Some of you might have noticed a hiccup on the blog this morning. That was caused by one of the disks failing in a way that hung up everything on the system. The system uses software mirroring, but when a disk gets serious read errors, it still blocks the process trying to do the read while it waits for a timeout. Unfortunately for us, the process was the kernel. I have checked the disk since this morning and it’s remapping sectors still. It has been spinning continuously for nearly 4 years straight, so I think it’s time to send it out to pasture. I’ll be doing the replacement tonight, probably around midnight.
5 thoughts on “Failing Disk”
Comments are closed.
Cisco might be having a sale?!?!!?
Wondered if the Brady Bunch were DDOSing you; I couldn’t even ping you.
Couldn’t ping? Thanks. That’s another piece of the puzzle.
And I don’t think the Brady folks are sophisticated enough to do a DDoS attack :)
I’m surprised you host this blog yourself. Seems like you could find a hosting company with reasonable rates and leave problems like failing drives and DDoS attacks to them to worry about.
Why would I do that? I trust my IT skills a hell of a lot more than I trust the support at any provider who charges rates I’m willing to pay. There’s also less danger of getting my site switched off because someone filed a complaint or claimed I violated some copyright. I’ve heard enough horror stories from other bloggers dealing with hosting outfits that really don’t care about them or their blog to think it’s a good idea. I already pay for a 25 megabit circuit anyway.