I spent a morning trying to debug my LTSP server, which wasn’t booting after it’s feisty upgrade.

I assume it was something to do with RAID & LVM on the root initially, but after I got through that, it still wouldn’t boot.

Aah, but the recovery option worked

Ok, so maybe it’s something with usplash, restart and boot without quiet but with splash. Boots.

Eventually I found that it just needed a few extra seconds of spin-time in the grub screen (increase timeout to 10), and now it works.

Bloody hardware!

I had the pleasure of trying to get an AVM Fritz ISA card working under linux 2.6 this week. While drivers for AVM’s premium equipment is in Linux mainline, these drivers are not. The used to be obtained from AVM’s ftp site, but these days, if you look there, all that you’ll find is old 2.4. drivers, and a notice saying future drivers are available in SuSE.

Before I continue, I wish that all hardware vendors that write linux drivers would get them mainlined. I’ve begged AVM to do this before.

I tried porting the old driver to a modern 2.6 kernel, but while I could get it to compile, I never actually got it to work :-)

For the record, and for googlers, here’s how you build these buggers:

• You read gentoo’s ebuild
• This points you to somewhere on ftp.suse.com where you can download a 2.6 compatible version of the driver. The URL given didn’t work for me, but I poked around with similar URLs, and came up with the goods
• You’ll see a sed command that adds some includes to everything. Run this.
• You’ll also find a whole pile of patches, needed for recent kernels.
• Really old cards might not be actively patched, but many of the changes in other drivers are applicable to all of them, i.e. in the patch for fritz.pcmcia, the entire atomic_xchg function is removed; this must happen for fritz.classic, too.

I got the fritz.classic driver working with Linux 2.6.21.1, thanks to Gentooers :-)

My old (hand-me-down) Sony Ericsson P900 has been dieing for a while. The screen was scratched to death, the stylus missing, and the keypad cracked. (The Top row of buttons didn’t work, and the rest occasionally returned the wrong number). Oh, and the phone was crashing a lot.

I don’t do spending money on cellphones (because the ones I’d like cost a fortune, and I would never buy something that I don’t want), I’ve upgraded to a “new” hand-me-down SE P910i from my brother :-). It came Orange-branded, and after I firmware-updated it, it Orange-locked, grr.

I wish there was a free way to sort this out, but I couldn’t find one, so I paid my £16, and used Total Multi Server. It doesn’t require any cables beyond the docking station, and most importantly it worked.

Overall, the P910i is much like the P900, but better. The keypad is sturdier, the phone has more RAM, more applications, and it seems to get better reception.

I “bluetoothed” all my contacts across (because if you do it the syncing route, you loose information when a contact has more than one cell phone number, etc.).

I’ve installed a few apps, that I like:

• Opera Mini - the best web browser for a phone
• GMail - this may come in handy (all my incoming mail gets CCed to GMail)
• Handy Day 2004 - It makes the phone much more usable as a PIM, and I had a copy lying around.
• MGMaps - I’ve tried using maps in opera on my P900, and it was a pain.
• MidpSSH - Never needed it, but just in case, and anyway, it’s cool :-)
• Mxit - I rarely use it, but it’s jabber-peering makes it occasionally nice to have.
• quirc - Who ever wants to be stuck without IRC?

Finally I got it to synchronise with Google Calendar. The phone only does proprietary (windows) syncs or SyncML over GPRS. But Google Calendar doesn’t do SyncML. The OpenSource SyncML server is a Java monster which I wasn’t going to install on any of my servers, So I opted for ScheduleWorld as an intermediary. (don’t you love their shameless google-style website? :-) ) It works really nicely.

There is also a midp Java program for syncing to Google calendar, gcalsync, but it requires “JSR 75: PDA Optional Packages for the J2ME Platform”, which the P910i doesn’t have :-(

I spent a lovely weekend morning setting up monitoring on servers - yes, what fun :-)

I like all my servers to run logcheck, smartmontools, sysstat, and lm-sensors.

logcheck means watching your email every hour, and adding in yet more ignore rules for things your server thinks it’s perfectly OK to spit out.

smartmontools means waiting to see which attributes it’s going to complain about, making sure it’s set up to mail you about bad sectors, and getting this all in inside the 128-character line-length limit.

And lm-sensors, well that takes a lot of tweaking, to get all the alarms to stop ringing, labelling the right temperatures, and ignoring the disconnected pins.

Ugh, it’s painful work, but it helps in the long-run…

I’ve been enjoying our server at UK2.net. It’s a pretty speedy machine (although a little light on RAM - I suspect that they don’t want people running Xen), and it’s connected to a fat pipe. But I’ve been experiencing a lot of bad lockups.

I traced the problem to postmaping the uceprotect.net RBL file. They recommend that you rsync this file from them, and then postmap it into a fast lookup database for postfix, rather than using their DNSRBL service. But running the postmap was taking my box 40 mins. The same operation, on a loaded, lower-spec, 2 year old server took 2 mins (yes this server also has RAID1 on the volume concerned). On my UK2 box, while the postmap was running, the machine became totally unresponsive, and it could take a minute or two to log in, serve a web page, or even execute a basic command like ps.

Clearly something wasn’t right. And it was something in the IO system. The only answer is the 3ware RAID controller. (It’s a 8006-2, doing RAID-1) I know these controllers have a big buffer, so I looked up the 3ware website, for tuning guidance. I followed it to the letter, and things didn’t really improve. I tried the deadline scheduler, and tweaking the buffers, but it only got marginally better.

Personally, I’ve always used software RAID, even for RAID-5, and I’ve never had bad performance like that. And having the RAID in a portable format has really helped with recovery in the past. I understand that Windows monkeys have to use hardware RAID (because their software RAID sucks so much), but is this kind of performance normal?

I’ve asked UK2 to chuck my controller and give me software RAID :-)

Update

I’ve now got software RAID 1, and postmap runs in 25 seconds. That’s what I call a 60x speed improvement :-)

Oh, and the system is totally responsive while the postmap runs.