Howto: Siemens Gigaset SE567 Bridge Mode and PPPoE

Well after about 6 years of rockin a Dlink DSL-504 I decided to upgrade. The previous modem has worked very well, sure it degraded and lost all but 1 of its Ethernet ports… but I ran a fantastic European firmware on it for years and had it perfectly tuned. Too bad I could never get the stupid thing to bridge… The strange thing about this device is it started to block certain websites. I was very confused on this one for months! First I thought that it was my server, but everything seemed to check out. Then I learned only a few particular computers would be blocked. This really messed things up! How does this happen?! Then I was 99.99% certain it was Frontier Internet. So I left it at that. But since this DSL modem is so old I decided to upgrade. I called up Frontier and asked specifically for a modem that was NOT wireless, but they sent me the Siemens Gigaset SE567 one anyways. Plus a really lame monthly modem rental charge. What happened to free modems?! These “Internet” companies are getting savage! Either way, on to what I found about bridging this modem, and oh yes, it CAN be done!

From testing this device, I found that 1. The wireless sucks, 2. This modem will not bridge and do PPPoE at the same time. So the first thing to do would be to get a client that can do PPPoE. I used my server, but any router, whether it is wireless or not, should do the job just fine.  I will explain how to setup Linux below as well.

This is with Frontier Communications personalized firmware:

1.  Attach your computer directly to a LAN port on the modem and log in as admin to the modem interface.
2. Click on ‘ISP Connection.’
3. Click ‘Advanced Settings’, then click ‘Configure the ATM Virtual Circuit.’
4. Click ‘Add a New VC.’
5. Set the VPI to 0 and the VCI to 35 (These could be different, check with the previously used VC to confirm). Click ‘Next.’
6. Select ‘RFC-2684 Bridged’ and Click Next.
7. Enter a name for the connection (or just leave the default) and click Next then click Finish.
8. Delete all of the VCs except the one you just created. (you can always restore defaults)
9. Make sure the new VC is Enabled and Reboot the modem.

You now have a basic modem with no more annoying second NAT. Another reason is if you have a server, it is good to get your outside IP address directly to your computer and have your beefy firewall take care of everything.   At this point, acquire a router and plug it into the modem. Set it up for PPPoE and enter your credentials. You should be connected to the Internet!

Now to make this work on Linux, you need a package called rp-pppoe. For Fedora 9 as root you can install it as follows:

yum install rp-pppoe

This will install the proper package. From then, there are 2 ways to setup the PPPoE connection. Graphically or through the command prompt. I did this graphically and that is what I will explain, but I know that there is a script at /sbin/pppoe-setup that is a good starting point.
To setup the PPPoE connection go to ‘System” -> ‘Administration’ -> ‘Network’. From there, edit the Ethernet card that the DSL modem is connected to (typically eth0). Set the card to ‘Staticly Set Ip Addresses’ and leave all of the fields blank. Click Ok, then ‘Deactivate’ and ‘Activate’ the device, saving all changes.
To insert the PPPoE connection:

1. Click ‘New’, Select xDSL connection, Forward.
2. Select the blanked out Ethernet device from earlier, Enter provider name (Frontier), Account type Normal, and your credentials, Forward
3. Then click Apply

After that, this is all you need for a simple PPPoE connection. To further configure it, open up the configuration on the xDSL connection and head over to the advanced tab. In this tab, make sure ‘Restart if connection dies’ and ‘Make this connection the default route’ are checked. You may also check ‘Use synchronous PPP’ but not all providers support this so I would test for this function later. After this, go back to the General tab and check ‘Activate device when computer starts’ only if you want to. And if you run your own DNS, uncheck ‘Automatically obtain DNS information from provider’.

There! Now watch your messages log when you first connect to make sure no errors pop up. You can do this by running this command in terminal:

tail -f /var/log/messages

This will give you live updates of your messages log. If you have problems, you may want to check to see if you bridged the modem correctly. To do this, issue this command in terminal:

pppoe -I eth0 -A

Of course, replace eth0 with whatever Ethernet port is connected to your Modem. What this command will do is query your service providers servers for a connection route. You should get an output as such:

Access-Concentrator: SERVER.SERVER.AC
Got a cookie: xx xx xx xx xx xx xx xx xx xx xx xx xx xx xx xx
AC-Ethernet-Address: 00:00:00:00:00:00
--------------------------------------------------

Of course, yours will be different. If you do not get anything like this, or you get a “pppoe: Timeout waiting for PADO packets” error, then it may be wise to go back and check the configuration of your modem.

Enjoy and happy PPPoE’ing! =)

Sources:
Siemens Gigaset SE567 Port Forward Problems
PPPOE setup on Fedora Linux

Fedora: Convert Your Partition From ext3 to ext4

Well today, I decided to upgrade my Fedora 9 server to Fedora 10. All went well there. Then I read about the new partition format called ext4. Now since this server is a live server, I wanted to convert the partition as quickly and easily as possible.

First, you want to make a backup of your data, THIS IS IMPORTANT! Any time you mess with a partition it can become a nightmare…. I am not responsible if you screw up your computer!

Important stuff aside, we can move on! This stuff needs to be done as root.

All of this needs to be done from the terminal, so open one up and open up /etc/fstab and have a look at your partitions:

nano /etc/fstab

Of course you can use any txt editor you would like, I just prefer nano.

fstab stores the information about your partitions and how to mount them. Be careful as you can mess things up! It should look similar to this:

#
# /etc/fstab
# Created by anaconda on Wed Apr  1 01:21:11 2009
#
# Accessible filesystems, by reference, are maintained under '/dev/disk'
# See man pages fstab(5), findfs(8), mount(8) and/or vol_id(8) for more info
#
/dev/VolGroup00/LogVol00  /                    ext3     defaults,usrquota,grpquota 1 1
/dev/md0                  /boot                ext3     defaults        1 2
tmpfs                     /dev/shm             tmpfs    defaults        0 0
devpts                    /dev/pts             devpts   gid=5,mode=620  0 0
sysfs                     /sys                 sysfs    defaults        0 0
proc                      /proc                proc     defaults        0 0
/dev/VolGroup00/LogVol01  swap                 swap     defaults        0 0

Looking above, each line has these columns: Drive, mount point, file system type, mount options, dump options, file system check options. We would like to change the file system type. So change the third column on the line that corrisponds to your root partiton (/) to ext4

/dev/VolGroup00/LogVol00 / ext4 defaults,usrquota,grpquota 1 1

Once this is done, exit and save the file. At this point, if the computer is rebooted it should still mount and boot. But a few commands will be confused. The mount command will report it as ext4 while the system-config-lvm command will correctly display the file system as ext3. Now we must remake initrd, or the /boot files, so that they include the ext4 driver. Do this step now as you may not be able to boot if you convert the drive before. First lets make sure the proper tools are updated:

yum upgrade mkinitrd e2fsprogs

Updating initrd works as follows:
mv /boot/initrd-`uname -r`.img /boot/initrd-`uname -r`.img.old
Then:
mkinitrd -v --with=ext4 /boot/initrd-`uname -r`.img `uname -r`
At this point, if you wish to maintain backwards compatibility with ext3, it is a good place to stop. Even though it is not fully converted, you still will have benefits such as delayed allocation which improves performance and reduces fragmentation and multiblock allocation which also improves performance, especially for older/slower hardware. If you wish to maintain backwards compatibility, all you need to do is restart the computer. If you wish to fully convert, continue below. If you fully convert, you will not have backwards compatibility!

Now, in order to continue, I would HIGHLY recommend that you boot up into the Fedora Live, or Fedora Install CD/DVD. I did not do this because It was a live system and I did not have a CD/DVD ROM drive at the time. So this does work without booting up into the Install/Live CD/DVDs but I do not recommend it. From the Install DVD select “Rescue Installed System” from the boot menu and select your language and keyboard layout. Also, select No to starting the network interfaces as we do not need them. Next, at the “Find Linux Installation” dialog select Read Only. Press Ok and continue.

Good practice is not to mount the partition, but in this case, we need to be able to see the partitions easily, hence the read only. Do not differ from these instructions once you start. Just let the system do its work. We don’t need any partition failures.

Since you are in the shell at this point, go ahead and run the mount command. You should get something like this at the top:

/dev/mapper/VolGroup00-LogVol00 on /mnt/sysimage type ext4 (ro,barrier=1,noextents,data=ordered

This line shows that the drive would like to use the ext4 file system driver [ext4], is read only [ro], has barriers enabled [barriers=1], doesn’t use extents [noextents] and that data is being stored using the ordered method [data=ordered]. Once converted, some of these properties will change. But at this point, the drive is ext3 in an ext4 disguise. Next we will initiate the command that will preform the conversion:

/mnt/sysimage/sbin/tune2fs -I 256 -O has_journal,extents,huge_file,flex_bg,uninit_bg,dir_nlink,extra_isize /dev/mapper/Volgroup00-LogVol00

If this command does not tell you to run fsck and to reboot, then it did not do its job. What will probably happen is it will say that inodes are already 256, If so, just remove the -I 256 section. Do this for other sections that it has a problem about as well. This is just to make sure that the default options are applied. Next we run e2fsck to update the system.

e2fsck -pf /dev/mapper/Volgroup00-LogVol00

Once this is done (and it may take a while depending on the size of your drive) go ahead a reboot! If it gives you crap about having to run the program manually, just remove the -p option and fix the problems manually. After the system has rebooted, issue this command below to check to see that it worked:

as root:
tune2fs -l /dev/VolGroup00/LogVol00 | grep extent

Example output:

Filesystem features:      has_journal ext_attr resize_inode dir_index filetype needs_recovery extent flex_bg sparse_super large_file huge_file uninit_bg dir_nlink extra_isize

This conversion does not convert files that were previously ext3, but newly created files become full ext4. This is not that important, but until a defragmentation program is finished, this will be the case. But there is one in the works, its called e4defrag and you may be able to download and compile it. I have not because I would rather wait till it is fully operational on the repositories. Once it does, I will be sure to inform of its awesomeness.

You can find more information on the e4defrag here:
Ext4 defragmentation with e4defrag

Sources:
Converting an ext3 root partition to ext4 in Fedora 10
How to edit and understand /etc/fstab