OK - Possible bug with Slack Puppy and thumb-drive
Yesterday I worked on text files (and more) on a thumb-drive.
I shut the system down for the evening.
Next morning (Wednesday) I booted my PC and "opened" Geany and (after that) clicked on the file of interest (Essay_5.txt in "open recent"). (This is not ctrl+"O" - but below that in the menu.) Nothing. (Thinks ...) Drive is not mounted. OK! We all do THAT sooner or later. I have a 4-slot USB extension; inserted the drive and clicked on the drive icon. (This mounts the drive.)
Bingo! I see something like the ROX file manager (probably is the ROX file manager) but Essay_5.txt is not displayed. Bugger! I have lost everything. Grrr!
Thinking, thinking, thinking .....
So - I did nothing. (I'm not stoopid!!)
I closed everything then re-booted my (very old) laptop.
Joy of joys - and great alarm at the same time.
Every file I worked on yesterday is NOW THERE.
("There" means visible in the ROX file manager and available for use.)
And yes! I edited Essay_5 all day, saved/copied/closed/opened - working as expected.
Now Puppy (and Linux) are reliable systems.
I am surprised by my report; I'm confident in the details above.
Has anyone seen anything like this?
All the best from Australia
Les Smith.
Some info.
My system:
Laptop Manufacturer: Dell Computer Corporation
Product Name: Inspiron 8600
Motherboard Vendor: Dell Computer Corporation
Product Name: 0D2465
BIOS Vendor: Dell Computer Corporation
Version: A14
Release Date: 06/30/2005
Distro: Slacko Puppy 6.3.2
Window Manager: JWM v2.3.2
Desktop Start: xwin jwm
Development:
Bash: 4.2.53
Geany: 1.25
Gtkdialog: 0.8.4
Perl: 5.18.1
Memory 2GB. External kbd, Mouse, 4-port USB drive.
USB drive: 64MB (yes, very small). Generic brand. (Perhaps the USB drive is flakey?)
(the end)
Unreliable USB operation with Slacko 6.3.2
USB stick with Slacko
Never, ever, had any issues such as you describe.
Is Puppy installed on the HD and if so is it the recommended frugal installation?
Suggest you mount all the partitions that you can and use the Search Programme to find the missing files. If you have been saving them, then they should be somewhere.
Is Puppy installed on the HD and if so is it the recommended frugal installation?
Suggest you mount all the partitions that you can and use the Search Programme to find the missing files. If you have been saving them, then they should be somewhere.
I once had a problem where ROX didn't seem to be updating correctly, but it turned out that the flash drive had gone bad.
http://www.murga-linux.com/puppy/viewto ... 32#1016203
http://www.murga-linux.com/puppy/viewto ... 32#1016203
Unreliable USB operation with Slacko 6.3.2
To all: First, and most important - I did not lose a file. After rebooting the laptop the system ran for a full day to specification - i.e. Linux ran a variety of programs and accessed a small number of files flawlessly.
I had a problem with a completely different USB "drive" but did not report that experience here because I was uncertain about the sequence of operations prior to "the problem". And what was this problem? The disk was completely corrupt. Nada. I ran GParted / Device / Attempt Data Rescue. After that - I found the disk could be read. Every folder and all the essential "Puppy" files were gone and the folder called "Lost and Found" has many, many files in it. I do not wish to discuss this experience. I cannot document the circumstances leading to the DISASTER. HA! Perhaps I have a USB port with a shonky connector? Who knows? Guessing without a full set of facts is a waste of my time and yours.
o-o-o
To 6502 coder: Thank you for the link to your post on the topic of incorrectly displayed. Back in the dark ages, when the S-100 bus was "beez-kneez" and 8 bit systems were "big", I wrote a walking bit algorithm to test memory. Ditto for a FDD exerciser. The FDD exerciser interleaved writing and reading to random data to random sectors on the FDD, reporting any errors experience.
Question: Is there a program that can test a thumb-drive for reliable operation (at such a low level)?
Question #2. Is there a similar program to test RAM? Clearly such a test must operate "outside" the OS (or before the OS boots). Yes, my DELL laptop checks memory (either with a short test or a longer test) when I press the "boot" button, but I have seen memory failures where the fault was apparent only after hours of testing. In one instance I found memory where a single bit was "flaky". (This was on an 8 bit uP that permitted easy access to memory after the system had been booted. You can't easily do that with a modern PC.
o-o-o
To Peter W. Thank you Peter. Yes, I found the missing files. They were where they were supposed to be. I only had to re-boot da Puppy and they appeared. However, I suspect (strongly suspect) that I would have had a scrambled thumb drive had I used it instead of re-booting the OS.
Configuration: Frugal or Full? I boot (at first) from CD drive. After that Puppy looks for virtual machine vmlinuz on the hard drive. I think this is a frugal installation? It is the "boot-faster" option that is available when Puppy is installed on a PC the first time. I see (about) 7 files in the "top level" of the hard drive.
(-end-)
I had a problem with a completely different USB "drive" but did not report that experience here because I was uncertain about the sequence of operations prior to "the problem". And what was this problem? The disk was completely corrupt. Nada. I ran GParted / Device / Attempt Data Rescue. After that - I found the disk could be read. Every folder and all the essential "Puppy" files were gone and the folder called "Lost and Found" has many, many files in it. I do not wish to discuss this experience. I cannot document the circumstances leading to the DISASTER. HA! Perhaps I have a USB port with a shonky connector? Who knows? Guessing without a full set of facts is a waste of my time and yours.
o-o-o
To 6502 coder: Thank you for the link to your post on the topic of incorrectly displayed. Back in the dark ages, when the S-100 bus was "beez-kneez" and 8 bit systems were "big", I wrote a walking bit algorithm to test memory. Ditto for a FDD exerciser. The FDD exerciser interleaved writing and reading to random data to random sectors on the FDD, reporting any errors experience.
Question: Is there a program that can test a thumb-drive for reliable operation (at such a low level)?
Question #2. Is there a similar program to test RAM? Clearly such a test must operate "outside" the OS (or before the OS boots). Yes, my DELL laptop checks memory (either with a short test or a longer test) when I press the "boot" button, but I have seen memory failures where the fault was apparent only after hours of testing. In one instance I found memory where a single bit was "flaky". (This was on an 8 bit uP that permitted easy access to memory after the system had been booted. You can't easily do that with a modern PC.
o-o-o
To Peter W. Thank you Peter. Yes, I found the missing files. They were where they were supposed to be. I only had to re-boot da Puppy and they appeared. However, I suspect (strongly suspect) that I would have had a scrambled thumb drive had I used it instead of re-booting the OS.
Configuration: Frugal or Full? I boot (at first) from CD drive. After that Puppy looks for virtual machine vmlinuz on the hard drive. I think this is a frugal installation? It is the "boot-faster" option that is available when Puppy is installed on a PC the first time. I see (about) 7 files in the "top level" of the hard drive.
(-end-)
To test /dev/sdc
If nothing was written to the device, the md5sum should be the same each time.
To do 10 md5sums
fdisk -l should tell you what the device name is.
You can mount the drive and write to it before testing, for example
ram tester:
http://www.memtest.org/
Code: Select all
dd if=/dev/sdc | md5sumTo do 10 md5sums
Code: Select all
for j in {1..10}
do
dd if=/dev/sdc
doneYou can mount the drive and write to it before testing, for example
Code: Select all
dd if=/dev/urandom > /mnt/sdc1/dd.binhttp://www.memtest.org/
I have a 4-slot USB extension; inserted the drive and clicked on the drive icon. (This mounts the drive.)
Nothing showing in Rox for the USB drive.
then re-booted my (very old) laptop.
Every file I worked on yesterday is NOW THERE.
I assume the USB drive was plugged into the 4-slot USB extension, when you booted.
This probably has to do with how this 4-slot USB extension is connecting drives to the computer.
Just plugging one in is not getting a complete connection.
But if the USB drive is plugged in when you boot. It does fully get connected.
May want to try plugging in this 4-slot USB extension into a different USB slot of the computer.
Not all computer USB ports provide the same power to a plug in added ports USB adapter.
Especially on a very old computer.
The things they do not tell you, are usually the clue to solving the problem.
When I was a kid I wanted to be older.... This is not what I expected
YaPI(any iso installer)
When I was a kid I wanted to be older.... This is not what I expected
YaPI(any iso installer)
-
Mike Walsh
- Posts: 6351
- Joined: Sat 28 Jun 2014, 12:42
- Location: King's Lynn, UK.
@ LNSmith:-
Les:- I don't know if this is the sort of thing you're looking for. Neerajkolte found this a few years ago, and shortly after Bill (rcrsn51) built a wee GUI for it. It's fairly self-explanatory.....but you can read up about it in the thread:-
[ =========F3=========
http://murga-linux.com/puppy/viewtopic. ... 340#788340 ]
Home page here:-
http://oss.digirati.com.br/f3/
I believe it began as a project for simply checking for 'fake' USB drives, but a number of tools have since been added. Can't hurt to take a look.
Mike.
Les:- I don't know if this is the sort of thing you're looking for. Neerajkolte found this a few years ago, and shortly after Bill (rcrsn51) built a wee GUI for it. It's fairly self-explanatory.....but you can read up about it in the thread:-
[ =========F3=========
http://murga-linux.com/puppy/viewtopic. ... 340#788340 ]
Home page here:-
http://oss.digirati.com.br/f3/
I believe it began as a project for simply checking for 'fake' USB drives, but a number of tools have since been added. Can't hurt to take a look.
Mike.
Last edited by Mike Walsh on Thu 18 Jul 2019, 21:37, edited 3 times in total.
I used two flash drive testers during my investigation of my problem. Both are Windows apps. I chose to use Windows apps for the testing so that I could eliminate Puppy itself as the source of the problem. In addition, I (eventually) ran these apps on my Win7 desktop machine, rather than on the WinXP/Puppy dual-boot laptop+USB hub where I was having the problems. This allowed me to eliminate the laptop itself and the USB hub as the source of the problem.
The first app I tried is h2testw, which is widely recommended. Unfortunately, it kept insisting that my flash drive was OK. So I no longer trust it, although it's handy for checking your flash drive's read/write speeds.
The app that finally detected that my flash drive was going bad is CheckFlash, which I got here:
https://downloads.tomsguide.com/Flash-C ... 22182.html
NOTE: although others have correctly pointed out that the USB hub or the USB port on the computer itself should also be considered suspects, I want to point out that due to the intermittent nature of the errors I was encountering, and the false negatives I got from h2testw, at various times during my investigation I came to suspect first my USB hub, then the USB port on my laptop, as the culprit. As I mentioned above, moving the testing to my Win7 desktop machine and running CheckFlash finally showed conclusively that it was the flash drive itself that was bad.
The first app I tried is h2testw, which is widely recommended. Unfortunately, it kept insisting that my flash drive was OK. So I no longer trust it, although it's handy for checking your flash drive's read/write speeds.
The app that finally detected that my flash drive was going bad is CheckFlash, which I got here:
https://downloads.tomsguide.com/Flash-C ... 22182.html
NOTE: although others have correctly pointed out that the USB hub or the USB port on the computer itself should also be considered suspects, I want to point out that due to the intermittent nature of the errors I was encountering, and the false negatives I got from h2testw, at various times during my investigation I came to suspect first my USB hub, then the USB port on my laptop, as the culprit. As I mentioned above, moving the testing to my Win7 desktop machine and running CheckFlash finally showed conclusively that it was the flash drive itself that was bad.
Unreliable USB operation with Slacko 6.3.2
Thanks to all who replied to this posting.
I have more info to post and will do this soon.
Briefly, the point of interest is what happens when I mount a thumb drive.
Here is my procedure:
I plug the thumb drive into the USB extension module. It's there but unmounted. Next, I mount the drive.
I can mount the drive in two different ways.
(1) By clicking on the icon for sdc(n) on the desk-top (OR)
(2) Opening Pmount from desktop > file (menu) and selecting Pmount.
When I mount the thumb drive by (1) I see an 8 character name for files on the drive. eg I might see Essay_~1.txt.
When I mount the thumb drive using Pmount (2) I see the full file name displayed. so I see Essay_#4_(The_Shop).txt Yes, many of the files on this disk have a similar name, eg Essay_#5_(Covert_Operations).txt
Yes, these are many files with similar names in my working directory on this thumb-drive. I think this is partly responsible for the corruption.
On this drive: I use a FAT16 filesystem with 64MegaBytes capacity. Yes, Megs, not Gigs.
I cannot say if I can reproduce this problem using a FAT32 file-system.
Provided I mount the drive using Pmount I can work all day on this drive and everything happens as expected. (No faults).
I am confident that I will corrupt the file-system on my drive if I access the files using the 8bit name.
I have only corrupted the files and the file-system once and I will not run a test to do this a second time.
I will try to repeat this exercise using a FAT32 file-system.
I have a more complete report on this process but that is on a different (not-available) drive.
Again, thanks to all who contributed and I hope this report adds to the general knowledge "out there".
To those who contributed info about software to test USB drives I say "Thank you".
Leslie Smith.
My system: A 15 year old laptop - Dell Inspiron 8600 1.4GHz clock
Memory: 2GB RAM (greatest amount possible)
Drives: sda1 (about 40GB). sdb1 (USB thumb drive (64MB see above) sr0 - garden variety CD/DVD drive (as supplied by Dell).
Slacko 6.3.0.
Yes, this is a primitive system, but I spend a lot of time working with vanilla type text files. Mostly this system matches my need for CPU power AND it consumes a far fewer kW hours than a standard tower. Typically 80W. A low-powered tower will consume 250W (maybe more with a monitor).
That means I save approx 1kWh every day. Now factor that across 10,000 (or 100,000 users) and figure the annual saving in CO2 reduction.
I have more info to post and will do this soon.
Briefly, the point of interest is what happens when I mount a thumb drive.
Here is my procedure:
I plug the thumb drive into the USB extension module. It's there but unmounted. Next, I mount the drive.
I can mount the drive in two different ways.
(1) By clicking on the icon for sdc(n) on the desk-top (OR)
(2) Opening Pmount from desktop > file (menu) and selecting Pmount.
When I mount the thumb drive by (1) I see an 8 character name for files on the drive. eg I might see Essay_~1.txt.
When I mount the thumb drive using Pmount (2) I see the full file name displayed. so I see Essay_#4_(The_Shop).txt Yes, many of the files on this disk have a similar name, eg Essay_#5_(Covert_Operations).txt
Yes, these are many files with similar names in my working directory on this thumb-drive. I think this is partly responsible for the corruption.
On this drive: I use a FAT16 filesystem with 64MegaBytes capacity. Yes, Megs, not Gigs.
I cannot say if I can reproduce this problem using a FAT32 file-system.
Provided I mount the drive using Pmount I can work all day on this drive and everything happens as expected. (No faults).
I am confident that I will corrupt the file-system on my drive if I access the files using the 8bit name.
I have only corrupted the files and the file-system once and I will not run a test to do this a second time.
I will try to repeat this exercise using a FAT32 file-system.
I have a more complete report on this process but that is on a different (not-available) drive.
Again, thanks to all who contributed and I hope this report adds to the general knowledge "out there".
To those who contributed info about software to test USB drives I say "Thank you".
Leslie Smith.
My system: A 15 year old laptop - Dell Inspiron 8600 1.4GHz clock
Memory: 2GB RAM (greatest amount possible)
Drives: sda1 (about 40GB). sdb1 (USB thumb drive (64MB see above) sr0 - garden variety CD/DVD drive (as supplied by Dell).
Slacko 6.3.0.
Yes, this is a primitive system, but I spend a lot of time working with vanilla type text files. Mostly this system matches my need for CPU power AND it consumes a far fewer kW hours than a standard tower. Typically 80W. A low-powered tower will consume 250W (maybe more with a monitor).
That means I save approx 1kWh every day. Now factor that across 10,000 (or 100,000 users) and figure the annual saving in CO2 reduction.