Archive for August 2nd, 2010

And now its just silly.

August 2, 2010

Well behind on the blog posting with the recent burst of work. Sounds like good news hey? However of course the news is not all good. Is it ever? Its all a bit mixed in… as one of my more annoying friends says “point of actual fact”:”

Our install in the city met a snag. Some issue with the multiple websites hosted on their IIS server meaning it did not respond in exactly the same way that our test website on our test server did. Some issue nearly impossible to duplicate or test for but something that has to be resolved. Of course hosting one website should have nothing to do with hosting another… address resolution and redirection issues aside but there we go. Its a MS issue .. if I were inclined to be generous I would say its a programming issue.

Also my carefully worked out backup procedure did not work. Well it does.. so obviously I documented something wrong. That won’t relate to a software difference, its just a poor documentation by yours truly issue. I,ll wear that. I will sort that today.. the plan was to do it over the weekend but things got in the way. Namely data recovery things

So today I shall mainly talk about data recovery. This particular mac was reformatted and rebuilt from a factory install image. Right over the top of a working machine. Lost data

Sometimes you can go back in time to an earlier format version and retrieve the machine entire. Or at worst entire directory structures. The problem in this case seems to be that the install disk installed a much worked on and modified image complete.. even to ghosts of former partitions. At least that’s what I think happens.. my software detects 50 or more former partition schemes. Not easy.

So pull the data off anyway you can. Photorec pulls everything that looks like a file off the complete drive. Everything. A very very small percentage of which is useful. And for reasons I am not clear on dumps it all into a vast number of directories. Its a time consuming and slow process. And getting all that data into one single directory so we can do searches and separate out the useful stuff is really annoying. I must find out why it does not do this for us

Normally we pull the drive.. put it into one of the servers with whatever adapters are needed to make it work.. and have the server.. probably the machine I am writing this on.. a linux box running ubuntu.. pull the data down. Its a slow slow process.. usually taking 40 hours or so.. with 200 Gb or better drives around. This was a new mac under warranty.. not really wanting to pull it apart. But with Macs their is an alternative. Most Macs have firewire. And they have this ability when connected to another Mac by firewire to be booted up in what they call Target Mode. In which the entire Mac just becomes a big external hard drive to the other machine. Very very cool. And so my macbook got the job.

Its a lot of data. I wrote a program (script) to drop it all into one dir and thats still grinding away. Meanwhile I am missing my laptop. I have to write blog posts sitting up at the server instead of lying on the floor. Makes me far less productive. You can see why I work for myself can you not? Anyway its so much data the poor little Macbook is seriously struggling. Which is itself depressing and time consuming.

Pulling the data off this way has one huge drawback. You lose the file names. The reason for this is that the file-name is essentially a label pointing at a location on the drive. We are pulling the data directly off the drive. So the label becomes another separate bit of junk. Start, and finish and whatever scrap of data is in between is salvaged and given some random file-name. Deleted stuff gets retrieved too. Think you zapped that bit of porn.. think again? Some files are easy to identify.. pics have the description of what they are within the file as do docs and the software is smart enough to pick up on that and label them with the appropriate file extension, which is how Unixy systems like linux and OSX do it anyway. When we hand this stuff back to the client we just politely ignore the stuff that is embarrassing. Not in this case btw.. this clients stuff is squeaky clean but in general. Rarely have to look at more than a few things… just have to confirm that I am recovering what I think I am recovering.

Recovered 200 Gb of pics for Kathy’s former partner last year. Not a single filename. Backups are such a good idea. Still with a browser allowing thumbnails quick sorting of data is at least achievable. Keywords can be searched for in docs. Nearly all the straight text recovered is rubbish. All better than nothing tho.

Kathy and I took the motorbike into town for this job on Friday. While we were their we bought a netbook for my friend Robyn’s visit to Sara in the UK. Sorry Sa… Apple don;t make a netbook yet and an Iphone was not going to cut it. We paid retail, wholesale seems pointless these days.. costs more than you can get retail a lot of the time, so many good deals out there We came back.. with Kathy on the back of the motorcycle with thousands of dollars of computer equipment between the two of us. Backed up at home of course

Anyway this post again reiterates my point. Backups guys… backups.

I,ll post again this evening.. much catching up to do.