Two strikes, and you’re out

Back on 11 January 2005, I had a bit of a problem with my primary drive, which I “fixed” two days later with the manufacturer’s drive utilities. At the time, everybody yelled “get a new drive, it’s going to fail any second now”. I didn’t, because part of the fun of running your own computer is keeping everything going on a shoe string when you can. The other part of the fun is buying nifty toys when you want to, which is the stage I’m in now. And the reason I’m in that stage is that this morning I got another email from “smartd”:

From Thu Aug 11 02:31:23 2005
Subject: SMART error (CurrentPendingSector) detected on host:
Date: Thu, 11 Aug 2005 02:31:22 -0400 (EDT)
From: (root)
X-Spam-Status: No, score=-2.5 required=5.0 tests=AWL,BAYES_00 autolearn=ham

This email was generated by the smartd daemon running on:

host name:
DNS domain:
NIS domain: (none)

The following warning/error was logged by the smartd daemon:

Device: /dev/hda, 1 Currently unreadable (pending) sectors

For details see host’s SYSLOG (default: /var/log/messages).

You can also use the smartctl utility for further investigation.
No additional email messages about this problem will be sent.

I looked in /var/log/messages, and it had been reporting this every half an hour since 02:31 this morning. After a bit of googling, I found that a “pending” error means that there is a block that the hard disk wants to remap somewhere else, but can’t because it can’t read the block. It will remap it automatically when it can read it, or when it needs to write to it. I also found The Bad Block HowTo, which told me how to find out which file it is that’s giving the problem. Turns out it’s a bogofilter database that I don’t use any more. So I can remove it, secure in the knowledge that the hard drive will remap that block when it tries to reuse the now empty block.

But this means that the drive is getting worse, so I think it’s probably time to replace it. The drive is 80Gb, and surprisingly enough, I don’t think I need any more space than that. /home is only 52% full, /usr is 45% full, and the rest of the partitions are down below 25%. I’m not even sure you can still get drives that small. The problem I’ve found is that bigger drives run hotter, and don’t last as long. Oh well, it’s off to the web I go.

2 thoughts on “Two strikes, and you’re out”

  1. …a cow orker just bought a 110GB PATA Seagate for $30.00 post-rebate.

    This is a 8MB/7200 RPM model with *5 year* warranty. Just last year; they were all going back to 1 year. Guess sales were off…

    Me? I have a rack full of 9GB F/W SCSI’s in hardware RAID5 I’m soon going to deploy. I like the idea of multiple hot spares.

Comments are closed.