As announced in the previous article, we now have a company blog. Best thing for you to do is
to monitor blog.bytemine.net as well as my private website. :)
I will keep this blog alive for a month, so that people have a chance of adjusting their feedreader.
Afterwards it will just redirect to my private site.
well, it was just a matter of time. Just like every-other cool-hyped .COM company we just had to launch our own blog. You should point your feedreader over to http://blog.bytemine.net/, unless you can live without the technical ramblings from us ;)
No, seriously. We will be publishing various technical things related to what we do over at bytemine as well as announcements and alike.
I am not sure, on how I will continue with this blog. For now, I suggest keeping it in your feedreader as well.
While updating machines at our colocation in Frankfurt last friday, Bernd and I hit a pretty harsh bug. Because of changes in the disklabel code, the disklabel of our main customer mailserver got updated by the OpenBSD kernel upon booting with the new OpenBSD kernel. After updating all packages and configurations on the machine we gave the server another reboot. Next the machine got stuck in the first-stage bootloader, complaining that there is no OS to boot.
Quite baffled, we booted bsd.rd kernel just to see that there is no disklabel present. Now, usually you have backups of /var/backups, where OpenBSD keeps informations like this around. Of course, these files were not in the backup (time to blush on my part). Thanks to a hint from Henning from OpenBSD, we tried out scan_ffs(8). Since scan_ffs is not present in the ramdisk kernels, we compiled it statically and placed in on a USB stick. scan_ffs(8) scans a disk and outputs the partitions it finds. With a bit of thinking, we could then create a disklabel with the correct data. Once we had /var mounted, we could use the saved disklabel there to make sure the label is correct.
The manpage for scan_ffs(8) contains the following advise:
The basic operation of this program is as follows:
1. Panic. You usually do so anyways, so you might as well get it over
with. Just don’t do anything stupid. Panic away from your machine.
Then relax, and see if the steps below won’t help you out.
Very true. While we did not panic, the best thing todo in such situations is to think and evaluate your options and possible recovery methods and don’t be quick on trying to fix it, likely you will make it worse ;)
Thanks scan_ffs!
The german Newspaper Sueddeutsche has an interesting story (it’s from last year, still most certainly of interest) about xerox machines with internal harddisks. Once you’ve read the article, it’s quite obvious that xerox machines that only scan a document once but copy it many times, must have some internal memory. Anyways, full story here: Kopiergeraete voller Geheimnisse.
I’ve had my ThinkPad x60s for a few months now, but since the built-in 3G broadband device from Sierra Wireless was acting up a bit, I just got around now to actually set it up properly with OpenBSD. The device is a PCI MiniCard (in Germany the card is sim-locked to Vodafone. However, you can call IBM/Lenovo support and request the unlocking code, only effect will be, that you will not receive further support from Lenovo regarding the broadband device and configuration.).
Very common these days is, that these devices hold a USB hub behind the actual broadband chip sits:
[0:24] fkr(ripley):~ %> usbdevs -v
….
addr 1: full speed, self powered, config 1, UHCI root hub(0×0000), Intel(0×8086), rev 1.00 port 1
addr 2: full speed, self powered, config 1, Mini Card(0×6804), Sierra Wireless, Incorporated(0×1199), rev 0.01
The PCI Mini Card is a Sierra Wireless MC8755, in OpenBSD this device is driven by the umsm(4) driver that was initially written by Jonathan Grey. Since then I’ve added quite a few device ids (and have support for the Huawei E220 almost ready, but that is a different story ;)
From what I gathered, it will only work if you use a baudrate of either 115200 or 460800 when talking to it. All others yield a proper connection to the device, but ppp connections will fail.
My options file for pppd is linked here.
More interesting is actually the chat script:
ABORT BUSY
ABORT 'NO CARRIER'
ABORT VOICE
ABORT "NO DIALTONE"
'' AT
OK AT+CGDCONT=1,"IP","web.vodafone.de"
OK AT+CGQREQ=1
OK ATD*99***1#
Until I added the AT+CGQREQ=1 the connection succeeded, but was rather slow. The Sierra Wireless MC8755 is capable of HSDPA (High Speed Downlink Packet Access) that allows speeds up to 1.8 Mbit/s. Vodafone offers HSDPA in various cities in Germany and it is quite fast. The AT+CGQREQ= command sets the requested service profile. Since I added that, the connections are way better. (still not as fast as possible, but I have to digg further for that).
so, the only thing that is missing from the new soekris stuff we expected to be available in April are the SATA kits. Since today,we can actually ship the net5501-60 as well as the kerberos.si 19″ cases. For the dual cases, you can actually load them up with two 5501 plus two lan1621. Makes seven ethernet ports each :) Quite nice.
I will post pictures tomorrow. happy ordering!
Following (only on german available) story from the german online magazine Telepolis was sent to a mailinglist I read.
Paranoia als erste Bürgerpflicht.
Reading highly recommended. I will post a summary here in a bit.
The openbsd atheros (ath(4)) driver made it into the linux kernel. This is good news as apparently the linux people finally see again, that its good to have free drivers around. However, as undeadly reports, they also “changed” the license to GPLv2, while they were at it. (commit message). Oh wait, right, they can’t. Its Sam Leffler’s and Reyk’s Floeter’s code. While Sam originally released the code dual-licensed (BSD, GPLv2) the parts from Reyk (free HAL among them) are ISC-licensed. Judging from the questions on undeadly, quite few people seem to think, that BSD license does allow re-branding of the license. na, only the author can do that ;)
Anyways, unlike our (OpenBSD) own little mishap (see Slashdot covering te bc43xx incident) I’m sure, this was no accident.
Update:
KernelTrap as some coverage as well.
Finally we started to ship the new Soekris modell, net5501-70, with the small configuration modell, net5501-60 being on the horizon (not yet available). The new modell is quite a change from the 4801 line, more ram, four ethernet ports (yet still missing gigabit ethernet), way bigger cpu and crypto hardware.
Soon we will have rackmount cases for the new modell as well, currently only the standard case is available for ordering. Since Soekris still has issues with the SATA port and the current version of the Bios, there is no SATA Kit available for ordering as well.
If you want to be notified, sign up with up with our newsletter: kunden-embedded.
Tyrel from SitesCollide did an Interview with me during Confidence 2007. Here is a direct link to the mp3.
