lee/hpr-knowledge-base
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Initial commit: HPR Knowledge Base MCP Server

Browse Source 
- MCP server with stdio transport for local use
- Search episodes, transcripts, hosts, and series
- 4,511 episodes with metadata and transcripts
- Data loader with in-memory JSON storage

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Lee Hanken
2025-10-26 10:54:13 +00:00
commit 7c8efd2228
4494 changed files with 1705541 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

90
hpr_transcripts/hpr0011.txt Normal file
View File

@@ -0,0 +1,90 @@
Episode: 11
Title: HPR0011: dd_rhelp
Source: https://hub.hackerpublicradio.org/ccdn.php?filename=/eps/hpr0011/hpr0011.mp3
Transcribed: 2025-10-07 10:16:44
---
.
Hello and welcome to this episode of Hack and Public Radio with your host, Operator.
This episode is going to be over DD underscore rescue and DD underscore our help,
which is a helper script for DD rescue.
And DD underscore rescue is not to be confused with DD rescue.
So here, from here on out, when I say DD rescue, I'm referring to DD underscore rescue.
So first off, what state does a drive have to be in?
You have a drive that picked up by the BIOS first off.
It needs to be at least picked up by the BIOS and that something can talk to it.
Be it hopefully, if it's an NTFS partition, windows can at least mount the drive as a readable drive
and see the partitions.
Or secondly, you can use something like F is to be able to display inside of units Linux
what the partition information is and see if you can at least pull that much off of it.
So the setup I have is IDE zero or the first IDE is the CD-ROM and second unslave
is going to be the backup drive.
And then that lead the other IDE channel open to I put the backup drive that I'm going to drive.
I'm going to actually image the bad drive and then the drive that's known good that I'm going to image two,
which is going to be the same size or larger, that's going to be the secondary.
So this will allow you to once you've actually made a good image of the drive,
you can go ahead without rebooting anything and push that image straight on to your known good drive.
Along with this, I've got the laptop to IDE and also the SATA IDE converter plugged into that same chain.
So if I need to do any of that prior, then it'll all be set up good to go.
Also, you'll want to, if you can format the drive, you're going to put the image on to something Unix Linux compatible,
EXT3 or UFS, whatever.
This will work a little bit better as far as mounting it and then we'll have to worry about closing it or cross-platform stuff.
So the drive's picked up, you've got everything set up.
And the first thing you want to notice is check FDisk or whatever partition manager you want to use.
And look and see where your partition started and where they stopped and things like that.
So this way, if you have to, you can just back up from the beginning.
You can back up the actual partitions and master boot record and all that good stuff before you start to actually touch the disk.
Next thing is, it's going to take some time.
Just depending on how bad the drive is, where the bad parts are, how often they are, how slow it is, transfer rate, how big the drive is.
160 gig drive took at right at about three days to make an image of.
And probably it's a total of four days to once restoration and putting it on the new drive.
So, easiest thing, I use a live CD called G-Parted.
It's similar to a commercial version of partition magic.
And also a USB stick with a few noted items on there about different things where boot sectors are.
Then just notes I have on the USB stick.
And also, I have the DD underscore R help, which is also in the show note.
And this is what the cast is mainly going to be about.
The helper application for DD rescue called DD underscore R help.
Like I said, I had no problems running it off of G-Parted live distribution, off of a thumb drive.
Once you've verified that you're riding to the right disk and trying to read from the right bad disk or known disk with errors on it.
You'll see the basic command line options for DD underscore R help.
Just pretty much the same thing as DD rescue, except there's a few other switches for it.
The first thing is it will automatically create a log file.
And this log file is sort of kind of hard to read at first.
But essentially, if you read the documentation and the FAQ and everything for DD underscore R help,
it'll tell you exactly what it's doing, what it's doing in the log files.
And it'll be a little bit clearer, good, you know, halfway into it.
You'll understand how it works.
Mainly, you're going to look through FDISC and see how big the partition is or disk is that you're copying.
How many blocks it has total?
From there, you'll be able to read where DD rescue is at any given time and how much it's actually copied.
If it goes from one end to the disk to the other, it will show the image will show a full size of the actual disk.
Being as that, it's skipping over parts that it kind of fails at.
It's going to display kind of a false information that's, oh, my disk is completely backed up.
No, what you want to do is read the log files and look for a line called EOS in caps.
This will tell you where DD underscore R help is at how much it's, how many blocks it's copied or at least tried to copy,
and all that good stuff.
And after a while, you'll see something inside of the log and also from the actual command line, you'll see bar.
And it'll detects representation of where it's at on the disk.
At first, basically, you'll see a bunch of dots.
And as it goes over, they'll turn it to x's.
So the dots are not parsed.
Jump points are going to be stars.
And parsed is means parts that it's already tried to go over or has gone over.
And how it works, if you go to the documentation, it starts off, hits an error, skips a little bit,
and then once it's gone through the whole disk, it goes back and picks the biggest chunk that it missed.
Starts in the middle and goes from the middle to the left to the middle to the right, meaning from the middle reverse and from the middle forward.
And I will leave some, in the show notes, I'll leave some example of the bar output.
So you'll kind of see what it looks like further down the road.
Also, there's some other things I didn't have to do, tweak any settings.
But you can do things like disable or enable DMA to make things go faster or slower.
The whole idea of DD underscore our help is to make the process faster.
And instead of sitting there, then, for on the same, you know, 500 blocks, it's going to skip over those portions and come back to it later.
So this utilizes the time that it takes to run a full DD rescue scan or DD scan or copy.
Pretty much wraps up this episode of Packer Public Radio.
Anybody has any questions about the episode or any other previous episodes?
You can contact me from the website, and I will try and help you. That's like it.
Thank you for listening to Packer Public Radio.
We are sponsored by caro.net. So head on over to caro.nc for all the hopes and needs.
Thanks for watching.