Quote:
Originally posted by VAXman
Yikes! IDE! What a complete brain<casual obscenity obscured>!!!
|
heh. I actually was able to convince a local company to toss their alleged 'High availability clustering solution' from Our Favorite Vendor by the simple expedient of pulling the drive on the central test server. After the windows admin required 12 hours to rebuild the cluster (the entire thing crashed) from the ground up, they listened to reason and installed something a little more robust.
Their windows consultant tried to say "yabbut you pulled the main DRIVE! You can't expect a cluster to survive that!" to which the head of IT said "we should be able to expect the cluster to survive a drive failure, no matter what the drive.
The windows consultant was heard cursing my name for weeks afterward. Not my problem.
Funny about that. While I was working at HP, they moved their five node VMS cluster to four different datacenters: one in California, one in Georgia, one in Dublin, Ireland, and another in Tokyo, Japan.
While individual nodes suffered downtime (they're moving the hardware, ferchrissake), the cluster suffered ZERO downtime. My sources that are still there report that it hasn't suffered a second of downtime, in spite of the fact that it's spread out over three continents (on two planets, if you consider California...well, we don't need to go there, do we?
)
I was impressed.
Quote:
Anyway, this geekery is beyond the subject of this thread.
|
Like that's ever stopped us.
Quote:
Respectfully, (VAX who ported VMS Host Based Volume Shadowing driver to Alpha circa '93 while M$ was still mucking with DOS)
|
Geh. I remember those dark days. That was about the time I was sitting at SGI, pondering how to make good my escape. God, that place was WAY too PC.
Roger -Dot- Lee, even the PC crowd thought so.