Help resurrecting my Sage server..

[osx-addict]

As some of you may recall.. I was looking for a new Sage server earlier this year (you can read up on that thread here).

So, Last night my wife calls to me asking why Sage is down.. I try to VNC over to it -- no luck.. my laptop indicates that the host is down -- ping fails.. Hmm.. Not good!

I drag a monitor/keyboard/mouse over to the Sage server plug it all in and find it stuck in a boot-loop.. Apparently XP hard crashes about 3 seconds after putting up the XP logo and the system reboots -- gets to the safe-boot menu, times out after 30 seconds, re-loops again.. Doh!

I drag out my XP install media which loads OK to a point, after which the installation kernel crashes indicating that I should check for viruses or something.. Can't see how that would have anything to do with it since I'm loading from a read-only DVD. That's about where I left it .. looking for suggestions/advice..

I'm thinking my first plan of attack is to remove the SSD that works as the boot media and then load the XP installation media.. If it loads OK then perhaps my motherboard is good.. However, I realize that sometimes these motherboards can get problems.. my hunch is that may be the case but obviously I've got no proof. All of my BIOS settings are set for "normal" -- not overclocked or anything..

Any suggestions/ideas/directions that you can point me in would be greatly appreciated!

[jorton]

I had something similar recently with a bad stick of ram on a Win7 64 HP install. I removed the two older sticks and the boot loop ended and everything back to normal.

I also had an SSD in this machine and thought the same but in the end just the ram that was the issue.

Good luck.

[osx-addict]

Hmm.. I had pondered the 'bad-ram' thought briefly also -- I guess I could fire up the old 'memtest' cd-rom and see what it finds -- did you try that in your case? I don't recall how many sticks of ram I put in this thing off the top of my head.. I know I didn't skimp with no-name ram -- I've got Corsair Vengeance RAM.. I'm just hoping I can find my receipt IF I need it for some sort of parts warranty..

[panteragstk]

The ram is a good place to start. Sounds like some sort of hardware failure. Those are always fun to figure out. Hope it isn't the motherboard.

[Dargason]

A failing power supply might cause weird things too.

[osx-addict]

Thanks guys.. It's pathetic to think that a PC starts to fall over at 6-7 months of use -- albeit 24x7..

I'll keep you posted on my diagnostics..

[rgroves]

Are you getting a Stop error code? Like other said... probably bad or failing RAM.

[osx-addict]

No.. No error codes -- windows XP starts booting and after perhaps 3 seconds the screen goes blank and a few seconds later the BIOS is up again.. Then it just loops over and over and over..

I've currently got memtest86 running..without error so far -- but its only been running about 1 hr. I might need to let it run for a day or so..or at least until tomorrow morning..

[RedR]

Based on the info you have, I''d guess memory, hard drive, or power supply. Since the XP install CD craps out, I'd venture to say the 3 items are still in question. I'd start testing easiest to do first..
Check BIOS settings, make sure they correctly support the hardware you have installed.
Memtest is a good start, and while assuming you are running two sticks of ram.. Try getting the machine up with just one stick installed. If it works, try the other stick. This should quickly rule out memory issues.
Download and burn a rescue CD from BitDefender or whoever you pref to do a virus scan. Disinfect anything found if possible, else delete it. If anything found and cleaned, try seeing if it'll boot. Worst case here Windows critical file(s) got removed then you'll have to the use the XP install CD to help repair the OS (and pray). Typical rescue CD's are linux desktops, perhaps now is a good time to backup your wiz.bin and whatever other files.
Assuming no virus issues, I'd disconnect all resources from the power supply that isn't NEEDED (CD drive, additional hard drives other than the OS, even the tuners if easily done). Try bringing it up.
If the problem still exist, then I'd move the main OS drive to another machine and run whatever SSD utility to test the drive. Else HDTune (or similar program) and check the drives health and check for errors via a full scan.

Sadly I've no experience with SSD's, tho from what little reading I've done on em, seems there is bit of effort to them vs standard hard drives. And yes you are correct, the up time of 6-8 months is horrid and far below acceptable rates I can't help but wonder what the state of the SSD is currently in. Worst case if one of these are your problem, at least it's likely under warranty.

FYI my server has been running 7yrs before I had any issue. On my current server 4yrs and still going.

Hope this helps

[osx-addict]

Thanks.. In this case the server has no external/internal dvd or cddrive -- all tuners are network tuners (hd-homeruns) so nothing there either. It booted up the memtest cd just fine (using an external USB dvd drive that is usually on my Mac).. I'll let it run overnight as I've got other stuff to do this evening anyway and will do further diags tomorrow after work..

[rgroves]

It may be a bad RAM slot on the mobo. Or even the SSD.

[osx-addict]

Well.. I ran memtest86 thru the night (on ALL of the RAM -- haven't tried removing any yet) and checked it this morning -- 9 passes so far and no errors seen yet.. starting to look like something other than ram at this point but I'll check when I get home (in 12+ hours).. Next thing to try is unplug the SSD and boot into the windows XP install media to see if it crashes or not.. If it does it would appear to be some sort of mobo issue perhaps..

One other thing I've got to try is going into the BIOS and have it reload the factory defaults in the off-chance some setting got messed up in some small way..

[Skirge01]

Don't forget to check for bulging capacitors on the mobo. While the issue isn't as prominent as it was a couple years ago, it definitely still happens.

[panteragstk]

May want to update the BIOS as well. I don't usually do this unless there is a specific issue I'm addressing, but since nothing seems to be broken I'd go ahead and do it just to check it off your list.

[osx-addict]

Thanks for the additional suggestions guys! I've been the recipient of the bulging capacitor issues in the distant past -- albeit on a work-PC -- a Dell if I recall.. Our company had bought hundreds of these particular Dell machines and they started exhibiting spontaneous resets.. Needless to say a big pain in the rear for the 'IT' guys..

As for the BIOS I was thinking the same thing -- I suspect there have been two revs of the BIOS since I built our server this past March.. I noticed several fixes tagged as "stability fixes" or similar..

[osx-addict]

Ok.. Quick followup.. I came home to find what is shown in the image below on the MemTest86 screen -- recall that this morning it was at pass #10 having completed 9 passes successfully at a bit before 7am PST :

photo.JPG

So, I updated the BIOS to the latest version -- without issue, now version 0304 IIRC (from the factory original BIOS -- original version -- dont recall the version #).. Worked like a charm.. However, trying to boot into Windows XP install DVD gets me the following AFTER it loads all desired drivers and indicates "starting windows..." :

photo-1.JPG

So.. I unplugged the SATA cable from all drives -- no change.. I then pulled all of the RAM out and left the bare minimum -- a single 4Gb stick -- no change. I then swapped that RAM board for one of the others -- no change. I then reset the RTC ram settings -- no change. I also did the mobo RAM adjustment (pushing a button on the board) -- no change. So, it seems to indicate that the MB has croaked.. Oh yeah.. I also reset the mobo's settings to factory as I recall.

I'll be contacting Microcenter tomorrow to see how to deal with it -- they may point to Asus since it's ~7 months old.

I'll keep you posted..

[osx-addict]

Could this be caused by a bad/failing PSU possibly? Would a bad PSU allow it to work fine in the BIOS? It behaves fine when the BIOS is up.. Although I was not in the BIOS for more than a handful of minutes..

[jerryt]

Bad power supply
Some bios have a screen showing all the system voltages. If not, you have to unplug from the MB and short two pin (Google) and then measure with a meter.

Could it not be also the processor? You could re-seat it?

(I have enough computers that I can always swap out all of the components)

[rgroves]

In my experience as a PC Service Tech, STOP 7B can be either Power Supply, CPU, RAM, or Motherboard. My first guess is that it's RAM, but more than likely is going to be the motherboard.

[panteragstk]

Bad PSU can cause very strange issues. The reason it may work fine until windows boots is because it is demanding more power to complete the boot. Same with the motherboard, chipset could be going out, could be a short somewhere. When this happens I usually take everything out of the case, grab components that weren't in there to begin with and try to boot windows with another hard drive, ram, psu, and cpu if you have a spare. If everything works, it isn't the motherboard, if it doesn't, then you get to have fun with Asus RMA. The process of elimination is a pain, but you wouldn't want to replace components that actually work.