[leaf-user] Odd DHCPC behaviour - Bigpond Australia
Hi. I have just decided to upgrade my lovely Bering firewall to take advantage of uCLib. I have put together my Bering 2 floppy and almost everything works fine! The fly in the ointment is the connection to my ISP. I am on BigPond Cable and I use udhcpc and BPALogin in my working Bering floppy. Udhcpc doesn't seem to have been recompiled for uCLib so I am using dhclient and BPALogin in my uCLib version. The oddness is that using udhcp I get a particular IP Address (call it IPudhcp) but with dhclient I get a totally different IP (call it Ipdhclient). But my hardware is the same so my MAC addresses don't change. This is repeatable ie if I boot with my udhcp disk, I get IPudhcp every time, then if I boot my dhclient floppy I get IPdhclient every time. This seems odd to me because all I have read indicates that IP addresses might be allocated on the basis of stored MAC addresses, not some characteristic of the dhcp client. The reason this is an issue is because BPAlogin for the dhclient floppy won't login! It says it can't find the authentication server. The settings are the same as for the udhcp floppy and I use the dotted quad IP address to point to the server. However, I find that if I turn my modem off for about 5 minutes with my router off, then turn the modem back on, and reboot the router, it will connect. This sort of behaviour is typically associated with clearing the MAC address stored by the modem, but I don't see why it makes any difference to me because my MAC Address doesn't change! Then after about 5 minutes my connection disappears! When I look at my Shorewall logs following the disconnection I find that there are entries showing rejection of connections from the Authentication server to port 5050 (which is the BigPond heartbeat port) but to Ipudhcp, not IPdhclient! So the rejection logs show that the authentication server has tried to contact my router on the IP address that would have been allocated if I had used udhcpc! And because the connection was rejected, I am logged out! This is all very weird. It appears that dhclient is connecting ok using the allocated IP address but 'something' in the Auth server is still associating my MAC address with the IP address I would have been given if I was using udhcp. I am after any advice anyone can offer on what might cause this or how to get around it. Is there any connection between MAC Address and the dhcp client? Does the dhcp client send a MAC address to the auth server? I would like to try another dhcp client and I see Pump has been recompiled so I will try that first. Lynn, are you planning to recompile udhcpc because I would like to try that as well. Thanks folks, as usual! David Pitts IT Services Manager Reid Library University of Western Australia Telephone: (08) 6488 3492 Fax: (08) 6488 1012 --- SF.Net is sponsored by: Speed Start Your Linux Apps Now. Build and deploy apps & Web services for Linux with a free DVD software kit from IBM. Click Now! http://ads.osdn.com/?ad_id56&alloc_id438&op=click leaf-user mailing list: [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/leaf-user SR FAQ: http://leaf-project.org/pub/doc/docmanager/docid_1891.html
[leaf-user] Odd DHCPC Behaviour
Hi. I have just decided to upgrade my lovely Bering firewall to take advantage of uCLib. I have put together my Bering 2 floppy and almost everything works fine! The fly in the ointment is the connection to my ISP. I am on BigPond Cable and I use udhcpc and BPALogin in my working Bering floppy. Udhcpc doesn't seem to have been recompiled for uCLib so I am using dhclient and BPALogin in my uCLib version. The oddness is that using udhcp I get a particular IP Address (call it IPudhcp) but with dhclient I get a totally different IP (call it Ipdhclient). But my hardware is the same so my MAC addresses don't change. This is repeatable ie if I boot with my udhcp disk, I get IPudhcp every time, then if I boot my dhclient floppy I get IPdhclient every time. This seems odd to me because all I have read indicates that IP addresses might be allocated on the basis of stored MAC addresses, not some characteristic of the dhcp client. The reason this is an issue is because BPAlogin for the dhclient floppy won't login! It says it can't find the authentication server. The settings are the same as for the udhcp floppy and I use the dotted quad IP address to point to the server. However, I find that if I turn my modem off for about 5 minutes with my router off, then turn the modem back on, and reboot the router, it will connect. This sort of behaviour is typically associated with clearing the MAC address stored by the modem, but I don't see why it makes any difference to me because my MAC Address doesn't change! Then after about 5 minutes my connection disappears! When I look at my Shorewall logs following the disconnection I find that there are entries showing rejection of connections from the Authentication server to port 5050 (which is the BigPond heartbeat port) but to Ipudhcp, not IPdhclient! So the rejection logs show that the authentication server has tried to contact my router on the IP address that would have been allocated if I had used udhcpc! And because the connection was rejected, I am logged out! This is all very weird. It appears that dhclient is connecting ok using the allocated IP address but 'something' in the Auth server is still associating my MAC address with the IP address I would have been given if I was using udhcp. I am after any advice anyone can offer on what might cause this or how to get around it. Is there any connection between MAC Address and the dhcp client? Does the dhcp client send a MAC address to the auth server? I would like to try another dhcp client and I see Pump has been recompiled so I will try that first. Lynn, are you planning to recompile udhcpc because I would like to try that as well. Thanks folks, as usual! David Pitts IT Services Manager Reid Library University of Western Australia Telephone: (08) 6488 3492 Fax: (08) 6488 1012 --- SF.Net is sponsored by: Speed Start Your Linux Apps Now. Build and deploy apps & Web services for Linux with a free DVD software kit from IBM. Click Now! http://ads.osdn.com/?ad_id56&alloc_id438&op=click leaf-user mailing list: [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/leaf-user SR FAQ: http://leaf-project.org/pub/doc/docmanager/docid_1891.html
[leaf-user] 3rd week of trying to LEAF
Yeah, a dreary three weeks of trudging through computer debris ... gigo: all good things in good time. I can ping out, (ping test to ping 64.29.201.21) this must mean that my eth0 is properly configured. My windows box cannot auto detect ip address from LEAF router, green light on my (PCMCIA) eth1 is off. When I type ifup eth1 I receive a message of RTNETLINK answers: FIle exists /etc/init.d/dhcpd: No such file or directory ip addr says 4: eth1 mtu 1500 qdisc pfifo_fast qlen 1000 link/ether 00:95:5d:37:7f:65 brd ff:ff:ff:ff:ff:ff inet 192.168.1.254/24 brd 192.168.1.255 scope global eth1 Also, syslog reports firewall dhcpcd[2649]: dhcpStart: ioctl SIOCGIFHWADDR: No such device I am the blind seeing with my fingers, xcept not really seeing, dho! Can any help me with what is happening with my current Bering-uClibc 2.1rc1 set-up? Are there other commands I can type to investigate what is happening? Any other logs I can set up? Thanks. _ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.msn.com%2f%3fpage%3dmisc%2fspecialoffers%26pgmarket%3den-ca --- SF.Net is sponsored by: Speed Start Your Linux Apps Now. Build and deploy apps & Web services for Linux with a free DVD software kit from IBM. Click Now! http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click leaf-user mailing list: [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/leaf-user SR FAQ: http://leaf-project.org/pub/doc/docmanager/docid_1891.html
Re: [leaf-user] 3rd week of trying to LEAF
This is pretty fragmentary reporting. I have some ideas about what might be going on, but only guesses. See below. At 04:36 AM 2/17/2004 +, joah moat wrote: Yeah, a dreary three weeks of trudging through computer debris ... gigo: all good things in good time. I can ping out, (ping test to ping 64.29.201.21) this must mean that my eth0 is properly configured. Depends. What is 64.29.201.21 ? If it is your own external (eth0) IP address, it only confirms that eth0 is properly configured in part. You still need to verify that you can reach actual Internet sites ... your external gateway address (whatever that might be) and some well;known site that answers pings. And you don't say where you are ping'ing from ... a final test woould ping from a LAN host. My windows box cannot auto detect ip address from LEAF router, green light on my (PCMCIA) eth1 is off. This sounds like it could be a hardware problem ... with the NIC, the cable to it, or the hub or switch it connects to. Is the appropriate LED on or off at the hub/switch end? But LEDs on NICs are anything but standard, so I don't really know what the "green light" on an unnamed NIC is or when it is supposed to be on. Consult the docs that came with the NIC. When I type ifup eth1 I receive a message of RTNETLINK answers: FIle exists /etc/init.d/dhcpd: No such file or directory So the script to start dhcpd cannot find something it needs. Is the dhcpd daemon itself present -- probably in /sbin or /usr/sbin ? Is the config file for dhcpd present (probably /etc/dhcpd.conf)? ip addr says 4: eth1 mtu 1500 qdisc pfifo_fast qlen 1000 link/ether 00:95:5d:37:7f:65 brd ff:ff:ff:ff:ff:ff inet 192.168.1.254/24 brd 192.168.1.255 scope global eth1 Well ... this says that the kernel thinks it had loaded whatever modules it needs to operate some NIC as eth1, since the interface is both created and configured. Also, syslog reports firewall dhcpcd[2649]: dhcpStart: ioctl SIOCGIFHWADDR: No such device dhcpcd is the DHCP client; Bering uses it (when appropriate) to get an IP address for the external interface. Is dhcpcd present on the system (again, check /sbin and /usr/sbin)? Are you supposed to be using DHCP to get your external address? If so, do you get an address assigned to eth0 (probably, if the eth1 entry really is the 4th interface "ip addr" reports, but the rest of that output will make it apparent either way). In any case, this entry by itself means nothing. I don't recall about Bering specifically, but sometimes it just means that dhcpcd is trying to get a lease too soon, before the actual interface is created. When I see this, I normally see a later lease attempt that succeeds. It's a bit sloppy, but no big deal, really ... unless there is an associated problem with the relevant interface actually failing to come up. I am the blind seeing with my fingers, xcept not really seeing, dho! I don't know what this means. Are you really using a Braille UI, or is this just a muddled metaphor for something? Can any help me with what is happening with my current Bering-uClibc 2.1rc1 set-up? Are there other commands I can type to investigate what is happening? Any other logs I can set up? Thanks. How familiar are you with Unix/Linux generally? If not very, there are a few basic commands you should know about. "ps ax" will give you a list of all processes that are running. You can use it to see if dhcpd and dhcpcd are actually running on the router. "netstat -l" will tell you what processes are listening on what ports. "dmesg" will display the current contents of the dmesg ring buffer, which can tell you more about what the kernel and init process were up to before syslogd was started. The file "dhcpd.leases" should contain information about what leases the router has issued. I'm not sure where Bering keeps that file; you might try /var/lib/dhcp/dhcpd.leases . I'm skipping commands you already have used, and ones that are described in the SR FAQ mentioned at the end of this message. --- SF.Net is sponsored by: Speed Start Your Linux Apps Now. Build and deploy apps & Web services for Linux with a free DVD software kit from IBM. Click Now! http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click leaf-user mailing list: [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/leaf-user SR FAQ: http://leaf-project.org/pub/doc/docmanager/docid_1891.html
[leaf-user] udp masq entry and dns abuse
Lately i'm having problem with udp masq entry in my internet leaf-router with a lot of messages like this: IP_MASQ:ip_masq_new(proto=UDP): could not get free masq entry (free=36864) here's my network looks like : - --- |leaf-router| |RH9 squid & dnscachex| to -|eth0 eth1|---|HUB|--|eth0 eth1| internet| | | | - --- | | |switch| | | | subnet A - | | | subnet B --- | | subnet C --| everytime i stop dnscachex, the messages also stop. am i having dns abuse from my internal network? or is it because there are too many clients in my internal network? how do i deal with it? any suggestion will be very appreciated. regards, gregor __ Do you Yahoo!? Yahoo! Finance: Get your refund fast by filing online. http://taxes.yahoo.com/filing.html --- SF.Net is sponsored by: Speed Start Your Linux Apps Now. Build and deploy apps & Web services for Linux with a free DVD software kit from IBM. Click Now! http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click leaf-user mailing list: [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/leaf-user SR FAQ: http://leaf-project.org/pub/doc/docmanager/docid_1891.html
