jaredhirsch

diff --git a/lockdown.json b/lockdown.json
index 9f12177..a9441b3 100644
--- a/lockdown.json
+++ b/lockdown.json
@@ -27,7 +27,7 @@
     "0.4.2": "ed0a831f7b39156b63ae2e9d95e7c1289bb5c73e"
   },
   "bcrypt": {
-    "0.7.3": "7523db1fdd8b0bda337bade63b5b90a7ee9c448a"
+    "0.7.3": "*"

jaredhirsch

this supposes your computer + phone are behind the same wifi.

initial setup

on the computer:

1. npm install -g weinre
2. create .weinre/server.properties following docs, except inserting your computer's LAN IP (eg 192.168.0.5):

boundHost: 192.168.0.5 <-- your compy's local IP here

jaredhirsch

/*
First column: test name
Second column: time until onready is fired from communication_iframe
Third column: time until "can_interact" from button press (dialog ready time)
*/

var data = {};

// 2G
// ======

jaredhirsch

{
    "state": {
        "name": "first-try",
        "defaultGraphParams": {
            "width": 400,
            "from": "-30minutes",
            "until": "now",
            "height": 250
        },
        "refreshConfig": {

jaredhirsch

This gist explains how to create a generic ELB monitoring dashboard in graphite 0.9.10, which allows import/export, but not via dashboard UI.

The python script attached to this gist should work, but it didn't immediately work for me, so I just did the following via the python repl:

Python 2.7.2 (default, Oct 11 2012, 20:14:37) 
[GCC 4.2.1 Compatible Apple Clang 4.0 (tags/Apple/clang-418.0.60)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import json
>>> import sys

jaredhirsch

summary

Suppose we have a 5.1 cluster and are upgrading to 5.6

• Cross-version incompatibilities to worry about
  • 5.5: backwards-incompatible binlog changes
    • you cannot put a 5.1 slave on a 5.5 master
  • 5.5: changed string/number conversion algorithm
  • 5.6.4: backwards-incompatible timestamp/datetime storage format changes.
    • upgrading isn't required, but restoring from backups is 'more difficult' unless you have a .frm file
• todo figure out wtf this means

jaredhirsch

linkdump

schema migration and data modeling

CQL catchup: http://www.slideshare.net/patrickmcfadin/become-a-super-modeler

great advice from @bbangert:

• things to remember with C*:
  • writes are super cheap, everything is super denormalized, data is written for optimized reads
  • the last part implies that if you need different ways to access all or part of the data, you write it.... again, in the other style of access you need
• on migrating existing data:

jaredhirsch

helenus at the node repl: (because i keep killing my session with uncaught exceptions...)

// start connection
h = require("helenus")
p =  new h.ConnectionPool({hosts:['localhost:9160'],cqlVersion:'3.0.0'})
p.connect(console.log)
p.cql("USE browserid", console.log)
  
// couple of helpers
function getValue(r, key) { return r && r[0] && r[0].get(key) && r[0].get(key).value }

jaredhirsch

moving a blog to S3 + cloudfront

1. have a static site generator
2. move data into S3?
3. make S3 bucket the cloudfront origin server?
4. flip DNS?
5. 🍻?

jaredhirsch

// remap exports.foo to exports._foo, log arguments
// passed to foo, then call _foo.
for (fun in exports) {
  (function(f) {
    // bail if it's not a function I added to the exports object
    if (typeof exports[f] !== 'function' || !exports.hasOwnProperty(f)) { return; }
 exports['_' + f] = exports[f];