Archive for July, 2010

Why pay for a NetApp system?

Friday, July 30th, 2010

Simple.  Because disks can and do fail at 3:21pm on Friday.  With NetApp, the replacement disk is being delivered to the datacenter at the same time the overworked sysadmin is reading the email notification of the failure – which is at 4pm the same day.  Under 40 minutes from failure to part delivery!  Storage questions?  The Answer is NetApp.

And I had such hopes…

Wednesday, July 28th, 2010

I thought you were starting to get better.  I thought you were starting to gain a clue.  Now I know that it was a disillusionment on my part.

So, let’s go over a few things you’ve done wrong.

First, your major outage this week.  You planned to outage, notified everyone, then you let Nagios alarm CRIT when you powered the systems down.  Then you went and ACK’ed the alarm.  Well, stupid, you’re doing it wrong.  What you should have done is to schedule downtime for those hosts inside Nagios, not allow them to alarm and then ACK the critical down state.  In effect, you’re saying “it’s unplanned, it’s not supposed to be down, I’m trying to get it back online”.  No.  You’re an idiot.

Now I find another virtual machine that you installed and that you didn’t install VMware tools on.  Hey, stupid, I’ve told you before – not installing VMware Tools make the guest’s performance go to complete crap.  You really do need to buy a clue.  I’ll even give you a quarter for it!  And just for kicks, why the hell did you install CentOS on this machine, when we have effectively unlimited copies of RHEL available – you know, the version that has actual, *support* available?  The one that all the vendors test against?  The one that doesn’t take 5 extra months to release a major update because one person got into a snit and couldn’t be arsed to do his part for the release?

You are a moron.

Another day, another moment…

Friday, July 23rd, 2010

What else can you do but stare in wonder at someone who thinks that a VMware guest machine will automatically add another processor to itself if it notices the system load go above a certain level?  And someone who is supposed to be a technical resource, to boot.  Amazing.

How to be a paranoiac

Thursday, July 22nd, 2010

It’s simple, really – just ask one question:  “What is our plan of action if command <foo> fails?”

Seriously.  It’s good management.  Or something.  But yes, I got asked that question.  I said something along the lines of “Well then we try A, B, and C, and maybe D.”.  The real answer?  “We call the vendor and tell them ‘It broke.  Fix it.  Now.'”.

Really, how much planning for failure do you want me to waste my time on?

Hey, stupid…

Wednesday, July 21st, 2010

When your boss tells your sysadmin to sign up for a list you manage, you probably want to approve the signup.  I know you’re paranoid, I know you don’t want to admit you have problems, but generally letting a sysadmin onto your special closed developer email list isn’t an issue.  Especially when your boss tells the sysadmin to sign up for the list because that sysadmin is the only one who can actually solve any problems.

Today’s random thought

Wednesday, July 21st, 2010

This isn’t really sysadmin-specific.  Or even computer-specific, although it does reflect directly on a particular developer I have the misfortune to work with.  It’s more of a general comment on all work environments.

Your office is not your home.  You do not need all the comforts of your home.  You are not here to sit back and suck down a beer while watching your favorite sports team in a background window that happens to be full-screen on the 24 inch secondary monitor.  You are here to work.  That’s why they call it “work”.  You do not need to have your already-budget-restricted employer buy you your own sharding desk fan in order to do your work.  If your work environment is really that intolerable, then maybe you should seek a different job – or a different career.  In the meantime, how about you do your job and stop whinging about not having an ionic fan on your desk?  That would be good.

No, that’s not a good idea…

Thursday, July 15th, 2010

Sharing password between multiple users because you’re all having problems getting to a given shared folder is a Bad Idea.  Especially when there’s known network work going on in your building that has been advertised to everyone and has already hit at least one other group with similar symptoms.  Just don’t share your password.  Ever.  It’s bad juju.

It just keeps getting better…

Wednesday, July 14th, 2010

No, idiot, you do not disable network name lookups by removing the resolv.conf file.  That’s just stupid.  That breaks all sorts of things.  Nor does nslookup pay any attention to the nsswitch.conf file – it’s *designed* to do only network-based lookups, so of *course* it hangs when it’s invoked after removing the resolv.conf file!

Safety Clown says “Be Smart, don’t be Stupid!”

Wednesday, July 14th, 2010

So you need a server for multiple users to log in to because they need to use Adobe Professional software to deal with PDFs.  Why, in the name of the Little People, would you use a server from a different project that is designated as a non-interactive-login application server that bears no relation to the project you need interactive logins for??  Cripes, that’s just stupid!

Oh, really?

Tuesday, July 13th, 2010

No, you do not add an admin address to a personal blocklist because it is telling you that you need to reset your password because it is about to expire.  If you do that, the admins will get notified with an SMTP bounce, and will come down on your insignificant incompetent pseudo-programmer ass like a ton of bricks.  Congratulations ${FOOL}, you’ve just found the fastest-yet-known way to piss me off.