similar to: missing ipconfig.exe

From: Geert Stappers <stappers at debian.org> To make klibc-ipconfig transmit the DHCP vendor class "d-i" and get the DHCP preseeding filename in a text file, it needs the attached patch. belows patch allows ipconfig to be used in Debian installer once the hostname is sent too, see http://blog.andrew.net.au/2006/05/17#d-i_dhcp_hacking Signed-off-by: maximilian attems <maks at

this is just on the common path, need to checkout others too, but a start: commit 5501109f6597f2bbed17a264b247c36220362dfb Author: maximilian attems <max at stro.at> Date: Mon Apr 26 08:07:51 2010 +0200 [klibc] ipconfig: Fix valgrind errors valgrind was rightfully complaining on ipconfig eth0: ERROR SUMMARY: 5 errors from 4 contexts (suppressed: 0 from 0) all

klibcs ipconfig tries to guess the nameserver if no nameserver is provided. This may happen due to misconfigured dhcp or the use of the ip= parameter. This patch removes the guessing from ipconfig as it is not possible for any other script to determine if the provided nameserver is valid or a guessed one. If the old behavior is really needed this could be easily implemented in an external script.

Hi, Here are two fixes for ipconfig/dhcp, the second of them being required for my dhcp3 server to accept ipconfig's requests. Thanks, Louis Louis Rilling (2): ipconfig: Fix missing dhcp_end field ipconfig: Fix null ciaddr on DHCPREQUEST during SELECTING state usr/kinit/ipconfig/dhcp_proto.c | 4 ++-- usr/kinit/ipconfig/packet.c | 2 +- 2 files changed, 3

Hello, I found a bug in ipconfig and I'm sending a proposed patch for it. I've only seen it happen in 2 dhcp-server environments. Scenario: ipconfig sends a DHCP_DISCOVER, server A answers with a DHCP_OFFER, server B answers with a DHCP_OFFER, ipconfig sends a DHCP_REQUEST for server A, ipconfig accepts the DHCP_OFFER from server B instead of DHCP_ACK from server A. <== BUG The reason

Hi, I had the problem that 'ipconfig' as DHCP client in Debian-squeeze could not get DHCP Offer reply from the built-in DHCP server of the cheap route that I have. The name of the router is 'MegaBitGear TE4571E' which I got at contract of ADSL Internet access service in 2009. The router is not too old and can reply to other DHCP clients like ISC-dhclient, WindowsXP and Vista.

Hi!! I found your ipconfig stuff in the udev rpm in SuSE 9.1. I'm about to setup a initramfs (pretty cool - much better than initrd) to boot linux diskless clients over the net. All the tools I would need, seem to be there with ipconfig/nfsmount ... I like to use the information I get from PXE/syslinux or Etherboot writing something like:

V1->V2: add the documentation. V2->V3: change the split sign from ":" to "@" for IPv6. Add the the ipconfig for vir-prep. Usage like below: [root at Allen ~]# virt-sysprep --ipconfig="eth0:192.168.1.2,255.255.255.0,192.168.1.1" --enable=ipconfig -d clone-6u1 OR [root at Allen ~]# virt-sysprep -d clone-6u1 Signed-off-by: Wanlong Gao <gaowanlong at

I had an odd situation yesterday, klibc 1.5.17 ipconfig was trying DHCP on all 3 NICs in the server, but one has a bad switch that causes the link on eth2 to go up/down every 10 seconds. IPconfig would just time out. If I tell ipconfig to only try eth0, then DHCP address is configured, and the netboot proceeds 100%. I'll take a peek at the code but maybe someone here is familiar with the

Hi, it seems that ipconfig has a problem if multiple devices are up and connected to the same network. It seems that it uses the wrong socket/device index to compare it to incoming packet. To be more precise, the packet gets discarded in do_pkt_recv as the ifindex from state differs always from the incoming packet To reproduce create two tap devices: $ sudo tunctl -u uli -t tap0

Hi Rich: I just send a patch to request for your comments, then will do further work about this. things like this: [root at Allen ~]# virt-sysprep --ipconfig="eth0:192.168.1.2,255.255.255.0,192.168.1.1" --enable=ipconfig -d clone-6u1 OR [root at Allen ~]# virt-sysprep -d clone-6u1 Please comments. Thanks -Wanlong Gao Signed-off-by: Wanlong Gao <gaowanlong at cn.fujitsu.com>

Thank you Maximilian, this was much easier! I still have problems with my email setup, I haven't found the time to migrate my mailbox from Windows yet. And gmail (webmail) doesn't allow many customizations, so I'm also uploading the patch to http://users.sch.gr/alkisg/temp/0001-Signed-off-by.patch I'm sending a patch regarding the discard_packet() function, please consider it for

Hi, ipconfig is used in ltsp (thin clients) initramfs. Some of the users have to have 2 dhcp servers in ltsp environments, so they hit on this problem: The first dhcp request is from the pxe rom boot code. This code ignores any dhcp offers that don't have a filename field. So they always get the correct address, from the ltsp server. The second dhcp request (in initramfs) is done by

In general for what most people do ipconfig is equivalent to ifconfig (getting phased out) or ip in linux. What I am trying to do is create ipconfig script that acts identical to the windows behavior's output so I can create better scripts in powershell that work on both platforms. ipconfig has few options that are not listed with ifconfig or ip. The Windows IP Configuration section has two

When attempting to boot a system using Infiniband interfaces, ipconfig does not recognize Infiniband NICs as having valid MAC addresses, so they are silently ignored. As I would like to be able to netboot using Infiniband, I have patched ipconfig to support Infiniband interfaces. Fortunately, it's a pretty simple and straightforward patch. Below is the patch for your consideration. ---

Hello, if ipconfig is run as ipconfig :::::: (i.e. long spec with all fields empty) and the computer has multiple network interfaces and only one of them is connected to dhcp managed network, ipconfig ends up in an infinite loop (see the log bellow). The following patch solved this for me. According to README, if the <device> in the long spec is empty, the first interface should be

On Fri, 27 Aug 2010, Vagrant Cascadian wrote: > > ipconfig apparently assumes that the dhcp server's ip address is the > nameserver when dhcp gives it no nameservers, dhcp gives the nameserver > 0.0.0.0, or the nameservers are not ip addresses (i.e. ns1.example.org). > while these scenarios are probably not extremely common, i'm not sure > this is a good default with no

[ adding relevant cc's ] On Tue, 10 Aug 2010, Julien-externe BLACHE wrote: > Hi, > > Attached is a second patch for ipconfig, which further improves (actually, > restores) ipconfig's behaviour on busy LANs with UDP noise. > > Commit 4efbcf90f60e27febe883ef052454d8cfded5c15 really is the root of all > evil, it badly broke ipconfig by doing a lot of changes all

Hi all, I've recently been experimenting with Xen VMs booted using DHCP/NFS, and that's looking good. One nuisance is that I'm unable to get the DNS to update on the server since there's no hostname in the DHCP request packet. I'm aware that I can specify a hostname using the long-format device name, but this doesn't seem to play well with initramfs. So, I'd like to

I'm trying to get nfsroot via kinit to work, and I'm increasingly of the opinion that the ipconfig code unfortunately is in much worse shape than I thought. I've tried to clean it up quite a bit, but it's still rather messy. Anyway, one of the thing that I think is probably broken is the handling of "ip=dhcp" or anything else without a specified interface in the