Our Dumb Mistakes

Puppy related raves and general interest that doesn't fit anywhere else
Message
Author
User avatar
Amgine
Posts: 231
Joined: Thu 22 Sep 2011, 01:27
Location: Washington State

#16 Post by Amgine »

I picked up some RAM at a yard sale it was about 4GB, I wasn't sure what it was for but he wasn't asking much, I go home take out my old sticks and try to fit the new ones, they did not match, no big deal I put my orginal RAM sticks back in, close up the computer get everything set back up, press the power button and nothing, but a little smoke comming from my motherboard.

Turns out I dropped something in the RAM slot and did not think to check over everthing to see if it was clean. :oops:

Lesson learned I clean and check over everything before setting back up again. :idea:

:arrow: The same sale I got a 200GB harddrive it was full of Linux partitions, nothing that I could access, I used Puppy+G-parted to clean it.
Ubuntu-One Cloud Storage, Know where your Puppy files are, Back-up & Share files. https://one.ubuntu.com/referrals/referee/2118185/"]Sign Up!

User avatar
8-bit
Posts: 3406
Joined: Wed 04 Apr 2007, 03:37
Location: Oregon

#17 Post by 8-bit »

I had a PC motherboard given to me that would just shut down after a period of time. That was in the summer.
I hooked up it in the winter and the PC seemed to work fine.
When the warm weather came, the PC again would shut down after a short period of time.
So I went to investigate at that point.
I found that the Processor was rated at 3.3volt and the power jumper for it was set to 5 volt.
I changed the jumper to 3.3 volt and the shutdown problem disappeared.

I also was trying to clone one drive to another of the same manufacturer and specs and had both drives go to junk.
During the backup, I kept hearing a high pitched noise.
That noise turned out to be the bearings in one drive on their way to a major failure.
What got written to the other drive evidently took it out in possibly the attempted write of corrupted data and sectors.

User avatar
prehistoric
Posts: 1744
Joined: Tue 23 Oct 2007, 17:34

#18 Post by prehistoric »

Yesterday's example, don't assume murga-linux will always be up when troubleshooting a network problem. I was using it because it was convenient to access from my home page.

I started off when murga-linux was up, using Puppy on one machine. After trying changes I would go back to the one machine where "nothing had changed" and see if the network was working.

I found my mistake when I wondered if Puppy was doing something that other OSes did differently. I booted Linux Mint 14 on yet another machine, and had no trouble accessing the home page. Different home page, different site. No problem in the network I was troubleshooting.

---

Someday I'll write up my experiences with "the network problem from hell" and Windows 7. It goes by the name "unknown network". The main problem can be traced to timing issues. Even testing every component in another location is not enough -- I did that. The problem did not appear under XP, yet M$ blames everything except W7 as a possible cause.

Some people have "solved" this problem by limiting the speed of their gigabit ethernet ports. Others have installed lower speed network adapters. Still others have replaced routers, power-line adapters and switches. It helps to disable every damn program which tries to gain access to the Internet first. If none of these things works you might change your Cisco router to disable the spanning tree algorithm, or set the "spanning tree portfast" option. (Don't have an industrial Cisco router? Too bad.)

My dumb mistake here was to trust M$. Their preferred solution is to upgrade every component, (death to XP!) remove all non-M$ software and use their servers. Just getting a clear explanation from them of what is happening probably involves signing a non-disclosure agreement which makes you a wholly-owned subsidiary.

User avatar
prehistoric
Posts: 1744
Joined: Tue 23 Oct 2007, 17:34

#19 Post by prehistoric »

Flash wrote:I once spent about a week doing all kinds of stuff, up to and including reinstalling the OS and the drivers, trying to get the speakers to work, only to discover I had plugged them into the microphone jack. :x
This is an example of a perennial dumb mistake to check for on "service calls" about sound. I just committed another version of the same yesterday, while checking out a newly built machine.

As detailed elsewhere, this is a close to state-of-the-art machine ( Asus F2A85-M pro mainboard and an AMD A10 5800K quad-core processor ) built for someone who persists in running Windoze, not necessarily a Puppy machine.

After some thrills in bringing it up to the point where I could see the BIOS screen, I immediately turned to Puppy for testing. All good, except no sound. Try Fatdog 611, in case there is some difference requiring 64-bit software. No sound. Try Linux Mint 14. Same.

In between these tests, I went over the wiring, in case I had plugged it into the wrong audio jack. (There are 6, plus SPDIF.) Also checked for muting at multiple places in software, or via the mute switch on the amplified speakers themselves. Also saw that speaker light was on, indicating power to speakers.

Now, I'm going to shock people by saying something good about Windoze. While I was installing W7 Home Premium, (to see if there was something on the ASUS installation disk or site which would fix the problem,) I had plenty of time to think. (Experienced installers will know just what I mean.) Sitting there, and looking around, I noticed that the physical volume knob on the speakers had no marking to indicate position. Turned volume up. Surprise! Sound has been working all along. :oops:

I'll keep a dumb mistake, made while connecting the mainboard, to myself, only noting that I did not let the magic smoke which makes computers work out of any chips.

Hint: just because a connector will fit a plug does not mean it belongs there.

This hint is right up there on my list next to the reminder that most cables have two ends.

User avatar
NeroVance
Posts: 201
Joined: Wed 10 Oct 2012, 23:00
Location: Halifax, Canada

#20 Post by NeroVance »

I'd say I've made my mistakes. Actually I made a pretty major one probably near this date a few years back. I won't go into much detail, but I will admit the time between then and me joining the kennels was a time of wisdom for me, and I learned much, and gained much maturity.

But hardware, I did buy a 2GB stick of RAM, and realized my desktop already had all it's slots filled with 2GB sticks. But I may as well keep em, I may need them if one of my sticks begins to fail.

User avatar
Flash
Official Dog Handler
Posts: 13071
Joined: Wed 04 May 2005, 16:04
Location: Arizona USA

#21 Post by Flash »

In my experience, the chance of a RAM failure after the initial burn-in period is approximately zero. If it lasts a week or two, it will almost certainly outlast the motherboard -- and then it will be obsolete and useless.

starhawk
Posts: 4906
Joined: Mon 22 Nov 2010, 06:04
Location: Everybody knows this is nowhere...

#22 Post by starhawk »

Flash, I hate to tell you this... but that's not quite right.

My mother has an HP Compaq tc4200 "convertible" laptop/tablet system. Runs XP because "you get what you pay for", etc. :roll: After a while it started bluescreening. I changed one of the RAM sticks out, and the blue screens have yet to come back.

Mind you I had purchased both of the sticks in that machine, about a year before.

User avatar
Amgine
Posts: 231
Joined: Thu 22 Sep 2011, 01:27
Location: Washington State

#23 Post by Amgine »

Flash wrote:In my experience, the chance of a RAM failure after the initial burn-in period is approximately zero. If it lasts a week or two, it will almost certainly outlast the motherboard -- and then it will be obsolete and useless.
Not true, my Dad had to send his back in after they failed, It was used RAM, from a gaming computer.
Ubuntu-One Cloud Storage, Know where your Puppy files are, Back-up & Share files. https://one.ubuntu.com/referrals/referee/2118185/"]Sign Up!

maddox
Posts: 454
Joined: Fri 28 Sep 2007, 20:37
Location: sometimes in France

#24 Post by maddox »

I think Flash was talking about an initial burn-in period of the RAM :run a looping ram test prog for a week or 2.
makes sense -> usually used in servers.

My mistake , been in france visiting, brought some 220V power supplies with me with continent plug adapters (aus/gb/us):
laptop,usb charger,oscillo,small TX/RX ...
wondered why some equipment smoked after a while... :(
I found out the painfull way $$$, france uses a 230-240V power grid

User avatar
Amgine
Posts: 231
Joined: Thu 22 Sep 2011, 01:27
Location: Washington State

#25 Post by Amgine »

maddox wrote:I think Flash was talking about an initial burn-in period of the RAM :run a looping ram test prog for a week or 2.
makes sense -> usually used in servers.
Ok!
Ubuntu-One Cloud Storage, Know where your Puppy files are, Back-up & Share files. https://one.ubuntu.com/referrals/referee/2118185/"]Sign Up!

postfs1

Re: Our Dumb Mistakes

#26 Post by postfs1 »

To reedit up to date.
Last edited by postfs1 on Sun 27 Mar 2016, 21:23, edited 1 time in total.

User avatar
greengeek
Posts: 5789
Joined: Tue 20 Jul 2010, 09:34
Location: Republic of Novo Zelande

#27 Post by greengeek »

This is not Puppy related - but years ago I was installing a computerised voicemail interface unit into a Telecom exchange. My field of experience was with the computer side of things - not the details of how the phone network operated. All I knew was that the phone network operated off 48 volts. "Thats not enough to kill you but take care anyway" was what they told me.

Anyway, these interface units ("Summa Four") had substantial red and black wires coming out of them, and these had to be connected to the massive buss bars running through the exchange. Naturally I connected the black wire to the Ground buss, and the red wire to the 48v buss. Everything was great until I turned on the new unit.

That was when I learned how much magic smoke they compress inside electronic components!

Turns out that telephone exchanges run on NEGATIVE 48 volts so the red wire goes on the ground buss. Doh! Who's stupid idea was that??!**!##%?

User avatar
Flash
Official Dog Handler
Posts: 13071
Joined: Wed 04 May 2005, 16:04
Location: Arizona USA

#28 Post by Flash »

When designing with vacuum tubes, it greatly simplifies the circuit designer's job to assume a negative ground, so negative ground became the natural choice as electronics grew into something useful. Before that, it didn't matter much which polarity was "ground". The phone system predates the development of vacuum tubes by decades. They probably flipped a coin to decide which side of the battery to ground, but once positive ground became the standard throughout the phone system, it would have been nearly impossible to change.

I remember when cars had 6 volt batteries and at least some of them grounded the positive side. I don't know if they all did. The original VW beetle put its 6 volt battery under the rear seat, and a heavy passenger could short it out.

User avatar
prehistoric
Posts: 1744
Joined: Tue 23 Oct 2007, 17:34

#29 Post by prehistoric »

greengeek wrote:This is not Puppy related - but years ago I was installing a computerised voicemail interface unit into a Telecom exchange. My field of experience was with the computer side of things - not the details of how the phone network operated. All I knew was that the phone network operated off 48 volts. "Thats not enough to kill you but take care anyway" was what they told me...
You wouldn't have lasted long in the environment where one of my OLD friends learned about digital electronics, logic and computers. He started off working INSIDE a computer -- Whirlwind I.

The microcoding was done with diodes clipped between bus bars about as big around as your little finger. For electronic convenience some of those bus bars were floating at 400 volts.

Oh, BTW, the machine blew a tube almost every time it was turned on, so debugging was done with the power on.

I don't dare tell him about how bad I have it today.

Dewbie

#30 Post by Dewbie »

First I thought several floppy drives were bad...
Then, no, it's the floppy disks themselves...
(Neither assumption was correct.)

There's often enough variation in head alignment between drives that a floppy disk written with one won't work with another. So you have to write the disk with the drive that it will be used with.

User avatar
greengeek
Posts: 5789
Joined: Tue 20 Jul 2010, 09:34
Location: Republic of Novo Zelande

#31 Post by greengeek »

prehistoric wrote:You wouldn't have lasted long in the environment where ....... some of those bus bars were floating at 400 volts.
True enough. I wonder what the frying time of human flesh at 400v is. 60 seconds maybe?? Mmmmm, smells like chicken...

That reminds me of another mistake I made - I was given the job of replacing the faulty "ON/OFF" switch in an 80 column card punch and I happily yanked the spade wires off the back of the switch - having forgotten to unplug the cord from the wall. Zzzzzt.

I managed to throw the switch right across the room and swore like a trooper, much to the customers amusement. Lesson learned.

User avatar
puppyluvr
Posts: 3470
Joined: Sun 06 Jan 2008, 23:14
Location: Chickasha Oklahoma
Contact:

#32 Post by puppyluvr »

:D Hello,
LOL...
I recently had a box go down after moving it.. Video failed completely.. Nada..
Had a geforce card in it, which I assumed had failed.. Removed it and attached to the onboard intel, reset the jumpers, and rebooted.. Still nothing at all.. Not even the bios framebuffer.. Crap.. Removed the disks, the cd/dvd, and replaced the ram, to no avail.. Checked all the jumpers, the ram slots, even tested the MB clip on the power supply.. Only after giving up in disgust after a few hours, when I removed the tower from the desk, did I realize I had plugged in the wrong monitor cord.. :roll:
I once turned off the wrong breaker to a 440 volt 60hp power source, and then stuck a screwdriver into a box, and blew the end off it, and my finger..
The video card thing still pissed me off more.... :wink:
Close the Windows, and open your eyes, to a whole new world
I am Lead Dog of the
Puppy Linux Users Group on Facebook
Join us!

Puppy since 2.15CE...

User avatar
8-bit
Posts: 3406
Joined: Wed 04 Apr 2007, 03:37
Location: Oregon

#33 Post by 8-bit »

At one time, I worked as a millwright at a small lumber mill that made agricultural stakes.
I was also a regular worker there since it did not require a full time millwright.
I had another worker complain to me that he was being shocked by a machine he was feeding.
I touched the machine and did not get shocked so I thought I would investigate further.
When I did not get shocked, I was standing on a wooden floor.
I then went to clean out some wood on a trough conveyor and happened to touch the machine in question.
I now know what it feels like to be the ground for a 440 volt 3phase circuit.
It turned out that when the mill was originally wired, the conduit was used as a ground.
The conduit had separated and wore through one of the power wires.
And with the conduit separated the machine had what they would call a floating ground.
So anyone that was grounded and happened to touch the machine completed the circuit.
Not a good thing!

I had the machine then shut down, locked out at the breaker, and established a good ground for it as well as repairing the bad wiring.
But I did learn that when one goes to test that a meter is a lot safer than touching the machine to check for a short!

I guess you could say I got recharged that day!

User avatar
lwill
Posts: 171
Joined: Fri 13 Jun 2008, 04:00
Location: City Of Lights
Contact:

#34 Post by lwill »

Not Puppy, but linux.
I updated to Fedora 18 with KDE on one machine and since then the "device notifier" applet would always ask for root password to mount cd/dvd/usb/sd drives when inserted. Never had before. No biggy, but annoying.
Spent half the day searching for reasons. Permissions, settings, mounting settings. Lots of google.
Turns out in the "default applications", the default file manager was set to "Konqueror ROOT" for some reason. I have an icon on the tool bar set to open my home folder that uses Dolphin (which is the normal fedora default) so I never try to open a directory directly and never actually used the "default" file manager.
.
.
On further thought, I may have done it to myself at some point. I was working with Android / linux on a Mele 2000 using sd cards to transfer stuff and needed to be root to modify files on them.

Still half a day I lost fixing it!!

User avatar
prehistoric
Posts: 1744
Joined: Tue 23 Oct 2007, 17:34

#35 Post by prehistoric »

lwill wrote:...Still half a day I lost fixing it!!
Only half a day? Hardly worth mentioning.

How about the better part of a stressful week?

This story relates to the earlier business about high voltage in a strange way, we were simulating a device which used a traveling-wave tube -- because not only did these things cost about $15,000 apiece, there was a real problem of losing students if they reached into the case without turning the power off. (We're talking thousands of volts, with kilowatts of power.) It also has some relevance to assumptions about the polarity of ground.

Naturally, we were dependent on an expert instructor for testing our simulated device. Nobody wanted to fool with the real thing, even if we were allowed to. (The device was classified, which made it hard to access data about it.) The one expert we had that week told me I had one control "backwards" on the simulator, as opposed to the real device. I questioned him about this with a graph of the performance of some parameter versus voltage on that knob in front of us. When I thought I understood, I went off and reprogrammed. This was not a simple task, because of other factors that would take us too far afield. Any substantial change cost me hours of work. I went through this drill three times. Each time it remained "backwards", and I became more confused.

On the last day of this fiasco I walked into the room housing our simulator, and saw a technician resoldering the wires on the potentiometer controlling the (simulated) voltage which was the problem. It turned out what our expert meant by "backwards" had nothing to do with the voltage on the graph; he meant the real knob rotated counterclockwise to increase voltage. My original program had been correct, but under pressure of all the changes made in a hurry I was no longer sure which version I should use.

I ended up reprogramming that control from scratch one last time. It only took me a little over an hour, because of all the practice I had had. A simple question about clockwise versus counterclockwise could have saved most of a week. Fortunately, errors with simulated voltages don't kill people.

Post Reply