Karl B

Just a short post made from work because I’m amused.

A tool staff member walked into the IT room today and said “I need to get some data off a Jaz”. Those of you who were involved with computers in the mid/late 90′s may remember the failed Iomega Jaz drive. Apparently this person has some client data back from 2001 (a year before the Jaz drive was discontinued) which he needs to get off the cartridge.

The conundrum now is whether we charge him the £200 that a data recovery service would, while buying an old drive off ebay for £20 or not. Decisions, decisions.

The age old problem of, how do you remove the password off a Microsoft Word document when you don’t know the password!

At work today I found a fairly easy solution. In previous versions of Office (we’re using 2007) you could save a document as a .rtf file, reopen it and the protection would be gone. Unfortunately in Word 2007 it looks like they’ve added some meta data to the rtf file to prevent just that, also wordpad/rtf files did not retain all the formatting information which was a pain.

Anyway, the solution! Save your password protected document as a web page, edit the html file and remove the lines which looks like this:

<w:DocumentProtection>ReadOnly</w:DocumentProtection>
<w:UnprotectPassword>59D8CC1A</w:UnprotectPassword>

Save it, reopen it in Word, save it as a doc. Done! The great thing about this solution is that Word puts SO much formating data in the webpages (at a guess so they can be converted back) that you don’t lose any formatting like you would with the rtf route.

After mulling over my previous post where I talked about server monitoring at work a thought came to me. I’ve been wanting to get back into programming as I haven’t done any serious coding since I was at university over 4+ years ago (I feel that I’m stagnating in the learning department) so I took the rant and decided hey, a project!

My thought is this: Monitoring agents should not really be needed on a server. From my workstation I can bring up a list of events from a remote server using eventvwr or a list of services by managing a server so why can’t monitoring tools?

I busted out Visual C# and after bumbling my way around for awhile I found System.ServiceProcess.ServiceController which you can use to find all the services and their current statuses on a local or remote machine that you have permissions on. In the coming weeks I expect I’ll be talking more about this as I try to write some rudimentary monitoring software.

Karl Out.

PS: Please don’t point out free software online which already does this, I’ve been trying to think of a good project for awhile and now that I have one I don’t want to lose it.

Lately I’ve become more and more  disillusioned by the server monitoring at my work.

I work in the IT department of a company which has a couple dozen servers supporting ~500 people. For reasons unfathomable to myself (politics) an out sourcing company does the monitoring/patching/etc of the servers at our work, something which I know we’re perfectly able to do.

I’ll give you some examples of poor monitoring

• Drives failing in raid arrays and not being picked up until one of the staff on site goes into the server room and notices the error lights and then notifies said out sourced support.
• Services dying on important (database) servers and no one being notified about it, only for the staff to find out about it on Monday morning and having to scramble to fix it.
• Using ping as a (sole) check of whether a server is available.
• Insisting that monitoring agents be installed on servers to monitor them and once they are installed the monitoring still being substandard(/non existent)

And what’s worse is all that happens is they get a slap across the wrist (a talking to) when on multiple occasions they’ve broken the SLA.

Rant, rant rant. (Please note: These are my opinions only and do not represent the thoughts of my employer)