Re: [gentoo-user] VMware not running well
Kevin O'Gorman wrote: When I run as root, things seem better, but it complains that it was unable to extablish an IP number for the simulated ethernet card (networking is set up for NAT). I had lots of problems initially with running the perl script which configures the network, but eventually after trying about four times it just worked. I'm still having timer problems, vmware complains on startup about /dev/rtc not being available (though it is, and the module is compiled into the kernel). A FreeBSD guest keeps complaining that time is running backwards. Playing Gnometris on a Ubuntu guest is really really easy since the blocks take a minute to descend :-) -- gentoo-user@gentoo.org mailing list
[gentoo-user] Anyone run Gentoo on a Dell PowerEdge SC430?
They are so, so cheap right now. Dell UK has them at £199+VAT for the entry level configuration, with free delivery to the end of the month. I'm tempted. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Anyone run Gentoo on a Dell PowerEdge SC430?
kashani wrote: Jim Hatfield wrote: They are so, so cheap right now. Dell UK has them at £199+VAT for the entry level configuration, with free delivery to the end of the month. I'm tempted. I've got three of them running Gentoo in our dev environment. 2005.1 and 2006.0 disks worked just fine and had no kernel or driver issues running them as servers without X. Thanks. I can feel my credit card itching already... -- gentoo-user@gentoo.org mailing list
[gentoo-user] Does the automounter support direct maps?
The man page for auto.master(5) refers to direct maps, ie The mount-point for the direct map is always specified as /- in the auto.master map. My auto.master map looks like this: speyburn ~ # ypcat -k auto.master /home auto.home /mp auto.mp /- auto.direct and my auto.direct map looks like this: speyburn ~ # ypcat -k auto.direct /users -rw,soft banff:/export/home/banff So a reference to /users/username should result in banff:/export/home/banff being mounted on /users - but this does not happen, I get no such file or directory. Also I only have two automount daemons running, for the two indirect maps. Can the Gentoo automounter handle direct maps? BTW while experimenting with this I found a funny in pam. /etc/pam.d/xdm refers to pam_console.so, but there is no such module in /lib/security. Is this provided by a user package? I can't remember the incantation for searching the whole package set for a single file, even if that package isn't installed. Also BTW is it worth reporting on errors in man pages? The man page for autofs refers to /etc/autofs/init.d/autofs, where it should be /etc/init.d/autofs. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] PHP and files over 2 Gb
On Mon, 22 Aug 2005 15:45:23 +0100, in local.gentoo.users you wrote: It appears that PHP as built can't handle files over 2Gb - I get warnings whenever the application peruses a directory with such files in it. I had a look at the USE flags for PHP but didn't see anything obvious. Is there an easy way to build PHP to have large file support? http://bugs.gentoo.org/show_bug.cgi?id=24373 That was useful, thanks. It is indeed not enough for PHP to be able to handle large files, Apache has to as well (I think the video files are delivered directly by Apache and not via a script). Looks like I will have to wait for 2.2. -- gentoo-user@gentoo.org mailing list
[gentoo-user] PHP and files over 2 Gb
I just bought a Pinnacle ShowCenter and am using it with some Apache/PHP/MySQL code to play mpeg movie files to a TV. It appears that PHP as built can't handle files over 2Gb - I get warnings whenever the application peruses a directory with such files in it. I had a look at the USE flags for PHP but didn't see anything obvious. Is there an easy way to build PHP to have large file support? -- gentoo-user@gentoo.org mailing list
[gentoo-user] Where to find smbmount?
A system I have which is a year or so old has a /usr/sbin/smbmount, but if I use qpkg -f to find out which package owns it I get nothing. A recently-installed system doesn't have this file and I can't find any obvious package which would have it (except maybe Samba). Do I need to install Samba to get smbmount? I would not have thought so since it is referred to in the man page for mount. -- gentoo-user@gentoo.org mailing list
[gentoo-user] Ybind problems - doesn't like YP server
We have a Solaris based YP server. On a fresh install I set /etc/yp.conf to: domain insignia broadcast and set NISDOMAIN to insignia in /etc/conf.d/domainname. If I run ypbind -debug I get: speyburn ~ # ypbind -debug parsing config file Trying entry: domain insignia broadcast parsed domain 'insignia' broadcast add_server() domain: insignia, broadcast [Welcome to ypbind-mt, version 1.17.2] do_broadcast() for domain 'insignia' is called Answer for domain 'insignia' from server 'panther.internal.local' leave do_broadcast() for domain 'insignia' Pinging all active server. Pinging all active server. The YP server is correctly identified. I can ping it OK in another window. However if I do a ypwhich in another window I get: speyburn ~ # ypwhich can't yp_bind: Reason: Domain not bound It doesn't make any difference if I use the -broken-server flag. This was working fine on the machine which failed and I'm in the process of replacing, and I'm pretty sure I didn't do any more than the above. -- gentoo-user@gentoo.org mailing list
[gentoo-user] Interesting install experience
The machine I posted about earlier (GRUB GRUB GRUB...) is dead. It hung booting the 2005.0 CD, and if I booted a DriveImage CD with a DOS partition, every key on the keyboard was echoed ^A. Ah well. So I just installed another machine, using the 2005.0 CD and using the new instructions. It has a Matrox G400 so I added support for that in the kernel. This may have been a mistake. Everything is fine until I reboot, when after the GRUB screen and kernel selection, the screen goes black with lots of pretty blue squares all over it. Nice, but not helpful in logging in. And of course sshd isn't enabled by default. No matter, I reboot off the CD, mount everything, do the chroot thing and add sshd to the default runlevel, reboot and I can get in remotely. I guess I will rebuild the kernel with Matrox support removed and see if that fixes. BTW, what is the received wistom wrt building things into the kernel or building them as modules? As well as the G400 I have an Intel NIC and a VIA sound card, and this time round chose to build them in, though before I built them as modules. I'm not clear as to the pros and cons. -- gentoo-user@gentoo.org mailing list
[gentoo-user] Finally got mplayer updated
I posted ages ago about failing to update mplayer due to compile errors which seemed to be caused by it using header files under /usr/src/linux. So after a while of headscratching I did the obvious: renamed /usr/src/linux to /usr/src/linux.NIU and re-emerged. No problem. Still don't know why mplayer wants to use the header files under /usr/src/linux rather than /usr/include, but at least it's fixed. -- gentoo-user@gentoo.org mailing list
[gentoo-user] Why the changes to /etc/nisdomainname etc
I'm trying to move from a Unix-based DNS to an AD setup, changing DNS domain name (but not NIS domain name) in the process. I'm testing by switching DHCP servers on and off. I had a problem with NIS and while investigating noticed that some key files seem to have changed since the last time I rebooted the machine some months ago: /etc/hostname - /etc/conf.d/hostname /etc/dnsdomainname - /etc/conf.d/domainname /etc/nisdomainname - /etc/conf.d/domainname Is there any reason for this? Was something broken with the old way of doing it? Given that the DHCP server gives out the hostname and DNS and NIS domain names, should these files be left empty? /etc/conf.d/domainname is a bit confusing: #DNSDOMAIN= # This only set what /bin/hostname returns. If you need to setup NIS, meaning # what /bin/domainname returns, please see: # # http://www.linux-nis.org/nis-howto/HOWTO/ # NISDOMAIN=insignia Which line does that comment refer to, the one above or the one below? Jim -- gentoo-user@gentoo.org mailing list
[gentoo-user] ypbind man page incomplete?
The man page for ypbind does not specify if domainname needs to be set (it is only mentioned in the SEE ONLY section). The man page implies that all settings are in /etc/yp.conf. However if I have /etc/yp.conf set to: terminator ~ # cat /etc/yp.conf domain insignia broadcast but domainname not set: terminator ~ # domainname (none) then ypbind will not start: terminator ~ # ypbind -d domainname not set - aborting. jim -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Why the changes to /etc/nisdomainname etc
On Thu, 23 Jun 2005 12:35:46 +0100, Jim Hatfield
[EMAIL PROTECTED] wrote:
Given that the DHCP server gives out the hostname and
DNS and NIS domain names, should these files be left
empty?
Answering my own mail.
If /etc/conf.d/hostname and /etc/conf.d/domainname are
both blank, we get:
(none) ~ # hostname
(none)
(none) ~ # nisdomainname
(none)
(none) ~ # dnsdomainname
dnsdomainname: Unknown host
(none) ~ # cat /etc/resolv.conf
nameserver 172.16.64.1
nameserver 172.18.1.1
search isltd.insignia.com
(none) ~ #
So even though the DNS domain name was delivered by the DHCP server
it was only used for /etc/resolv.conf. And the hostname seems to be
set from /etc/conf.d/hostname well before DHCP is activated.
So I set /etc/conf.d/hostname and try again:
terminator ~ # hostname
terminator
terminator ~ # dnsdomainname
isltd.insignia.com
terminator ~ # nisdomainname
(none)
terminator ~ # cat /etc/resolv.conf
nameserver 172.16.64.1
nameserver 172.18.1.1
search isltd.insignia.com
Hmm. So if I set the hostname, the DNS domain name is set from DHCP,
but not if I don't. Weird. The same happens if instead of setting
the hostname in /etc/conf.d/hostnanme, I set dhcpcd_eth0=-H in
/etc/conf.d/net.
It would be nice if the NIS domain name could be picked up from DHCP,
is there a way to do that? The info returned is picked up because
/etc/yp.conf is set to:
domain insignia broadcast
but that doesn't seem to be enough for ypbind.
Also /etc/init.d/domainname looks like it is supposed to put a domain
line in /etc/resolv.conf:
[[ ${OVERRIDE} == 1 ]] \
resolv=${resolv}$'\n'domain ${DNSDOMAIN} \
|| resolv=domain ${DNSDOMAIN}$'\n'${resolv}
echo ${resolv} /etc/resolv.conf
but as you see above this does not happen, or it is overwritten later.
Once I can sort this out I can move on to my real problem, which is
that changing the DNS domain name in the DHCP server stops NIS
working. But that's for another mail.
Jim
--
gentoo-user@gentoo.org mailing list
[gentoo-user] Build failure when updating mplayer - header version problem?
I did a 'emerge --update world' and the mplayer compile ended like
this:
cc -c -I../libvo -I../../libvo -I/usr/X11R6/include -fno-PIC -O2
-mcpu=pentium4 -pipe -frename-registers -fno-pie -fno-pie -D_REENTRANT
-D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -I/usr/include/gtk-1.2
-I/usr/include/glib-1.2 -I/usr/lib/glib/include -I/usr/X11R6/include -I..
-I../loader -I/usr/include/gtk-1.2 -I/usr/include/glib-1.2
-I/usr/lib/glib/include -I/usr/X11R6/include -I../libavcodec -I../libavformat
-I/usr/src/linux/include -o stream_vcd.o stream_vcd.c
In file included from /usr/src/linux/include/asm/byteorder.h:57,
from /usr/src/linux/include/linux/cdrom.h:14,
from vcd_read.h:7,
from stream_vcd.c:25:
/usr/src/linux/include/linux/byteorder/little_endian.h:43: error: syntax error
before __cpu_to_le64p
/usr/src/linux/include/linux/byteorder/little_endian.h: In function
`__cpu_to_le64p':
/usr/src/linux/include/linux/byteorder/little_endian.h:45: error: `__le64'
undeclared (first use in this function)
The relevant bits of /etc/make.conf are:
CFLAGS=-O2 -mcpu=pentium4 -fomit-frame-pointer -pipe
CHOST=i686-pc-linux-gnu
CXXFLAGS=${CFLAGS}
MAKEOPTS=-j3
USE=-kde -gnome alsa oss avi cdr dvb dvd dvdr encode java mmx quicktime samba
sse sse2 truetype usb xine xv xvid
I don't recall seeing anything like this on the list over the last few
weeks.
I saw a mail this morning about not using headers from /usr/src/linux,
but this seems to be doing just that. Could that be the cause of the
problem? I see that there are differences between
/usr/include/linux/byteorder/little_endian.h and
/usr/src/linux/include/linux/byteorder/little_endian.h, and the diff
listing shows these very macros which are failing in the compile.
If this is the problem, how do I make mplayer build against
/usr/include/linux rather than /usr/src/linux/include/linux?
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] XML parsing error when downloading files with F irefox
On Wed, 11 May 2005 14:28:24 +0100, in local.gentoo.users you wrote: 1) if you're specifically talking about vtun, it's in Portage, so you don't even necessarily have to download it separately. I'm building a statically linked version for a remote Redhat box to which I only have ssh access and which has no compilation tools and doesn't have the LZO library installed. Fortunately it seems to work fine - phew! So this is probably one of the ever-popular make sure to create a new profile when upgrading issues. That would be the first thing I would try. Done, and it worked. Many thanks. -- gentoo-user@gentoo.org mailing list
[gentoo-user] Setting up a route through a point-to-point interface
Scenario: using vtun to build a Lan-to-Lan VPN. One end FreeBSD, one end Linux. On the BSD box I can set up a route directly through the tun0 device without having to assign it an IP address, like this: ifconfig tun0 up mtu 1450 -arp route add 192.168.10.0/24 -interface tun0 and it works just fine. On Linux (I've tried Gentoo and Redhat) it would seem that the following should work: ifconfig tun0 up mtu 1450 pointopoint route add -net 192.168.10.0/24 dev tun0 However I get a: SIOCADDRT: No such device which suggests that the tun device doesn't support the right ioctl variants. Has anyone done this? It's easy enough to do it using throw-away IP addresses for the two ends of the tunnel, but it's neater if they can be avoided. -- gentoo-user@gentoo.org mailing list
[gentoo-user] XML parsing error when downloading files with Firefox
I think this may have started happening since a recent Firefox upgrade. If I try to download (say) vtun from vtun.sourceforge.net, it points me at the mirror list and I pick one, say Heanet. I get the what should firefox do with this file dialog and select Save To Disk and when I hit OK I then get: XML Parsing Error: not well formed Location: chrome://global/content/filepicker.xul Line Number 1, Column 22: all, dialog=no, url, null, null, line); ^ ie the uparrow is pointing at the second comma. Is there an easy way of fixing this bar removing and reinstalling Firefox? -- gentoo-user@gentoo.org mailing list
