0xDEADBEEF

Well, it’s set back up at least

Uncategorized | Posted by attriel March 30th, 2013

OK, so, I’ve reconfigured things and picked out a new nice looking theme again.

Most of the themes are white backgrounds. Or massively over-engineered. Or photoblogging. Which isn’t really what I’m doing. And I hate the light backgrounds. Which reminds me, I should look at changing my terminal windows to green on black again on my new system.

Anyway, one of the updates I did was to hook up the Google Authenticator. Which then (apparently) doesn’t let me use the wordpress iPhone app. Actually, I’m not even sure why that’s a plugin not a builtin at this point. Two-factor is all the rage.

The effect is there are now “two” accounts posting. My primary and a mobile. Which seems stupid. So I may just end up not using the iphone app at all. Not sure yet. Because that’s dumb.

But, it’s only the first day back up. I have no idea what the schedule will be, since I have to decide what I’m posting about and if I want regular “features” or irregular posts or some combination.

Luckily, I’m probably most of the people reading this right now, so it’s not like I’m risking losing my audience by waffling. Honestly, if you’re seeing this, you have me on a feed-reader that suddenly updated. Because it’s been over a year.

Last year, shit went down. This year, hopefully nothing happens, and I have time to dedicate to this notion of blogging again!

Comments Closed

AWS Instance Types

Site Maintenance | Posted by attriel February 26th, 2012

So, currently I’ve been running on a t1.micro on-demand instance for my three blogs (yeah, I keep the topics distinct, since cooking, programming, and generallized life, don’t really have the same audiences; between the three of them I probably make one-two posts a week when I’m remembering to write anything)

Anyway, I’ve been using that setup b/c it’s inside the AWS free-tier ; but with the problems I’ve had recently (see last post), I’m not sure if it’s the on-demand-iness, the micro’s virtual disk instead of an s3 distributed drive backing (my money is on the on-demand bit).

So I’ve been looking at the prices. I looked first at the low-utilization micro instance, since I really don’t see enough traffic to qualify as “high utilization,” but apparently in AWS-speak, utilization is “how much would this system be online.” Since the blog/sites would be on all the time, that’s “high utilization”

Current prices are on-demand at 2c/hr, so 1yr is 175$

Light Reserved is 23$ registratoin and 1.2c/hr, total 130$

Heavy utilization is 62$ and .5c/hr, 105$

To go up to a standard small (instead of micro) for the s3 backing:

on demand – 8.5c/hr, 745$

light – 98$ & 5c/hr, 535$

heavy – 276$ & 2c/hr, 450$

So … yeah, apparently it’s not quite as cheap as I was thinking for the year on AWS, but even if I split out across three instances for the various bits (my blogs, a couple sites we host, and a forum we run) it’d be ~300$/yr, opposed to currently we pay ~80$/mo on the business internet to run those things . Still a net-savings of 400$ give or take. And a massively higher throughput ; Not to mention that we can upgrade our home network from 1.5/1.5 business to probably 25/5 and STILL save 50$ a month (there’s more on that bill, so I’m not sure which parts exactly would be removed, but I’m sure it’s more than the 300$ I’d be paying on the new structure)

So, there will likely be some flickering as I move to the reserved instance in the next month or so; I’m really hoping it helps the occasional choke-out I’m seeing. If not, then I will DEFINITELY be putting in a support question

Tags: aws
Comments Closed

AWS Issues v2

From The Lines, Site Maintenance | Posted by attriel February 20th, 2012

So, the setting I have is for the “On Demand” instance, b/c that’s where the free tier is AFAICT.

But it seems like maybe it’s having some issues with the 24/7 uptime concept, b/c the other day it was mentioned that my main blog had flaked and browser was declaring “unknown!”

Well, sure enough, couldn’t ssh into the server either, weird. All the status checks said good, but the system was definitely not responding well. So I bounced it from their admin interface, it came back up, and has been happy for a few days now.

So, I’m guessing that it doesn’t actually like being on all the time, maybe it swaps out when it’s not being used, and eventually gets GC’d even though it’s “active”

Not sure why the status notices said clear though …

Anyway, that’s a potential issue I’ll need to figure out better before moving anything bigger over.

Tags: aws
Comments Closed

Server Upheaval

From The Lines | Posted by attriel December 24th, 2011

Be careful what you ask for, thats the moral here.

Months spent asking after a new server with more drivespace. What happens? There is no server, make do. Can we have additional storage here? Well. … we could put you on the NAS. GREAT! Except we don’t know how to hook that up to anything, and that seems so tedious to figure out. And then we wouldn’t have all these pristine unused drives!!! OK, fine, USB? Oh ,no, you don’t want USB. It would be too slow!

How is NOT HAVING MY SYSTEM an improvement over “it’s too slow”???

So, we find things we can cut and trim, seriously months of this, trying to get it to fit within the specifications we were given as already available to us. It’s not pretty, huge chunks of functionality are marked off with “there’s no space to store the data this requires, come back later”

Well, two weeks before launch, someone got a bee in their bonnet that “OMGZ! the data isn’t on the system! POMG!!!!” and lit a fire under people.

This led to, a week before launch, we get told that they’re giving us a new server. Somehow they FOUND one, and it has 1.5T instead of 700G of drives. IT’s got more memory, more HDD, everything. It’s like two generations ahead of what we got the first time. GREAT! A little late, can we … NOT move the system? Since you need time to set it up still?

Oh, no no no! That’s not acceptable! We need to move to the new system so all the data can be available immediately! Oh, and since it’s so big, we’re switchign to 64bit architecture. And recompiling all our stuff for 64bit is silly, so we’re going to upgrade everything to latest and greatest.

And it should be up by the day before launch. We’ll move you over then. No, there’s nothing in the dev environment.

So, seriously, we’re looking at a brand new server, with all new software versions, on a new architecture, no more than a few days before we launch.

I’m so sorry we asked for more drive space, because this is going to turn into a COMPLETE CLUSTER!

Comments Closed

AWS issues

From The Lines, Site Maintenance | Posted by attriel November 21st, 2011

So, apparently your IP is static as long as your instance is running.

BUT! The instance hung/crashed a week or so ago, and when I restarted apparently it got a new IP! oops.

So I finally hooked up the “Elastic IP” settings, and we’ll see if that’s static or not.

If this works for another week or two I might look into offloading more stuff. Because honestly, the home business line is costing way more than it’s worth. I can cut the price by half and raise the speeds by 10x …

Comments Closed

Amazon AWS/EC2

Site Maintenance | Posted by attriel November 15th, 2011

So I’ve been looking at Amazon’s AWS tools, specifically EC2, and they look really neat. It’s tempting to move some of my home based sites (specifically the ones I host for other people that get enough traffic to slow down my access) off onto an amazon site.

I started with experimenting with some of my blogs, and didn’t find any significant issue with the process.

I want to look at all the docs, find out the benefits of using amazon’s RDB system, elasticache (memory caching) etc.

Charges are based on having the system, then network traffic, disk space and disk access, for the simplest construct. So the question arises — if I hooked up a cache, could I keep everything in cache and have less disk access? would it be more efficient costwise? at a guess, no, I doubt all my sites combined could combine to anything noticeable on the scales those are probably structured at.

OTOH, running a business, it seems a sight simpler than the old business line concept, and then you can just hook it into their CDN and even distribute the site amongst their data centers so processing is done in local spaces (which I presume is when cache and their DB system come in really handy)

Comments Closed

Clearly … failing

Site Maintenance | Posted by attriel September 14th, 2010

So, I haven’t posted in nearly two months. I’ve missed 3 of my intended post-times.

See, here’s the problem I’m having. When I get home, I play computer games with my son, then frequently spend the rest of the evening trying not to move too rapidly a my daughter sleeps on my back. I haven’t been having the time to play with fun things lately at home. Heck, I only manage to do the dishes a couple times a week (and we really have enough to do it every day and a half)

At the same time, I decided early this year that I wasn’t going to post about the stuff I’m doing at work. First, there’s all the various clauses about what we can or can’t say and how we can or can’t represent our selves, our companies, clients, allies, and bitter enemies. And a lot of what I’m doing right now is very specialized bits of stuff. It’s also quite cool.

But, if someone went looking, it would be trivial to link my work to this blog. Not that I work that hard to hide my name, but if I start talking about all the widgets and sprockets and cogs, it becomes a blog about work. And that’s where the problem comes in

So, for now, I’m again going to put this blog in deep-freeze. Add it to your google reader or feed subscriptions, and then when it comes back you’ll know about it.

Originally this blog started as a place to talk about the seriously geek stuff I wanted to discuss on my main blog, but that none of the people I knew would really be interested in. It was more for me to have someplace to talk. Then I decided I was interested in explaining and tracing crypto. A kind of how-to primer. B/c I can’t be the only kid who found that kind of thing interesting

But when I got to DES … well, that stalled the blog the first time. And the second time. So I broadened it out. Time constraints mostly killed it that time. And time again now. I have a bunch of interesting stuff I’d like to post, but I don’t know when I could make it live anyway, so it hardly seems worthwhile.

And it’s not like I post on my personal blog either, so, maybe I’m just past the blogging phenomenon and i’m “old” now … So maybe deadbeef will come back in a new form. Not sure what that would be, but since it still requires that I have time to do it … could be some time

Comments Closed

Regex is COOL

Programming | Posted by attriel June 16th, 2010

We started on a Tuesday this month didn’t we? damnit!

Well, on an unrelated note, Regular expressions kick ass. Here’s one I cribbed together to break things on :-separated pairs with optional preceding un-paired text

We started on a Tuesday this month didn’t we? damnit!

Well, on an unrelated note, Regular expressions kick ass. Here’s one I cribbed together to break things on :-separated pairs with optional preceding un-paired text

We started on a Tuesday this month didn’t we? damnit!

Well, on an unrelated note, Regular expressions kick ass. Here’s one I cribbed together to break things on :-separated pairs with optional preceding un-paired text

‘/(.*(?!s+w+:|$))(w+):(.+(?=s+w+:|$))/U’

Also, zero-length lookahead is sick and twisted and I liky!

Comments Closed

KCacheGrind

Tool Tips | Posted by attriel June 1st, 2010

Well, here’s the large KCacheGrind pics. In order to get good pics, I installed WordPress and logged it!

So, here we go.

The first image is from the installation script. I created the db & user, and unzipped the files. This is what happened from the wp-admin/install.php script:

The top half has the display of “caller map”; it shows everything being called. Each box is scaled to the quantity of time used by the function.

The second half shows the “caller graph”; at the top is “main” and then it trees to each call, which calls down.

The left-hand is the plain listing. It shows a lot of the same data from the trace files. Inclusive-time, is the first column. This is the complete time spent in this function, including all subs. The second is the amount of time spent actually in this function, followed by total number of calls. The last one is the function name, the very beginning is a bar-graph representation of the total time.

The second image was from the admin login page.

This is a similiar picture to the first. The top is the callermap again. The bottom shows the functions called. In this entry, we’re seeing the callees off of main.

The third image is from adding a new post the the blog.

This time I zoomed in on a function (WP_Object_Cache->get) , which is why the upper panel is so empty. There are actually two sub-calls up there on the bottom edge, but they’re really small, hard to see.

The lower pane is the “caller map”, which is kindof the opposite of the upper panel. In this case, everything that calls the Cache Get, which in this instance is a huge number of calls from “main.”

The fourth picture is from the homepage of the new installation.

I cropped it to just the upper pane. This gives us a good view of exactly HOW MUCH goes into the homepage. With only two posts (the default post and the one I created), it is just a huge listing. The dark-blue speckled areas are where it ran out of things to fill into the field, because it couldn’t always slice the page in reasonable ways.

I never quite was able to configure the system to make thumbs for me, so the images above don’t link around to the full-size pics. The gallery is inserted below, but I couldn’t figure out how to get the thumbs out of the gallery, so everything is listed twice. Sorry.