Seekbrain.com

LOL, Domain Renewal

Perlboy — Mon, 31 Jan 2011 09:46:50 +0000

No, I will not pay $45 to renew a domain. I found it interesting that this came with an Australian postal address which is (for the Google bait: 189 Queen Street, Suite 209, Melbourne) a bit interesting given that this registrar is trying to take my legitimately held (and nowhere near due) domain from my chosen provider and then charge me 4x what I can get from a reputable provider. No thanks Domain Renewal Group.

Awesome, $45/year

Brisbane CBD @ January 12 23:50

Perlboy — Wed, 12 Jan 2011 13:36:34 +0000

Brisbane CBD is all but abandoned. Previously manned police outposts have packed up and left. Albert Street is currently under a foot of water.

The ferry stop ramp is completely submerged:

And the Jade Buddha carpark ain’t going to be taking traffic anytime soon:

Meanwhile water is starting to flow out of the drains at Albert St, if that goes under completely we’ll be surrounded by water:

Flood Levels unchanged @ 6pm Jan 12 2011

Perlboy — Wed, 12 Jan 2011 08:03:15 +0000

Latest photo, levels don’t appear to have increased again but that has to be put in the context that the tides actually going out and the water isn’t going down. Tonight will be very interesting indeed.

Brisbane Floods 2011

Perlboy — Wed, 12 Jan 2011 05:36:40 +0000

So, I’ll start this off, I’m currently in the middle of the Brisbane CBD monitoring one of the many datacentres located within the city. We’re on the 21st floor of a non-inundated (yet anyway) building but water is starting to lap all around us. The Brisbane river is starting to break into Eagle St while the storm water drains at Albert Street are starting to push water out. We’re lucky, we’ve got 24hrs of fuel loaded into the generator and at least another 48hrs of fuel stock piled here so the facility shouldn’t be affected. We’ll be isolated though so I’ll try to continue to provide updates.

For now, here’s a photo of the water as it stands as of an hour ago:

Cloud Central Performance Testing

Perlboy — Thu, 07 Jan 2010 11:29:49 +0000

So, Cloud Central is running a public beta at the moment so I grabbed an account and decided to do some tests.

Cross copying from the wiki page I created on their site:

Initial impressions
- Nice GUI
- Xen based virtual machines
- Asked for a Nano which has 16GB of storage but it was only provisioned with 1GB. Suspect the idea is that additional allocations are made on demand.

Bugs/Showstoppers:
- Resizing the console on Firefox 3.5 on Ubuntu causes a complete X11 crash. That’s a serious worry.
- Was never told my default root password had to trigger a reset. Maybe that’s intentional.
- Password reset seems a bit slow/buggy, it works, eventually but there’s no ’spinner’ saying it’s doing much other than ‘Loading’.

Storage

Doing stock performance benchmarking using iozone on a Nano product.

iozone -azcR -i 0 -i 1 -i 2 -i 8 -n 256 -g 524288 -f /512MB-1Thread -b /512MB-1Thread.xls > /512MB-1Thread.log

- IO looks slow, during the test load went to 4 (~100% iirc) on a Quad Core AMD 64bit CPU.
- I couldn’t do a 512MB test since it apparently ran out of space. My space still hasn’t been freed.

Working just with the 256MB outputs.

Log output here: 512mb-1thread

Network

Network looks quite nice from initial impressions. PipeNetworks peering in Sydney and managed to pull 8-10Mbit from an uncontended source at work. Consistent throughout.

Conclusions

- Xen isn’t really ‘elastic’ cloud it’s more ‘resource on demand’.
- Network seems to be nice
- Storage performance ain’t that great

Will modify this post as more points come up.

Wow..

Perlboy — Sun, 04 Oct 2009 03:30:27 +0000

Trevor Rose has written a nice piece on VC’s, Inventors & the business issues faced when taking an idea or invention into a product.

The timing on this article is surreal since I myself am facing what I’d call ‘inventive hurdles’ whereby I have to try and figure out the business/financial distribution of something I just want to get to a releasable product. In the most part I agree with Rose perhaps a little bit too boisterous for my liking but hey, that’s what makes it such a snappy article so maybe that was the point.

I think we’re switching to MobileMe

Perlboy — Wed, 24 Jun 2009 13:20:47 +0000

I’m not a Apple fan boi by any stretch of the imagination and have day-to-day experience with running in multiple email environments (including Exchange, Courier-IMAP and Zarafa) but for personal mail at least MobileMe looks like a serious contender.

For ~$200/yr both Sal and I can get integration with our iphones, synchronised appointments & contacts as well as iphoto integration for our files. For a long time now I’ve been running a dedicated mailserver installation for her and I beginning with an old dovecot->courier-imap setup then a community Zarafa install and most recently a licensed one (on a trial anyways). With all these setups I’ve had issues with integration, courier-imap lacked central calendar/contacts, community Zarafa severely limited my ability to do regular backups while licensed Zarafa provided this functionality it remained clunky and still didn’t give Sal access to full calendaring support on her Mac. Sure, Exchange integration IS good on the iphone but having completed a test Exchange deployment it was costing me more time then a solution for personal reasons made sense for (mainly because I’m a Linux/Unix guy! :)).

Sal had been bugging me about MobileMe for a while and finally I thought ‘umm, I do this every day at work, why am I producing undue time usage [when I could spend it with her :P]?’ so I signed her up. Sure, I could have gone to gmail or yahoo but as far as I’m aware that didn’t provide appointment booking support (and sync back) from the iphones themselves.

Nasty bits had to deal with:
- Pulling contacts & calendars out of zarafa’s cold hands was damn near impossible
For Sal at least I could forward all her contacts as VCF’s (MANUALLY i might add from the iphone) and back fill her calendar appointments. For anyone with an active calendar that’s ‘working’ (which I guess was the main reason I made such a drastic decision) this could be particularly painful. I couldn’t find an appstore app that allowed mass attach and send (only commercial looking ones for everything BUT MobileMe).

- Apple doesn’t appear to have too many migration tools
Our previous data on Zarafa (via IMAP) is still there. I don’t think there’s any sanctioned Apple mail sucking tools available there so I’m going to have to look at imapsync. That could be interesting since with all my mail in the Zarafa install it flogs the hell out of the mysqld zarafa stores everything in.

- MobileMe doesn’t seem to allow hard setting of From address

Dear Apple Inc,

Yes, we like me.com & mac.com, we might even use them but we also have personal domains too which we prefer as primary contact points. Please, I beg you, allow us to at least set Reply To on our mail account.

- MobileMe timezone seems a bit warped

Booking appointment on the iphone is the only way to get it to look like the right time. Yes, we set the Timezone setting but it’s still broken. Think that’s probably worthy of a support ticket.

Fixed: Every calendar has it’s own Preferences for applying timezone support otherwise it uses Cupertino time. Not sure why this is separated from the main Account Preferences.

That’s it so far with more to come if I do take the plunge of migrating my rather huge (nearly 10 years) mail archive to MobileMe. :-\

Done for now.

Stuart

WHT-US Exploited

Perlboy — Mon, 23 Mar 2009 14:33:21 +0000

Heya,

They’ve taken it down again now but WebHostingTalk.com has been hacked.

I have obtained a publically posted copy of the WHT-US user database and verified the content in it is authentic.

ANY Password used by WHT-US members is now ‘in the open’.

Change everything quickly, good luck to WHT.com.

Stu

Storage Musings

Perlboy — Sat, 21 Mar 2009 06:29:03 +0000

Someone asked on WHTAU about SAN stuff and I figured, well, I’ll post it here too.

Ok, talking about SANs this is an area I don’t know much about it… If you grab yourself a fiber channel SAN hock it up to a few of your servers which have the required cards, than the servers are running completely off the SAN and not the internal drives? If that’s the case what sort of options are available to backup the entire first SAN to a second SAN for fail over if the primary SAN dies?

Hmm, SAN discussion.. Yum.

Fibre channel SAN’s typically ship with 2 fibre channel uplink ports (you can get some which have more). There’s a few ways about integrating with your new found space & speed freedom.

In the situation where you don’t buy a set of switches but you do get a set of redundant SANs you generally look at getting 2 servers in failover and direct attaching things in there then sharing the space as required from those host nodes. There’s a number of distributions specifically for this but for performance I’ve always leaned towards Solaris. Nexenta do a Solaris derivation specifically targeted at this market and they allow you to do nfs/smb/iscsi from there. When it comes to CPanel I guess you’d go the iscsi option here and either:

a) Get a iscsi initiator card in there and initiate to your storage pair.
b) Put a bootstrap startup on the hardware itself which then does an iscsi initiation and starts your real OS.

Guess I should also state that iscsi support from Solaris & Linux has shown to be far more stable then iscsi support direct from the storage kit itself. That’s mainly cause the storage kit usually has a crappy CPU and the ‘i’ part of iscsi makes it relatively costly.

Now, the other method (the ‘Enterprise’ way) is if you DO get a pair of fibre channel switches. These cost around 10K each which relative to your SAN purchase isn’t THAT much. You setup your SAN’s so that their connections are redundantly split across your two FC switches (or ‘Fabrics’ as the storage people like to say). Now you’ve got 2 SAN’s connected via diverse paths to two switches.

From there you take your actual hardware and install a dual port FC card (~$700 each) in each host. You set that up to terminate one to each switch (diverse paths to both SAN’s via both fabrics).

For both the iscsi and FC option basically any decent card will offer a BIOS booting option. That’s where your hard drives onbox become irrelevent (except maybe for swap) and the BIOS treats the SAN/FC volumes like a normal hard drive.

Replication wise, as usual, you have the costly and cheap option. The costly option is to get SAN’s with replication capabilities out of the box. They take care of each other and you never have to think about it again (hopefully). Vendors used to charge a lot for this but the new series stuff (Hitachi & Equalogic for instance) have this shipping as standard. If you’re going down this path you want to look for a SAN which has the capability to make ‘virtual fabrics’ seamlessly migrating between the two on failure. Otherwise your boxes are going to need a reboot or reconfiguration if your backend fails.

The cheaper option is to export two independent LUN’s (1 from each SAN) then use whatever replication you want on box. On Linux this is software raid or mirrored LVM, on Solaris this is ZFS (and it’s awesome :)). Then if you have a SAN failure the idea is that all your boxes will register a ‘dead disk’ and recovering involves getting the SAN back to normal and resyncing from the good disk.

Hopefully that answers a few of your questions. There’s some more tidbits in a Cluster 101 doc I did.
Stu

And now..

Perlboy — Tue, 27 Jan 2009 14:15:36 +0000

I’ve upgraded Wordpress to the most recent version, ditched the broken template and loaded this one.

Cleaned up dead users, cleaned up crap comments.

Hopefully things are much nicer now.

Don’t you guys have anything better to do?

Perlboy — Tue, 27 Jan 2009 13:39:30 +0000

Ok,

So, I was looking at tarvos.seekbrain.com (the host that serves these pages) and it was sitting at a load of 1.5 constantly. This box is only a single proc hyperthreaded box so for something which overall gets very few hits this was strange and when I was trying to get some images (for my home photo collection) it plain locked up.

Initially, I thought, well, ok, what’s it’s uptime? 620 days. Hmm, at this point it’s not unusual to see ‘crap’ introducing 0.5-0.8 of the load average (specially with hosting machines constantly spawning processes). With that in mind I did a full update of all packages including a new kernel. A reboot later (after Redhat’s initrd.img creator was kind enough not to add support for my mobo) and it was back online BUT, still 1.0 load. What’s going on here….

A little investigation and it was revealed that my somewhat defunct Gallery install was copping an absolute flogging from comment spam. Joy, so, with comments disabled I went to clean up this mess. Running the uninstall option has triggered a DELETE and OPTIMIZE so things will be a bit slow but at least I’m not adding to the quagmire. After 660,175 comments in the past 48hrs it was no surprise this poor little box was choking.

So, the lucky top hitters are as follows are as follows:

[root@tarvos logs]# cat /tmp/newspams2 | sort -nr | head -n 10
128475 access_log.processed:195.225.177.40
122578 access_log.processed:195.225.178.21
67396 access_log.processed:84.16.227.86
36641 access_log.processed:83.233.30.39
18589 access_log.processed:200.63.42.133
15927 access_log.processed:67.19.22.194
9139 access_log.processed:83.233.30.34
6814 access_log.processed:62.90.102.146
6225 access_log.processed:62.90.102.148
5376 access_log.processed:195.225.178.19
[root@tarvos logs]#

With a bit of foo and absolutely no intention of wasting my time further with people that can’t control their botnets:

[root@tarvos logs]# iptables -I INPUT -s 195.225.176.0/22 -j DROP
[root@tarvos logs]# iptables -I INPUT -s 195.225.177.40/32 -j DROP
[root@tarvos logs]# iptables -I INPUT -s 195.225.178.0/24 -j DROP
[root@tarvos logs]# iptables -I INPUT -s 83.233.0.0/16 -j DROP
[root@tarvos logs]# iptables -I INPUT -s 195.225.176.0/22 -j DROP
[root@tarvos logs]# iptables -I INPUT -s 62.90.0.0/16 -j DROP
[root@tarvos logs]# iptables -I INPUT -s 200.63.40/22 -j DROP
[root@tarvos logs]# iptables -I INPUT -s 200.63.40/21 -j DROP
[root@tarvos logs]# iptables -I INPUT -s 84.16.224.0/19 -j DROP

Good bye a good million or so IPs, I don’t want you don’t need you and you can kindly pissoff and die.

Stuart

Goofi!

Perlboy — Wed, 21 May 2008 13:16:57 +0000

What do you do when you have a GPS receiver and a wireless card? Well, naturally, you hook it up with Google Maps and see what it looks like.

Introducing, Goofi! It’s all very Betaish now but basically it does the following:

- Collection occurs using some rather nasty collection scripts correlating GPS coordinates & iwlist scanning information and storing it into a flat text file. I could have used Kismet here but the target platform I’m thinking of won’t have an X11 terminal so that’s not too available.

- Data is dispatched to a DB upload script

- Data is displayed through an XML generation script and made clickable to the record data.

Things todo:

- Make collection much more automated. I’ve got a fanless 18V based machine I’m thinking I might be able to hook up to be powered permanently from the car. That’ll mean I can enable ‘constant’ wireless scanning to any destination driven.

- Correlating ‘commercial access’ wireless points into some sort of grouping component (ie. “Show me Telstra access points around this area”).

- Using MarkerManager better with some sort of ‘clustering’ component (ie. some circular geometry to identify ‘clusters’ and show them)

- Open sourcing this and inviting people to submit data. It’d be pretty cool to get a whole bunch of contributors submitting data around the world. Maybe then I can go wherever and NOT have to go hunting for commercial wireless points to check email!

As always, any thoughts, suggestions welcome (use email link on the page itself).

Stu

Toshiba M600 Xorg Configuration

Perlboy — Wed, 28 Nov 2007 16:05:05 +0000

Heya,

For those with a Toshiba laptop (one of the ones with a native resolution of 1280×800) and an inbuilt Intel i965, I thought I’d dump my xorg.conf here. I found this device reported a non existent TV output which was stuffing up X’s ability to maintain the correct resolution. Remember, you need to have i915resolution setup correctly as well, in Ubuntu this is available as a package. I have manually set the modes and depth settings and disabled auto but your results may vary.

With this configuration I have compiz working on Xubuntu (ie. XFCE Ubuntu). Now all I need to get working is sound!

Attached configuration file: Xorg configuration for i965 on Toshiba M600

Have a good night!

Stu

New Laptop

Perlboy — Mon, 12 Nov 2007 16:18:13 +0000

Well well, sending this from my new laptop.. Will be uploading last update soon.

Adios,

Stu

Final leg: Part 4

Perlboy — Wed, 07 Nov 2007 01:38:21 +0000

Bonjour!

On the 23rd we had a sleep in and in the afternoon explored Domfront which is an ancient walled village and was very pretty. The next day we went to Arromanche to see where the artificial floating harbour was constructed by the Allies and we also visited the 360 degree cinema which provided an insight into how tough it was for the troops. We went on to see the American War Cemetery and the futility of war was clear with the number of crosses which commemorated lost troops by name, and those known only to God. It was an excellent day and we will definitely need to go back to see the rest of Normandy. The 25th we had lazing around Adele and Ian’s house which was a welcome break and gave us a chance to sleep in and unwind in the fresh country French air. Howard picked us up in the afternoon and we had a delicious dinner with Carole and Howard and their friends.

On the 26th Howard dropped us of at Flers to catch the train to Paris and then we mastered the metro to get us to our hotel without too many wrong turns! We had a bit of a rest before heading out again on the metro to visit the Eiffel Tower, Arc de Triomphe and Champs Elysee. On the 27th we went again to the Eiffel Tower, this time to go up, it was so busy but the view was worth it. We then walked to the Rodin Museum which was really lovely and then caught the metro to the Catacombes which was pretty creepy. The catacombes were a lime stone mine but were converted to a graveyard in the 1800s for hundreds of thousands of Parisians when the cemetery was full. The 1.7km of tunnels are stacked with bones arranged in macabre patterns which was a bizarre tribute to the dead.

On the 28th Stuart and I decided to venture out to visit the Palace of Versailles (or Ver-silly as Stuart pronounced it). We waited for about 2 hours to get through the tiny entry doors and were royally sick of queues by the time we got inside and had to queue for the audio guides (10 minutes), the ladies toilets (20 minutes) but of course there was no wait for the hommes (or homies according to Stuart). The rooms were spectacular but the grandeur was a little lost due to the over crowding. For Sally’s birthday on the 29th her French friend Audrey came up for the day to visit and we had a great day, despite the rain. We visited parts of the Louvre and luckily avoided the enormous ticketing queues by buying our tickets elsewhere (the French are more helpful to the French!) After that we had a delicious lunch in the latin quarter, finally a traditional French bistro meal Stuart was after. We aren’t sure what we ate, even Audrey didn’t know what was on the menu, but we know it was an entree of foie gras something and then a chicken casserole with potato gratin. Audrey was quick to inform Stuart that pate is for the country people and foie gras is best. After lunch we went to Montmarte and much to Stuart’s disgust, traipsed up the stairs to Sacre Couer instead of taking a perfectly good lift.

On our last day in Paris we visited Notre Dame and went shopping where we bought Stuart a new suit which he plans to wear to work occasionally (show off). We had dinner on a budget that night because we only had a few euros left and didn’t want to withdraw anymore. We went wandering around off the beaten track and managed to stumble across the male gay bar district. Stuart couldn’t get out of there quick enough! We eventually found a nice looking Chinese buffet (seemed that way through the window) which was really cheap so we went inside and sat down, before noticing that people were heating the buffet food up in the microwave. I don’t think that would pass a food safety inspection in Australia but we didn’t get food poisoning!

Early the next morning we caught the train to England and were glad to be leaving Paris, where when you are wandering the streets with a crepe in your hand you need to be careful of the puddles on the footpath because you are unsure if it is water or pee. We were so glad to arriveat the hotel in London as the bathroom was spacious and tea and coffee were provided in the room with UHT milk! How exciting! We didn’t have a lot of time in London so we dumped our bags and headed out to catch one of the double decker buses around the city. The weather was great and we saw sights such as this and this and went shopping on Oxford St. Sally was excited to visit the places on the Monopoly board! That night we caught up with one of Sally’s friends and had dinner at Southbank which was great fun. The next day we got back on the bus to finish the tour before heading to Waterloo station for what we were expecting would be an easy trip to Heathrow. We originally only had to change trains once but due to a train being stuck in a tunnel somewhere the direct line to Heathrow was blocked and we had to go back up and down millions of stairs to catch another 3 trains. It took us 3 hours and Stuart was buggered after carrying what we found out at the airport was 50kgs of luggage. Not fun, but cheaper than a AU$130 cab ride!

The flight on Singapore Airlines was excellent, we still didn’t manage to get an exit row but Stuart had just enough leg room so was reasonably comfortable. Our time in Singapore was very busy, our hosts, Stuart’s friends from uni, didn’t disappoint and we saw a lot of sights in a short amount of time, and managed to avoid the tourists. The first night we had dinner at the best crab restaurant in Singapore and we had way too much food. Stuart went to a nightclub that night and paid AU$40 for a jug of rum and coke and was flabbergasted! The next day he slept in while Sally had coffee with a friend and tried egg jam, which was a specialty. That afternoon we went electronics shopping with our local tour guide and were blown away by the 7 floors of electronic stores. We then went to ‘Little India’ where the locals were celebrating their new year. Unfortunately we didn’t have much time to see the sights but we did have delicious curries for dinner. After that we went out to the Singapore Zoo for a night time safari which was excellent. No need to go to Africa now, we have seen elephants, tigers, giraffe, flamingos and hyenas in Singapore! We went for ’supper’ after that which is a normal Singaporean thing to do which basically meant having a second dinner at midnight. It was another feast where we tried stingray, sugar cane juice and some really different ice desserts. The Singaporeans were great hosts and it was a lovely end to our holiday.

Now we’re back at work! Was a crazy 5 weeks and we really enjoyed ourself!

Stu