I'd appreciate knowing whether everything was OK after the upgrade. jim ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Transparent proxying in 4.x with squid and ipf?
I used to do transparent http proxying with ipfw and Squid, but lost that when I switched from ipfw to ipf. The Squid port Makefile says: >#This option does not work on FreeBSD at the moment: ># >#--enable-ipf-transparent ># Enable Transparent Proxy support for IP-Filter systems (incl 3.0) ># (IPFilter headers are not currently installed to the base system, ># PRs ports/60700 and misc/44148 describe the problem; see ># http://www.squid-cache.org/Doc/FAQ/FAQ-17.html for information ># about how to do transparent proxying with ipfw) Is this ever likely to be possible again in 4.x or will I have to wait for 5-STABLE? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Growisofs and etc/crontab
On Wed, 21 Jul 2004 07:48:31 +0100, in local.freebsd.questions you wrote: >Thanks for the suggestion will check the paths at the beginning. The >crontab >I was using is the system one (/etc/crontab) and I did try specifying >the >whole path to the program as well. Not in this example you didn't. > The exact error message was >"growisofs: >not found". The command was "1 2 * * * root growisofs -dvd-compat -Z The path should be /usr/local/bin/growisofs. jim ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: mpd3.18
On Mon, 14 Jun 2004 20:14:23 +0100, in local.freebsd.questions you wrote: >I have a box that is having a 24th hour crash, I have narrowed it down >to mpd 3.18 as the cause, kill mpd and the box will hum right along for >days, restart mpd 24 hrs later it goes into kernel panic and reboots. >Anyone else have this problem?? (side note - swap file full is one msg) >As you can tell, I am a basic noob when dealing with this type of >problem. There is a mailing list for mpd - [EMAIL PROTECTED], I am sure they would like to hear about this. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Issues with large files on nfs-mounted filesystems?
I've made a large .mpg file on a Linux machine (because some tools, such as mplex, are newer than available in FreeBSD ports). Here's a directory listing: >-rw-r--r-- 1 jim users 4388444160 Jun 2 23:15 movie.mpg I want to burn it to DVD but the burner is on a BSD box, so I nfs-mount the /home partition. However when I look at the same file from FreeBSD 5.1 I get: >-rw-r--r-- 1 jim 100 93476864 Jun 3 00:15 movie.mpg (yes I know there is a time zone issue. I haven't worked out how to set it on Gentoo yet). The difference between the sizes is 4294967296, ie 2^32. Anyone know if this is an issue with the NFS implementations or whether the NFS protocols have a 32-bit size limit? Even copying the file from one machine to the other is proving a challenge! Trying to push from Gentoo to FreeBSD gets: >ftp> put movie.mpg >local: movie.mpg remote: movie.mpg >local: movie.mpg: File too large and trying to pull from the FreeBSD box gets: >ftp> get movie.mpg >local: movie.mpg remote: movie.mpg >227 Entering Passive Mode (192,168,100,2,128,64) >550 movie.mpg: File too large. Hm. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Max size of file on ISO filesystem?
I used to think that ISO filesystems were limited to a maximum size of 2 Gb for each file, but now I'm not sure sure. mkisofs from sysutils/mkisofs will not accept files over 2 Gb but the one in syutils/mkisofs-devel will without a problem. So if you install mkisofs-devel you can then use growisofs to make a DVD with a file over 2 Gb on it. The fun starts when you try to read the DVD! On a 4.9 machine, doing a "ls -l" shows the file size as negative, presumably just the result of using a signed 32-bit quantity to hold the size in bytes. However a 5.1 machine generates an error, something like "value too large for data type". I presume this is coming from the iso9660 filesystem layer. So there seems to be a discrepancy between what the latest mkisofs and the latest iso9660 filesystem code regards as acceptable. Which is right? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: IMAP server and client recommendations?
On Thu, 22 Apr 2004 05:42:00 +0100, in local.freebsd.questions you wrote: >Almost right, but not quite. You set up an IMAP server that stores >mail in the desired format, add the IMAP support to Outlook, and then >drag/drop the mail into the IMAP mailbox. There is no equivalent >client-side export needed. Not always possible, unfortunately. Outlook 2000 with a MAPI connection to an Exchange server lets you add a POP3 service but not IMAP. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: DVD -> SVCD with FreeBSD?
On Mon, 29 Mar 2004 16:13:18 +0100, in local.freebsd.questions you wrote: >On Mon, 29 Mar 2004, Jim Hatfield wrote: > >> I tried mencoder. The ports version won't produce mpeg2 (at > >Here's what I do: I decode the MPEG to YUV using mplayer, then >reencode back to MPEG using mjpegtools. Interesting - I read the man pages for mjpegtools and thought "what's all this yuv stuff?" thinking it was only of interest to someone using a capture device. It never occurred to me to use it as an intermediate file format. Thanks. jim ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
DVD -> SVCD with FreeBSD?
If anyone is successfully converting DVDs to SVCD (or MPEG2 formats in general) I'd appreciate knowing what tools they use. I have been fine producing MPEG4 files with mencoder, but I now have a networked video player (Hauppauge MediaMVP) which doesn't support MPEG4 but supports MPEG2 just fine. I tried mencoder. The ports version won't produce mpeg2 (at least not with libavcodec). The version from the mplayer site will, but the results look like "modern art". I got the same with transcode, though it's a while since I used it. ffmpeg won't even build on my AMD-Duron based 5.1 machine (though to be fair it will on my PIII Celeron 4.9 machine). The package of ffmpeg doesn't support mpeg2 encoding. mjpegtools has an mpeg2 encoder but not a decoder. It seems like no matter what I try there is some problem which blocks me. I would hate to have to revert to a Windows solution for this. jim ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: OT: transcode, ffmpeg, mencoder
On Mon, 8 Mar 2004 22:01:35 - , in local.freebsd.questions you wrote: >#convert a quicktime file into an mpg >mencoder -of mpeg -o example.mpg -oac mp3lame -ovc lavc -lavcopts >vcodec=mpeg1video example.mov >mencoder -of mpeg -o example.mpg -oac copy -ovc lavc -lavcopts >vcodec=mpeg1video >example.mov >mencoder -of mpeg -o example.mpg -oac pcm -ovc lavc -lavcopts >vcodec=mpeg1video >example.mov >mencoder -of mpeg -o example.mpg -ovc lavc -lavcopts vcodec=mpeg1video >-oac copy >example.mov > >I am able to get the video, on some of these, but never the audio. According to www.dvdrhelp.com, audio for vcd needs to be mpeg-1 layer 2, sampled at 44100 Hz and with a bitrate of 224 kbit/sec. If you use -oac lavc you should be able to use lavcopts acodec=mp2. The bitrate defaults to 224. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: S/KEY ftp logins
On Mon, 8 Mar 2004 15:31:50 - , in local.freebsd.questions you wrote: >Is there some way to tell if ftp logins are successfully using S/KEY or >falling back to cleartext? Is there some way to require S/KEY only? I believe the password prompt includes "required" if a static password would not be accepted. As I recall if you create /etc/skey.access then everything which is *not* mentioned in that file will require s/key. I think this also applies to shell logins so you need to be careful. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
(Re) mp4 spliter
On Wed, 25 Feb 2004 04:00:20 -, in local.freebsd.questions Tsu-Fan Cheng <[EMAIL PROTECTED]> wrote: >Hi freebsders, I have a movie file ended with mp4 and it's bigger than >700mb, does anybody know how I can split the file into 2 so I can make a >CD out of it? thanks a lot! > >Best Regards, :-) avisplit from the transcode port? Be warned though it has a *lot* of dependencies! ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
4.5 -> 5.1, rlogin works but rsh fails
I'm trying to use rdist to mirror a tree from a 4.5 machine
("speyburn") to a 5.1 machine ("banff"). So firstoff rsh has to work.
This is root-to-root.
I've set up the /root/.rhosts on the target 5.1 machine and also
on the 4.5 machine so I can check it both ways round.
Logged on as root on the 5.1 machine I can do "rsh speyburn ls"
and get a listing with no problems. But the other way round I
get "rshd: Login incorrect" and the auth.log file on the server
shows:
>Dec 19 14:36:29 banff rshd[74139]: [EMAIL PROTECTED] as root: permission denied
>(authentication error). cmd='ls'
But I can use "rsh banff" and I get logged on with no password
prompt.
It's not tcp wrappers since the hosts.allow is unchanged from
the as-installed version. Both .rhosts files are rw-r--r--.
It's got to be something simple but I can't see it.
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Any way to lock down disk errors?
Strictly speaking OT but the machine is running FreeBSD. While copying a file I got I/O errors. The console shows: >ad0: hard error cmd=read fsbn 31891359 of 31891359-31891486 status=59 error=40 >ad0: hard error cmd=read fsbn 31891231 of 31891231-31891486 status=59 error=40 Given that the disk is just under three months old, is it worth doing anything other than getting it replaced? I have no other disk big enough to old the data on it so unless the supplier sends me a replacement ahead of me returning the faulty one it will be a pain. I have enough space to empty the partition with the error in, but I couldn't find anything in newfs or fsck which would let me map out selected blocks or to do a full write test of each block and map out bad ones. Is there such a beast? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Confused about perl versions and perl modules
I'd love to understand the interaction between perl versions, FreeBSD versions and perl modules. Frinstance I have a 4.7 system with only the base system perl, and a number of perl modules from the ports collection. The perl modules seem to install themselves under /usr/local/lib/perl5/site_perl/5.005/ and put the man pages under /usr/local/lib/perl5/5.00503/man. Would this be different if I had perl 5.6 from ports installed and had done a "use.perl port" before installing them??? I note that on a 5.1 system under /usr/local/lib/perl5 there is both 5.00503 and 5.6.1 and site_perl and under site_perl there is 5.005 and 5.6.1. This seems kind of surprising given 5.x doesn't have a base system perl and the ports version is 5.6. I guess the real question is, if I move from base to ports version, do I have to uninstall and reinstall all the modules from ports? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Using poptop....
On Sat, 18 Oct 2003 05:52:57 +0100, in local.freebsd.questions you
wrote:
>It's really pretty easy. I beat my head on the wall about a year ago,
>so I
>don't recall where all the docs are, but this is what my configs look
>like:
>/etc/ppp/ppp.conf:
>pptp:
> set timeout 0
> set dial
> set login
> set ifaddr {IP address of internal interface} {IP address range
>xxx.xxx.xxx.xxx - xxx.xxx.xxx.xxx}
> set server /var/run/pptp_ppp_%d "" 0700
> enable mschapv2 mppe # <--- these two lines enable encryption
> set mppe * *#<---
OK thanks I'll try this one. I did eventually find an example which I
used and it did work but there were some errors logged. The
config I used included:
>enable mppe * stateful
in the ppp.conf file, but that generated the following in ppp.log
>Oct 20 12:52:38 field ppp[53548]: Command: loop: enable mppe * stateful
>Oct 20 12:52:38 field ppp[53548]: Warning: enable *: Invalid command
>Oct 20 12:52:38 field ppp[53548]: Warning: enable *: Failed 1
>Oct 20 12:52:38 field ppp[53548]: Warning: enable stateful: Invalid command
>Oct 20 12:52:38 field ppp[53548]: Warning: enable stateful: Failed 1
The connection did appear to be encrypted though, at least the XP
client said it was.
Jim
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Using poptop....
On Fri, 17 Oct 2003 16:00:00 +0100, in local.freebsd.questions you wrote: >If anyone can give me a recipe for setting up poptop >I'd be very grateful. I want to run it on a machine which >is a gateway/firewall. One NIC has a public address and the other >is on a private network, ie 192.168.1.x. I want to allow XP >clients to connect into the private network. > >I found the man pages a bit terse(!) and they seem to assume >that kernel ppp will be used whereas AIUI the port is built to >use userland ppp. Answering my own post, I know, but this: http://heyer.supranet.net/pptp/ is very useful, though it assumes that userland ppp can't do encryption, which I'm hoping is no longer true. If encryption is possible, then this: http://lab.digitol.net/openbsd-poptop.html albeit for OpenBSD, is pretty useful too. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Strange pkg_ behaviour
On Fri, 17 Oct 2003 17:55:57 +0100, in local.freebsd.questions you wrote: >On Fri, Oct 17, 2003 at 03:30:19PM +0100, Jim Hatfield wrote: >> Anyone know what causes this: >> >> >speyburn-isis(113) pkg_info -x php >> >pkg_info: read_plist: bad command '@conflicts php4-cli-4*' > >You're trying to use new packages with your old FreeBSD system, which >is not supported. No I'm not, these packages were installed ages ago. But that's beside the point. Is there anything I can do to fix it? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Using poptop....
On Fri, 17 Oct 2003 17:11:40 +0100, in local.freebsd.questions you wrote: >> >> I would use mpd but it has problems with XP clients. >> >> > >What type of problems are you seeing with MPD (Netgraph variety) and >WinXP? >That's my VPN terminator software and all of my remote XP systems seem >to >have no problems with it. Interesting. What I see is on the XP box, the connectoid which shows the packet counters shows lots of receive errors. If I (for example) make an ftp connection to a machine on the protected network, it is OK but if I do a dir then it hangs and takes maybe 30 seconds to list a dozen files. I'm not alone, other people on the mpd-users mailing list have the same problems. My gut-feel tells me it is something to do with fragmentation but I don't know how to make XP tell me what the errors are. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Using poptop....
If anyone can give me a recipe for setting up poptop I'd be very grateful. I want to run it on a machine which is a gateway/firewall. One NIC has a public address and the other is on a private network, ie 192.168.1.x. I want to allow XP clients to connect into the private network. I found the man pages a bit terse(!) and they seem to assume that kernel ppp will be used whereas AIUI the port is built to use userland ppp. I would use mpd but it has problems with XP clients. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Strange pkg_ behaviour
Anyone know what causes this: >speyburn-isis(113) pkg_info -x php >pkg_info: read_plist: bad command '@conflicts php4-cli-4*' Jim Hatfield ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Strange behaviour of pkg_ commands
While upgrading ports/packages on a 4.5 machine I noticed some strange behaviour of the pkg_ commands. For example pkg_info with no parameters produces output which ends: >bash-2.05b.007 The GNU Bourne Again Shell >pkg_info: read_plist: bad command '@conflicts bison-[0-9]+' And now I cannot pkg_delete any package which would appear after bison in the pkg_info list. Weird. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Can no longer install ports
When I do a make install in any ports dir, I now get this: >speyburn# cd /usr/ports/graphics/tiff >speyburn# make >speyburn# make install >===> Installing for tiff-3.5.7_1 >===> tiff-3.5.7_1 depends on shared library: jpeg.9 - found >===> Generating temporary packing list >===> Checking if graphics/tiff already installed >*** Error code 1 > >Stop in /usr/ports/graphics/tiff. make deinstall generates the same error. I tried make -d A but the output was voluminous, to say the least. This coincided with strange problems doing portupgrades, where I would end up with duplicate directories in /var/db/pkg, the duplicate having .tbz or .tgz attached to the directory name. A clue as to where to look would be appreciated. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
