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

143
hpr_transcripts/hpr0057.txt Normal file
View File

@@ -0,0 +1,143 @@
Episode: 57
Title: HPR0057: LPI Certifications Part 3
Source: https://hub.hackerpublicradio.org/ccdn.php?filename=/eps/hpr0057/hpr0057.mp3
Transcribed: 2025-10-07 10:51:16
---
.
Hello and welcome to Acro Public Radio. My name is Ken Fallant, so they will be continuing
this series on the LPI certification. Today we are going to be talking about hard drives.
The most common hard drive found in the PC today is the Integrated Drive Electronics
or IDE hard drive. They are also known as ATA drives. Other drives in use commonly on
servers are scousy drives and recently most drives being shipped to PCs are now serial
ATA. So the original IDE or ATA drives are now retrofitted with the name of parallel
ATA drives. When you can tell the difference, the parallel has a big ribbon cable, whereas
the serial ATA has a smaller data cable. As far as the LPI certification goes, what you
probably need to know is the limits that are imposed due to the original bias limitations
when dealing with hard disks. You are also going to need to know some stuff on the tools
to look at and tune a hard disk and you probably also going to need to know about the Linux
disk naming conventions. So that is more or less what we are going to cover today. First
of all some background on what a hard disk is and how it works. Hard disk is actually composed
of platters made of other aluminium or glass coated in a magnetic material and the platters
or disks are arranged one on top of each other on a spindle. The whole thing is enclosed
in a enclosure and there is a hard disk controller attached to the underside of the disk. The
read right head is mounted in an arm which moves from the outside edge of the disk to the
center that allows it to access any area on the disk. The disks themselves can spin at
very high speeds often reaching 170 miles per hour or around 270 kilometers per hour. That
is stored on the surface of a platter in sectors and tracks. The tracks are concentric circles
and the sectors are pie-shaped wedges on the track. The term cylinder is used to describe
the same track on all the platters as they are stacked one on top of the other. So if
you have track one on the first platter, you are also talking about track one and second
track one on the third track one on the fourth as you go up. It creates a sort of cylinder.
When the original hard disks came out, the way you would locate a piece of information
on the hard disk was by giving it a reference based on cylinders, heads and sectors of tracks.
So if something was asked position one, cylinder one and head five on track twelve, you would
be able to locate information. Another method for locating data on the disk is called
the logical block address or the LBA address. And rather than using geometry like CHS does,
it simply used number structure to name the sectors regardless of their physical construction.
Okay now what does all of this got to do with the exam? The important part is that every hard disk
is presented to the bias through its geometry, which tells the bias of many cylinder's heads
tracks that the disk uses. This is using the CHS method. For Windows, the boot loader is always
located at the master boot record. However for Linux, the boot loader, LiDOR group can either be
at the master boot record or at the root partition. And if it's at the root partition,
then you have to be aware of certain limits that are imposed by this CHS methodology
employed by the bias. And the most important limit that you need to know about is the
one, oh two four cylinder limited. Now where this comes about is the IDE, ADA specification
had specification for the number of cylinders, heads and sectors. And then you had biases
the bias in 13H standard had also specifications for the number of cylinders, heads and sectors.
Problem is that they didn't talk to each other and we ended up with having to take the smallest
combination of each. So although the IDE specification supports 16 cylinders maximum,
the bias only supports 10. And while the bias supports 8 heads, the IDE specification only
supports 4. And the IDE specification supports 8 sectors max but the bias only supports 6. So what
you end up is having maximum of 10 cylinders which corresponds to one or two four cylinders,
sorry 10 bits for a cylinder number. And that corresponds to one or two four maximum cylinders.
And then the maximum 16 heads with 63 sectors. Like gives a total maximum of 504 megabytes. So if you're
making a boot partition, you should always make it less than 500 megabytes. A lot of people will
say that anything more than 50 megabytes is a waste of time. Okay, there have been various
different ways to get around this problem by doing translations and all the rest will put the
generally most modern hardest supports LBA. Now the LBA just starts at zero for the first sector
and marks its way up. However, there is a limit of 137 gigabytes imposed by the old ADA standard
that only use 28 bits for addressing. That's more or less gone away now and to the use of 32
bits sector numbers. And that increases the limits to two terabytes. Using workarounds and mapping
they were being able to make the CHS method work up to a limit of 8.4 gigabytes. But after that,
you're just supposed to use LBA. So when querying any hard drive over 8.4 gigabytes,
they will always report through geometry as 16, 3, 8, 3, cylinders, 16 heads, and 63 tracks,
regardless of what their actual geometry is. And you should look in the LBA field for their
actual number of sectors. And that's exactly what we're going to do right now.
But first before we hit that, I want to talk a little bit about the hard drive naming convention
under Linux. First of all, you'll be able to see the devices under special directory called
the DEV file system. Those who are all devices are kept. And like the PROC file system,
which we talked about before, it's a pseudo file system, meaning that it doesn't actually exist
in the conventional sense. Following again, the Linux, everything is a file philosophy.
Now in there, the naming convention has typically been for IDE hard disks that they begin with
slash dev slash HD something. And that for scousy disks, they will go in slash dev slash SD something.
Now scousy disks are also considered to be serial 80 disks. So if you want to see what devices
are attached to your computer, you want to check for both HD and SD drives.
Okay, so jump out to a console and type the following command, LS, space,
forward slash DEV, forward slash HD and anastrix. And for me, that returned no such file or directory
and what that does is it asks for a listing of all the files beginning with HD
in the dev directory. I'm going to do the same thing for SD files and I am returned with a list
SDA, SDA1, SDA2 and SDA5. And that means that I have one serial 80a disk attached to my computer.
And the reason I know it's one is that the third letter of that LS, forward slash dev, forward slash
SD asterix returned SDA and there was no SDB. So that means I just have one disk attached to my
computer and that's correct I do. For scousy disks, they're assigned letters. The first disk is
A, the second disk is B, the third disk is C and continues on to the SDA, SDB, SDC.
For ID artists, it works in a particular order. So the master on the first ID interface would be
HDA, the slave on the first ID channel will be HDB, the master on the second ID channel will be
HDC and the slave on the second ID channel will be HDD. Okay, that's the naming system for the
disks themselves. However, each disk can have partitions and on an ID drive you can have up to four
partitions and on a limited number of logical partitions. Those partitions are donated by a number
after the first three letters designating the disk. So on my example from my computer,
we had a SDA1 and SDA2 and then an SDA5. Another physically amounts is one big root partition which
is SDA1 and then there's a swap partition which is actually an extended partition and that's SDA5
and because extended partitions need to be housed somewhere, it's housed under SDA2. That's
probably a little bit complex for now but it'll come in time. Now looking at my file server downstairs,
when I do Alice, I suppose four slash dev, four slash HD star, I get a HDA, HDC,
HDC1, HDC2, HDC3, HDE and HDE1 and HDG and I happen to know that the HDA and the HDG
are CD-ROMs, the HDC and the HDE are, well the HDE1 is part of an LVM volume which we'll talk about
later and the HDC contains the root partition. I also have scusy disks on that and SDA1 is part
of an LVM SDF and SDF1 are part of an LVM as it's SDG1 and just as a side note, an LVM is
logical volume manager and it's a way to take multiple disks, physical disks and put them together as
a logical volume making the addition and expansion of your partitions as they're shown to Linux a
lot easier but that's not something for the exam just right at the moment. Okay a good command
to find out information about your hard disk is called the HDPram command and that's actually
spelled HDPARM. Get slash set hard disk parameters and for now we're only going to be getting hard disk,
I don't want anyone messing about with the hard disk parameters unless you know what you do it
and I give you the description from the man page which I got by typing man space HDPARM
HDPARM provides a command line interface to various hard disk IO controls supported by Linux
serial ATA parallel ATA and SAS subsystems and older IDE driver subsystems some options may not
work correctly with the latest kernel and the most important command out of that is going to be
minus capital I which requests identification info directly from the drive which is displayed in
the new expanded format which with considerably more detail than the older minus low case i flag.
Now if I'm osy over here to my server and I do uh
uh
sudo hdparam minus capital I for slash dev for slash hda I get loads of information back on my cd
ram drive the model number is a ncdvd read writer it's got the firmware it's standards used
at api dash 1 it's de recue it's bloody bloody but that's all very interesting now if I do um
an ls of the slash dev for sd drives and I look at um
and I look at one of the serial ATA drives I'll do s
s hdparam space minus capital I space for slash dev for slash sda I get a lot more information
about the drive it's a max store something rather it's got a serial number firmware
what standards it supports um and here we are the uh max and current cylinders heads and sectors
are down as 16 303 heads as 16 in the sectors as 63 which we uh found out earlier in this podcast
that it was in fact uh dummy information the real information that we're looking for is the
LBA usable addressable sectors which is 2 4 0 1 2 1 7 2 8 which is 122 gigabytes so you can
happily play with that command and see what information is available on your system one good
way to find out what in what our disks are attached to your computer is using they out from
the system log as it boots up and that command well I think we've talked about it before is called
and from the description in the man page d messages used to examine or control current
buffer the program helps user to print out their boot up messages instead of copying messages by
hand the user needs only uh do de message do messages uh however so if we go to the command line
on my server again and I type de message space uh the pipe command which is usually above the
keyboard uh the entry key and a type grep space sd I get a whole list of sd devices that were on
my computer and in there I'll see that I have uh discuss the uh cd rom drive that I had
coming up and the other additional disks and if I did the same thing de message and I pipe that
to grep and I do hd then I see all the different id devices that have come up so I see a max store
80a disk what else your cd rom drive comes up here as a hd device so that's pretty much it for this
episode of acro public radio I want to thank you all for listening if you yourself come across
something cooler interesting consider sitting down for five minutes doing a podcast about it
we'll send it in uh know that we could do with some extra filler episodes thank you very much
have a good day thank you for listening to acro public radio hpr sponsored by caro.net
so head on over to caro.nc for all of the team
you