…and around it goes

April 2, 2007

After the weekend

Filed under: Personal — steve @ 9:11 pm

Unfortunately I did not get to spend as much time with my daughter over the weekend as I had hoped. Again I was sucked into the data center at inopportune times. I hope to make it up to her over the summer, winters have me in front of computers seven days a week and many hours a day. I’m not much fun to be around winters. I feel bad as I was looking forward to spending more time with her (I hate being a part time dad).

Winters are rough because I try to plan all new development work in the winter. It’s usually quite a bit of work. It’s not just development, but testing and support as well. Any change at all, no matter how minor, seems to break something for someone. For example, those critical security patches that have to be installed ASAP and always break something (PHP is good for this).

But even worse than those patches is building out a new back end. So much change, so many variables. Even with months of extensive testing I am bound to miss something. For the most part this one went smoothly, nothing exceptionally major. But it did have it’s issues, ram failure, suicidal daemons, file perms, and such. Enough to keep me hopping. The bad ram this am was just icing, especially since it tested fine before install and ran fine all weekend.

An additional aspect that also contributes to the load that some may not be aware of is the “every problem is a result of the upgrade” phenomenon that occurs after any change. Techs know it, if you work on someone’s printer one day and their CD ROM drive stops working the next, it had to be because of you working on the printer. The same thing happens with a service. Don’t get me wrong, I am not saying that no problem is service related after an upgrade, that is obviously untrue, but not every problem is related.

Someone changes something, tries to log in using all capitals, or has a network problem and it is very hard to get them to look at anything but the upgrade as a possible cause. This adds to the confusion when I really need them to test something so I can eliminate either end, and they refuse because it has to be on our end. I can’t adequately troubleshoot an issue with only one half of the information.

I admit sometimes these things are very tricky, just as it can seem all on our end to a user, it can seem all on their end to the service. The problem is that an honest 99% of helpdesk tickets come back to something on the user’s side. A configuration issue is the most common. I try very hard not to fall prey to that because I know what it is like to call support and have them refuse to look at anything but my end.

So I work to gather all info from both ends and look at it as a whole and as a new issue. But that gets frustrating if I can’t get the person on the other end to even look at anything on their side, leaving me with only half the data involved. It can make a rough week or so on the helpdesk after an upgrade. (please note that I am not pointing at anyone, I have fallen prey to it myself, I’m just explaining one of the frustrations involved with an upgrade).

So this is why I save this stuff for the winters. I used to ski, but had an accident racing (I used to race GS back when I was in college, well actually Bridgeton Academy, it was after HS and before college. My career was very short as I wracked up trying to show off for some local girl…Bridgeton was an all male school then, I would have tried to show off for the lunch lady).  I have not skied since, so I shut myself inside and code. Not much else to do in NE winters, especially NE city winters. I’m not a bar fan at all, consider going out to dinner a waste of money, and have a home theater system specifically to save me the costs of going to the movies.

Summers, however, are far different. Summers are for trips on the bike, 500 mile rides for no reason other than I wanted to know where a road went or wanted to see the sun rise or set from somewhere specific, taking the kids to local tourist traps (Plymouth Rock, Faneuil Hall, Salem, the fishing villages on the coast, the cape, etc), trips to the cottage, BBQs, sitting on a deserted section of beach with a fishing pole, and anything but sitting in front of a computer all day. So by summer all the new dev should be ironed out and then I run things from anywhere I am, even my bike. That is usually the plan, anyway.

But winters, they suck. I am very glad this one is nearly over.

2 Comments »

  1. Hope I’m not being too nosy in asking, but what kind of RAM was it that turned out bad? Or would you prefer not to mention the manufacturer? I’ve been lucky over the past several years, haven’t encountered any bad memory at all, now that I’ve said that I’ll probably have a streak of bad memory crop up, lol. Speaking of memory, what do you think about those copper memory ‘coolers’ that snap over the module? Are they worth the bother? I’m heading to bed, time for ‘zzzzz’s’! Adios!

    Comment by mkj — April 2, 2007 @ 11:20 pm

  2. To be honest, I have no clue. It’s been backup ram for a while, bought a long time ago. The server needed another gig, I tested it and it tested fine, I installed and no good.

    As far as the ram in sp1, also died same day, different data center, different type. I attribute it to gremlins.

    It goes in stages, I like the ram stage better than the disk stage. Failing disks are far less fun. It’s always something.

    Comment by steve — April 3, 2007 @ 12:29 am

RSS feed for comments on this post. TrackBack URI

Leave a comment

You must be logged in to post a comment.

Powered by WordPress