25.02.2012
Upgrade and revamp of site
13.09.2010
OpenWRT backfire with Optware
Reinstalling OpenWRT backfire requires one to have complete /etc backup as LuCi does backup only important files and thus misses custom scripts. Reinstall procedure for Optware can be scripted as:
# sed -i -e "s|src/gz*|src/gz packages http://downloads.openwrt.org/backfire/10.03.1-rc3/brcm47xx/packages" /etc/opkg.conf grep -vq "dest opt" /etc/opkg.conf && echo "dest opt /opt" >> /etc/opkg.conf opkg update opkg install kmod-usb2 kmod-usb-storage kmod-fs-ext3 opkg install e2fsprogs test -d /opt/var/lock || mkdir -p /opt/var/lock opkg install libpthread opkg install -d opt libstdcpp mkdir /opt /home swapon /dev/sda2 mount /dev/sda1 /opt mount /dev/sda3 /home cp /opt/etc/usbmount /opt/etc/init.d/ /etc/init.d/usbmount enable /etc/init.d/usbmount restart cp /opt/etc/profile /etc tar x -vz -C /home/backup/oleo/ -f /home/backup/oleo/etc-20100913.tar.gz cp /home/backup/oleo/root/.* /home/backup/oleo/root/* /root
08.05.2010
BEGUNJE WLAN
In our village I've been active mostly on wireless forum that was inspiring us for some time. We've taught that we can create community of wireless access points to share bandwidth. Besides Begunje we went also to nearby village Selšček and Topol. In the era of greatest enthusiasm we've had up to 30 computers interconnected. We faced technical problems, that compromised reliability of such network. Although this problems were solved by better antennas after a two years people just did not want to cooperate any more and went standalone. As I did not see technical problems my conclusion to this is: You can fix and problem except people mind. Now that even a forum died and it does not seem to be brought back I am putting backup of it on this website to remember all that fun time we've had with wireless. At that time I've started as Optware developer at NSLU2 website. I've learned a lot. And I am happy for that. But also got a lesson from a local people. I hope that they got it too.
Please follow the following link to enter backup of the Begunje WLAN forum. It's purpose is archival only and remembrance of great times now that official forum administred by Matej is down for at least half a year..
19.02.2010
Slice not recognized after upgrade from FreeBSD 7.2 to 8.0
Our server that lives from version 4.11 get hit with the following error when upgrading from 7.2 to 8.0
If you have invalid mount options, reboot, and first try the following from
the loader prompt:
set vfs.root.mountfrom.options=rw
and then remove invalid mount options from /etc/fstab.
Loader variables:
vfs.root.mountfrom=ufs:/dev/da0s1a
vfs.root.mountfrom.options=rw,userquota,groupquota,acls
Manual root filesystem specification:
: Mount using filesystem
eg. ufs:/dev/da0s1a
eg. cd9660:/dev/acd0
This is equivalent to: mount -t cd9660 /dev/acd0 /
? List valid disk boot devices
Abort manual input
mountroot>
After many hours of figuring the problem the following line solved the problem:
dd if=/dev/zero of=/dev/da0 oseek=1 bs=512 count=1
But before getting there I needed to boot with the memstick FreeBSD as booting the old kernel did not allow writing to the partition table. Backing up with
dd if=/dev/da0 iseek=1 bs=512 count=1 of=/tmp/sector1.ddand then transfering to some other host solved the things.
10.02.2010
Hylafax and FreeBSD 8
It seems that hfaxd binds to default listening socket on FreeBSD that defaults to ipv6 only. This prevents ipv4 clients to bind to it and sending faxses is not possible. Setting sysctl net.inet6.ip6.v6only=0 does not help. The only way to allow ipv4 clients is to patch with multiple -l -i pairs of socket binding.
--- /usr/local/sbin/hylafax.orig 2010-02-10 14:41:37.000000000 +0100
+++ /usr/local/sbin/hylafax 2010-02-10 15:02:21.000000000 +0100
@@ -154,7 +154,7 @@
$FAXQ; printf " faxq"
fi
if test $HFAXD_SERVER = yes ; then
- HFAXD="$HFAXD -i $FAXPORT"
+ HFAXD="$HFAXD -l 192.168.22.7 -i $FAXPORT \
+ -l 127.0.0.1 -i $FAXPORT"
HFAXMSG=" hfaxd"
if [ $HFAXD_SNPP_SERVER = yes ]; then
HFAXD="$HFAXD -s $SNPPPORT"
