How do you handle this...

[murph]

Mods, please move if this is too off topic, but...

If you have a client that asks you to "look at" their computer because it freezes up occasionally and aren't able to find anything obvious, what do you tell them or how do you handle this?

Basically I had someone give me their machine to work on but I was not able to reproduce any of the issues and found nothing significant otherwise. I gave it back to her and informed her of such. It ran fine for about a month, now she says it is "doing it again." I had the same results in that there was nothing I found that was out of the ordinary and couldn't reproduce the issue?

I'm not charging her for the second look but what do you all do?

Thanks.

 

[BollWeevil]

Working at a repair shop for a good many years, this was a VERY common problem. I would tell customers that it is one of the hardest problems to diagnose because it could depend on so many factors. I would let them know that I would need to keep their PC in the shop for at least a few days to stress test it, and even then the problem may not present itself.

I've seen lockup problems caused by anything from bad drivers, faulty hardware (ram/cpu/mobo), buggy software, etc.

Probably the most common causes are a weak power supply or heat-related issues. I got to where the first thing I would try if I couldn't find the problem was replace the power supply. I'd let the customer have it back for a couple of weeks to see if the problem ever re-occured.

I've had a few customers who would claim theirs would lock up hourly, but whenever it was in the shop it would behave fine. They would take it home and the same issues would happen. I was able to trace the problem to their keeping their tower in a closed desk drawer. Without proper ventilation, heat would build up and cause lockups.

So the short answer is that there is no short answer. Let the customer know that lockups take an extreme amount of time and patience to resolve, and that it's possible you might never find the problem. Also try to get them to record when lockups happen, i.e. what temperature the room is, what applications are running, what hardware (printer, usb camera) is plugged in, whether they are on the internet at the time, etc.

Those lockups that are easily reproducable are much easier to find the root cause, but unfortunately most lockups aren't easily reproducable.

 

[GeForceX]

Perhaps this is a learned lesson. To save yourself some potential trouble, why not examine the environment surrounding the computer and what he/she does/runs. It takes only 10 minutes or so. Then everything is in your hands. A great point there about the computer locking up tracing back to lack of ventilation-- that's why you should examine the environment.

-J.

 

[SJConsultant]

Perhaps this is a learned lesson. To save yourself some potential trouble, why not examine the environment surrounding the computer and what he/she does/runs. It takes only 10 minutes or so. Then everything is in your hands. A great point there about the computer locking up tracing back to lack of ventilation-- that's why you should examine the environment.

-J.

In addition, I have found numerous computer problems cease when you use a decent UPS such as an APC unit.

 

[Fark_Maniac]

In addition, I have found numerous computer problems cease when you use a decent UPS such as an APC unit.

on another issue related to power, you might want to crack open the case and make sure none of the capacitors are puffed up. We've recently had a rash of old 815 chipset (I know..I know...Abit will be repairing these) boards that will generally run fine, but on occasion...lock up.

 

[murph]

Ok, thank you all. Its hard to tell that to clients sometimes..."Its very hard to diagnose..." but it is really the truth. I know sometimes they don't believe you but I suppose that will happen occasionally.

Thanks again.

 

[murph]

In addition, I have found numerous computer problems cease when you use a decent UPS such as an APC unit.

As a sidenote also...she mentioned that it "won't turn back on" for several days. Any other ideas?

 

[SJConsultant]

As a sidenote also...she mentioned that it "won't turn back on" for several days. Any other ideas?

Is she even using a UPS? I had a client site with weird computer problems, only to find out the outlet the computer was plugged into was not grounded. The APC unit I plugged into the AC outlet immediately showed a "site wiring fault".

Once the wiring was corrected, all the issues disappeared.

Seriously we could spend all day doing "what ifs" and "possibilities" , but if she is experiencing the problem at her location, then you need to evaluate the computer at the location where it is having problems when it is having the problems.

 

[BobSutan]

Moved because its leaving the networking and OS realm and headed for misc hardware issues.

 

[_Durandal_]

As a sidenote also...she mentioned that it "won't turn back on" for several days. Any other ideas?

Yeah, certainly problems like this are extremely difficult to solve. All I can offer is that a friend of mine mentioned (a while back) that when reaching behind his tower to plug something in he bumped the power cord and the whole unit shut off. After this occurred, he could no longer power the unit back on unless he waited several minutes. I know you mentioned that your client can't do so for days, but perhaps this is somewhat similar??

When I looked at the power cord I noticed it was older and thinner than some of the new cables (namely the ones supplied with Antec power supplies). Anyone know what I'm talking about? If necessary I can post a picture of the two side by side.

Anyhow, all I did was replace his older PC power cord (which was actually fine) with a newer, thicker cord, and the problem went away. The fact is the connection end made much better contact with the PSU points. So, at this point I'm with the others on a possible power issue of some sort, whether it be from the actual outlet, cord, or capacitors. Good luck with it though, and if you find out I'd be interested in knowing the cause.