No, not the cowboy series, but the Fedora development repository. As part of my investigation into why large transfers are hard locking my PC, I was advised to install the kernel from Rawhide, which also meant enabling the RPMFusion Rawhide repository to pull in the Nvidia modules. So first we install the repo: yum install fedora-release-rawhide.noarch Then we edit the repo files to enable them but limit them to kernel/nvidia RPM’s (we don’t want to upgrade to Fedora 15 Alpha!

I’ve compiled some RPM’s of John The Ripper 1.7.6 with the jumbo-12 and early-release-md5-gen-v3 patches applied. I can’t get the hmailserver-02 or intrinsics-2 patches to merge nicely, or build reliably even with manual fudgery. Its got a bit crazy recently with five patches for 1.7.6 after jumbo-12 which are incompatible with each other and the omp-des-7 patch. They really need to be merged into jumbo-13 or even 1.7.7 or preferably ditch patches altogether and work from Git, but apparently this won’t happen.

I upgraded my main desktop machine to Fedora 14 the other day and noticed that it hard crashes when transferring over about 300Mb of data over NFS. rsync and ssh are fine. So I did a bit of searching and as usual there’s a shedload of Ubuntu users reporting the issue but none of them fixing it or raising it with the kernel devs instead of Canonical. Anyway I tinkered with my /etc/fstab entries and it seems to either be down to the buffer size or async setting as now it seems to be working fine, and faster than before – odd as async and a larger buffer should speed things up!

Today I’ve been playing with my Fon Simpl wireless access point. It has an “internet” ethernet port for WAN, a “computer” ethernet port for LAN and public and private wireless networks, so I wanted to minimise what goes through the internet port to just the heartbeat and disable routing between my wired/wireless LAN’s. Essentially I’ve got the internet port in a different VLAN and subnet to my LAN or the private wireless network and the computer port is currently disconnected – if I want to admin the router I can plug a laptop into the computer port.

I got fed up with my shitty web host and their downtime – apparently they fucked up the OpenVZ kernel when running “yum update” on their CentOS server, then had RAID controller issues, then their ISP null-routed them after they got blacklisted by Spamhaus, then they fucked up the config of their new host node, blah, blah, lots of excuses (including flooding in Australia depsite the servers being in Amsterdam/Munich/Kansas, and poor outsourcing).