Episode: 2203
Title: HPR2203: NOT SO SMART
Source: https://hub.hackerpublicradio.org/ccdn.php?filename=/eps/hpr2203/hpr2203.mp3
Transcribed: 2025-10-18 15:42:53

---

This in HPR episode 2,203 entitled Not So Smart, it is hosted by Opera Nero R and in about
8 minutes long, and Karina Cleanflag.
The summary is how I am failing at troubleshooting this kind of issues.
This episode of HPR is brought to you by an honesthost.com, get 15% discount on all shared
hosting with the offer code HPR15, that's HPR15, better web hosting that's honest and
fair at An honesthost.com.
So I was using an old version, portable version of cool edit that I created, and at the
end doing noise reduction it decided to crash.
So lucky you guys get a better, more organized podcast for this series.
So what I want to be talking about today is my troubleshooting issues around disk IO.
What I think it is, is that this IO is heavy and it reaches a big file for whatever reason
it craps out locally and over the network, SMB, SSH doesn't matter what I do, but if
that IO goes up, I'm thinking on large files, it seems to want to crap out.
So first thing first, we want to look at it for look for errors, right?
So you want to do a check disk on the system.
There's a tune 2FS, the number 2FS, that's tune to FOXTROT Sierra-C1 and then the disk,
the partition, and that all on next boot scan the disk.
So I have all three of my disks since everything it's crashing all the time.
All three of my disks are being scanned when the computer boots up.
There's also a smart, smart will tell you how many errors you got, I've got no errors
across all three disks, they're all Western Digital Lex, I haven't had any issues with
them.
So I knew that wasn't a problem, I thought about BIOS and firmware on the BIOS updates
for the motherboard and firmware for the hard drives and I said try to get Windows to
boot through craziness and I still couldn't get it to boot.
So before that I actually validated that it wasn't any of that stuff because I booted
off the Linux live and I did some copying to backup stuff and then I also did some check
disk FSCK on and ran those on two different two drives in there for six hours and had no
problems.
So it's got something to do with something running or the kernel or who knows what.
Second thing I started to do is look at the error logs.
There's a Windows app called tracer.32.exe, tracer.32.exe and a log expert, tracer.32.exe I
think might be part of Windows server or something, I don't know where I found it.
It was on a work computer at some point in time and then you have the kind of GZ default
error highlighting or error or something like that.
There's another one called log expert which you can put like rejects in there for like
error or error or fail or enable or cannot or not fault anyways.
So that kind of helps you look through it and see anything that really jumped out.
I started seeing applications that I didn't need like Cups which is for printers and all
kinds of crap like Bluetooth and turning all that off really unplug any devices that
you don't absolutely need.
I uninstall the video drivers out of the kernel, I tried all kinds of stuff which I don't
think, I don't think it's any of that, I think it's something to do with the current
build that I have that's not happy with apps that are running, I'm running mono and all
kinds of crazy stuff.
So I started doing, walking around, I found a post about system preases and there's
apparently different kinds of I.O. schedulers within Linux, there's NOOP and anti-crap
perpetuary and deadline which that's what I'm using now and you can set it inside of
the grub or you can set it for each disk on boot and I did both of those just to make
sure double the double the that I changed my my scheduler.
I did notice faster transmission speeds on our USB, it would start out really fast and
then drop down to like slow speed and it would take five minutes to push and then I did
the deadline scheduler and pushed an image pretty like three times as fast so I definitely
fixed the speed of at least transferring to USB one of those cruiser piece of crap the
USB drives.
So we talk about that, I want to talk about our sync and why it's horrible you need to
download the latest version of our sync and compile it.
There's an option called dash dash info equals progress to which will basically give you
what you want it'll give you a percentage of the total time of how long it's going to
take and how much it's transferred thus far and along with that you can add the dash
see option if you're copying verb VMs and it will actually check the VMs to make sure
because if you don't do a CFC check on the VMs you're going to lose data because sometimes
the dates don't change or whatever it doesn't happen to flag that file to be different
and I've lost three months worth of VM data because of that before.
So I mentioned firmware stuff I don't think it's that, it's not different file systems
it's all EXT and NTFS so I don't think it's any of that stuff, it might be a file system
thing I don't know because I am backing up to the NTFS from EXT and EXT4 so what else
I talked about TTFS, the schedulers, temperatures ranges for the dryers or anywhere from 40 to 50
you can use the smart control app to help you out with that to show that value.
You'll see I have a thing called temperature Celsius change from 112 to 111 basically
closer it gets to 0 that the hotter it is so the number less the number the hotter it
is but really you just need to look at the actual temperature within smart and it'll
tell you, no one on my dryers had any errors or anything it's almost limit when it hits
a big file and it's anywhere from 15 to an hour it it craps out on me so whole thing
freezes nothing no CPU anything so that's where I'm at so far I have throttled it down
from it was there ain't is there anywhere from 70 to 124 locally I throttled that down
to 30 and so far so good hasn't crashed yet I'm tailing out all the logs that I could
possibly see while it's running so if it does dump something out to a log I can have
that you can also turn on crash dumps and I don't know if that's on by default but I might
have to turn that on and do some tracing with our sync or something you can do tracing
to binary I really don't want to do that because that's going to eat up I have just tracing
I think that pretty much covers everything where I'm at now so I might be good with this
I'll definitely update the show notes I might be good with this IO scheduler being set
to deadline but we'll see but I will let you guys know in the show notes if this fix
it or not and if you have a quick thing to record just record on your phone send it up
or send it to me and I can give it to some of the noise for you and feel free to contribute
you've been listening to HECKA Public Radio at HECKA Public Radio dot org we are a community
podcast network that release the shows every weekday Monday through Friday today's show
like OR shows was contributed by an HBR listener like yourself if you ever thought of recording
a podcast and click on our contributing to find out how easy it really is HECKA Public
Radio was founded by the digital dog pound and the infonomicon computer club and is part
of the binary revolution at binrev.com if you have comments on today's show please email
the host directly leave a comment on the website or record a follow up episode yourself
unless otherwise status today's show is released on the creative comments, attribution,
share a light, free dot org license.