[davecom]

I'm getting constant random complete lockups on my G5 1.8 GHZ. I have 512 MB OEM, and 1 GB from Crucial from about 2 years ago. The lockups often lead to hard drive corruption. I've had to erase the drive completely (I zeroed all data) and restore my data to correct problems with it in the past. After a few hours of use though, it's back to locking up randomly. The only other thing I could think about it a virus in my core data files that keeps getting brought back after I copy back my backed up data, but I think that's unlikely. How can I find out if my RAM is causing the problem? Just taking it out? Also what is the correct order to have the pairs in? First the bigger ones, then the smaller ones right?

[mduell]

Remove half the RAM. Does it still lock up? Swap it for the other half. Does it still lock up?

Order doesn't matter on G5s; only on Intel machines.

[CanadaRAM]

Originally Posted by davecom 

I'm getting constant random complete lockups on my G5 1.8 GHZ...a virus in my core data files that keeps getting brought back after I copy back my backed up data?

Not likely to be any kind of virus or malware.
More likely to be RAM. You can do an Apple Hardware toest booting from the original CDs, you can download and run Rember (see Mac, Palm, and Windows Software Updates and Downloads - VersionTracker) or if you have TechTool Pro, run that.

Test your ram seperately as mduell says.

If that is not conclusive, what happens if you create another user, and log in with that user (this eliminates all of the data and preferences in your User folder from consideration)

[Sherman Homan]

I know for a fact that it is not a virus.
I also know that the G5's in general are really picky about ram, although the dual processors are worse. Do the mduell deal. That is probably the right direction.

[CatOne]

If your system is frequently locking up and that's leading to hard drive corruption -- there's a chance it could be the hard drive itself giving up the ghost. I had an issue with lockups (about once/week) where the machine would TOTALLY lock. I'd get a beachball in one app, and then over the span of about a minute everyone would hang, and then it would hard lock (iTunes was playing when I got the beachball, and after about 30-60 seconds it would stop playing).

Turned out it WAS a bad hard drive... AppleCare finally replaced it after a couple calls.

Try running this command from the command line:

sudo grep -i 'disk[0-9]s[0-9]' /var/log/system.log

Which will comb through your system log and pick out any I/O errors. If you see any in the log, your drive is failing (though unfortunately convincing AppleCare with system log I/O errors isn't typically what they will accept as evidence).

[davecom]

It apparently returned something. I just tried taking out the RAM I suspect is bad, then I ran the command you suggested. Does this mean I have serious errors?

dhcp-244-58:~ dave$ sudo grep -i 'disk[0-9]s[0-9]' /var/log/system.log

WARNING: Improper use of the sudo command could lead to data loss
or the deletion of important system files. Please double-check your
typing when using sudo. Type "man sudo" for more information.

To proceed, enter your password, or type Ctrl-C to abort.

Password:
Apr 11 15:01:02 localhost kernel[0]: BSD root: disk0s3, major 14, minor 2
Apr 11 15:01:02 localhost diskarbitrationd[38]: disk0s3 hfs 55E109B8-9BD8-3321-87A2-60B8C544B7EC HD /
Apr 12 21:31:03 localhost kernel[0]: BSD root: disk0s3, major 14, minor 2
Apr 12 21:31:04 localhost diskarbitrationd[38]: disk0s3 hfs 55E109B8-9BD8-3321-87A2-60B8C544B7EC HD /
Apr 17 11:57:52 localhost kernel[0]: BSD root: disk0s3, major 14, minor 2
Apr 17 11:57:53 localhost diskarbitrationd[38]: disk0s3 hfs 55E109B8-9BD8-3321-87A2-60B8C544B7EC HD /
Apr 18 22:25:08 localhost kernel[0]: BSD root: disk0s3, major 14, minor 2
Apr 18 22:25:09 localhost diskarbitrationd[38]: disk0s3 hfs 55E109B8-9BD8-3321-87A2-60B8C544B7EC HD /
Apr 29 01:28:17 localhost kernel[0]: BSD root: disk0s3, major 14, minor 2
Apr 29 01:28:18 localhost diskarbitrationd[38]: disk0s3 hfs 55E109B8-9BD8-3321-87A2-60B8C544B7EC HD /
Apr 29 14:33:58 localhost kernel[0]: BSD root: disk0s3, major 14, minor 2
Apr 29 14:33:59 localhost diskarbitrationd[38]: disk0s3 hfs 55E109B8-9BD8-3321-87A2-60B8C544B7EC HD /
Apr 29 14:34:51 localhost kernel[0]: BSD root: disk0s3, major 14, minor 2
Apr 29 14:34:52 localhost diskarbitrationd[38]: disk0s3 hfs 55E109B8-9BD8-3321-87A2-60B8C544B7EC HD /
Apr 29 14:39:23 localhost kernel[0]: BSD root: disk0s3, major 14, minor 2
Apr 29 14:39:24 localhost diskarbitrationd[38]: disk0s3 hfs 55E109B8-9BD8-3321-87A2-60B8C544B7EC HD /
Apr 29 14:46:20 localhost kernel[0]: BSD root: disk0s3, major 14, minor 2
Apr 29 14:46:21 localhost diskarbitrationd[38]: disk0s3 hfs 55E109B8-9BD8-3321-87A2-60B8C544B7EC HD /

[davecom]

I think that I've discovered the problem! The 1 GB of RAM that I've had for over a year is 2.6 volts, not 2.5 volts as suggested. Could this be what causes the seemingly random crashes? I tried memtest on it and it had no problems...

[mduell]

Originally Posted by davecom 

I think that I've discovered the problem! The 1 GB of RAM that I've had for over a year is 2.6 volts, not 2.5 volts as suggested. Could this be what causes the seemingly random crashes? I tried memtest on it and it had no problems...

2.5/2.6V is just the nominal rating... it's not a real difference and not going to cause crashes.

[davecom]

Well the 1gb is out and no random crashes yet, I'll keep my fingers crossed.