Hardware issue: possibly due to RAM not working anymore?

[vl33l]

Hello,

I discovered that my NAS was not working anymore when I was once at home and not able to access my files anymore (I suspect it was down many days before: I was very busy for many weeks due to lot of things at work and didn't really used my computers at home to access my NAS).

I then went to my NAS, and there was an error message on the screen saying an issue was detected on the power and my motherboard (Asus) shutdown to prevent damages, which was a good thing.

I then though it was the power supply that was not working, but then I tried to power on the NAS, and it went on, I got the Asus load menu but from there impossible to go on the boot menu. I then looked into my reports Freenas was sending me, and noticed this strange thing:

Code:

freenas.local kernel log messages:
> (ada2:ahcich2:0:0:0): WRITE_FPDMA_QUEUED. ACB: 61 18 78 c6 0c 40 17 01 00 00 00 00
> (ada2:ahcich2:0:0:0): CAM status: Uncorrectable parity/CRC error
> (ada2:ahcich2:0:0:0): Retrying command

It looks it's some kind of RAM error? But then, I tried to boot only with one RAM strip, not successful. I then switch to the second one, but on success also.

I don't think it is the motherboard causing an issue, otherwise I wouldn't see the Asus screen. I probably think it's the RAM, but I'm not sure. I was going to point the power supply, but it wouldn't be working at all if the power supply was having an issue right?

If you have any idea, that's very welcome here as I'm out of ideas :)

Thanks!

[EDIT] Below my configuration:

RAM: G.Skill RipjawsX Series, DDR3 1600MHz 16Gb, CL9
Motherboard: Asus A88XM-PLUS
Powersupply: Corsair CX750
Proc: AMD A4 5300
Case: Lian Li PC-V354B
HDD: 3 x WD 6TB Red in RAID5

Version of Freenas: If I'm not mistaken I was running 9.3

 

[Dice]

Could you be a bit more specific on what version of FreeNAS you were running and provide full system specification / configuration please.

 

[vl33l]

Sure, here is my configuration:

RAM: G.Skill RipjawsX Series, DDR3 1600MHz 16Gb, CL9
Motherboard: Asus A88XM-PLUS
Powersupply: Corsair CX750
Proc: AMD A4 5300
Case: Lian Li PC-V354B
HDD: 3 x WD 6TB Red in RAID5

Version of Freenas: If I'm not mistaken I was running 9.3

 

[darkwarrior]

Hello,

I discovered that my NAS was not working anymore when I was once at home and not able to access my files anymore (I suspect it was down many days before: I was very busy for many weeks due to lot of things at work and didn't really used my computers at home to access my NAS).

I then went to my NAS, and there was an error message on the screen saying an issue was detected on the power and my motherboard (Asus) shutdown to prevent damages, which was a good thing.

I then though it was the power supply that was not working, but then I tried to power on the NAS, and it went on, I got the Asus load menu but from there impossible to go on the boot menu. I then looked into my reports Freenas was sending me, and noticed this strange thing:

Code:

freenas.local kernel log messages:
> (ada2:ahcich2:0:0:0): WRITE_FPDMA_QUEUED. ACB: 61 18 78 c6 0c 40 17 01 00 00 00 00
> (ada2:ahcich2:0:0:0): CAM status: Uncorrectable parity/CRC error
> (ada2:ahcich2:0:0:0): Retrying command

It looks it's some kind of RAM error? But then, I tried to boot only with one RAM strip, not successful. I then switch to the second one, but on success also.

If you have any idea, that's very welcome here as I'm out of ideas :)

Thanks!

[EDIT] Below my configuration:

RAM: G.Skill RipjawsX Series, DDR3 1600MHz 16Gb, CL9
Motherboard: Asus A88XM-PLUS
Powersupply: Corsair CX750
Proc: AMD A4 5300
Case: Lian Li PC-V354B
HDD: 3 x WD 6TB Red in RAID5

Version of Freenas: If I'm not mistaken I was running 9.3

Hello,

this error appeared when communicating with the disk ada2.
You should definitely be running long and short tests on that drive and post the output of smartctl -a /dev/ada2

In the end you are also running non-ECC RAM, so it will not be a bad idea to run a full Memtest.

 

[SweetAndLow]

Looks like something is funky with the connection to the disks. Run smart tests and double check your connections.

 

[vl33l]

Hello,

this error appeared when communicating with the disk ada2.
You should definitely be running long and short tests on that drive and post the output of smartctl -a /dev/ada2

In the end you are also running non-ECC RAM, so it will not be a bad idea to run a full Memtest.

I would like to but I'm stuck at the Boot screen Asus, I do press Del or F2 to go on the boot menu but nothing happens.

Looks like something is funky with the connection to the disks. Run smart tests and double check your connections.

I do have couple of tests running regulary, I will post it here once I'm able to boot again but for now I'm stuck with (I guess) a hardware issue on the motherboard or the RAM.

Also, all the components are under warranty so for now I can exchange them. I tested to boot with each RAM strip separately, I was always stuck on the Asus boot screen without being able to jump into the BIOS menu. So I don't really know if it could be a RAM or motherboard issue. I doubt but, could it be related to the HDD? I think if it was an HDD related issue, I would still be able to boot and problems will happen later (typically while trying accessing my data).

Should I just simply call the shop describing my issue? I don't see anything I could do unless changing components since I'm stuck here at boot screen :(

I'm running out of ideas :/ Will check connectivity again

 

[SweetAndLow]

I would like to but I'm stuck at the Boot screen Asus, I do press Del or F2 to go on the boot menu but nothing happens.



I do have couple of tests running regulary, I will post it here once I'm able to boot again but for now I'm stuck with (I guess) a hardware issue on the motherboard or the RAM.

Also, all the components are under warranty so for now I can exchange them. I tested to boot with each RAM strip separately, I was always stuck on the Asus boot screen without being able to jump into the BIOS menu. So I don't really know if it could be a RAM or motherboard issue. I doubt but, could it be related to the HDD? I think if it was an HDD related issue, I would still be able to boot and problems will happen later (typically while trying accessing my data).

Should I just simply call the shop describing my issue? I don't see anything I could do unless changing components since I'm stuck here at boot screen :(

I'm running out of ideas :/ Will check connectivity again

Unplug everything you don't need to boot. This includes all disks, PCI cards, USB things and memory. Leave one stick in the first slot.

Sent from my Nexus 5X using Tapatalk

 

[joeschmuck]

Also make sure the CPU fan is running. The goal @SweetAndLow is getting at it to bring your system down to the minimum components to bootstrap your computer. And make sure you read the use manual for your motherboard to ensure you install the RAM in the correct slot. most motherboards will not bootstrap if the RAM is in not there. And if it fails to boot, try a different stick of RAM.

If you cannot get the system to boot, does it beep (assuming you have a speaker connected to the SPKR connection), and how many beeps is it? Beep codes may be listed in your user manual. If you do get your system booted, power it off and add the rest of the RAM, try again. Keep adding parts one at a time until you fine the faulty part.

You may not be able to bootstrap the computer at all and if that happens, try a different power supply if you have one.

 

[vl33l]

Unplug everything you don't need to boot. This includes all disks, PCI cards, USB things and memory. Leave one stick in the first slot.

Sent from my Nexus 5X using Tapatalk

I have unplug the HDDs, put in a usb stick with memtest and ran it against my two RAM, 0 error appeared so I think we can safely put the RAM outside of the problem.

Also make sure the CPU fan is running. The goal @SweetAndLow is getting at it to bring your system down to the minimum components to bootstrap your computer. And make sure you read the use manual for your motherboard to ensure you install the RAM in the correct slot. most motherboards will not bootstrap if the RAM is in not there. And if it fails to boot, try a different stick of RAM.

If you cannot get the system to boot, does it beep (assuming you have a speaker connected to the SPKR connection), and how many beeps is it? Beep codes may be listed in your user manual. If you do get your system booted, power it off and add the rest of the RAM, try again. Keep adding parts one at a time until you fine the faulty part.

You may not be able to bootstrap the computer at all and if that happens, try a different power supply if you have one.

I unpluged one HDD and tried to boot, until everyone was unplugged but it wasn't able to boot. I did the same with the RAM, but my motherboard requires RAM to boot (but the RAM can be put aside the issue based on the memtest I did)

I noticed a strange noise coming from the powersupply: I was closed to my NAS, and while pressing keys on the keyboard I started hearing a very strange noise, nothing coming from the speakers, but kind of an electrical noise coming from the powersupply and with the inital error message from my motherboard that shutdown my system to prevent damages, I may be related and my issue is maybe coming from my powersupply?

Unfortunately I do not have another powersupply, but I may go to the shop and check with them as all my components are under warranty as I now have isolated a bit more the issue. What do you think?

 

[vl33l]

Oh, one more thing: when I did had the 3 HDDs plugged, it wasn't able to boot on the usb stick with memtest, but when I unplugged one, it booted. This also makes me think it is probably an issue within the powersupply no?

 

[Dice]

I may be related and my issue is maybe coming from my powersupply?

yes.

 

[vl33l]

Ok so I got a Corsair rm650x, quite nice because ultra silent and modular. I was able to boot once, and then I did reboot and now it is not able to boot again. One network led is in orange, and the other keeps 'beeping' in yellow, which doesn't seems to be something good at all...

Probably the powersupply issue has damaged the motherboard, so I have decided to go tomorrow with it to the shop to change it (still under warranty). What do you think?

 

[darkwarrior]

Ok so I got a Corsair rm650x, quite nice because ultra silent and modular. I was able to boot once, and then I did reboot and now it is not able to boot again. One network led is in orange, and the other keeps 'beeping' in yellow, which doesn't seems to be something good at all...

Probably the powersupply issue has damaged the motherboard, so I have decided to go tomorrow with it to the shop to change it (still under warranty). What do you think?

Hi there,
You always have 2 LEDs on a LAN interface and it's perfectly normal to see it blinking. :)

 

[joeschmuck]

Here's the deal... You actually don't know what component is damaging which one. You could be going in circles. For all you know the power supply is now bad. It bothers me that you were able to boot up your system once but not a second time.

Read your user manual, it will tell you what lights mean, like the Ethernet port blinking light which is likely normal activity.

Also, are you certain you are connecting up the motherboard properly, both power connections mated? We have no idea of your understanding about computers and electronics and the way you indicated above the Ethernet port, I'm a bit skeptical. I'm not trying to be confrontational, I'm only trying to ensure that we provide good help.

 

[vl33l]

the Ethernet port was blinking but not as it is normally, as you and darwarrior said there is nothing wrong in the Ethernet port blinking, however it is wrong when it's blinking the way it was.

A quick update on that: I went with my system to the shop, they run couple of tests and they spotted an issue with the motherboard. It's now in the hands of Asus, probably in like 1 or 2 weeks I will get one repaired or a new one. I will update you once I got it back and have it running :) thanks for your help and hopefully my issue will be resolved.

 

[joeschmuck]

Please keep us informed. Sorry to hear about the board.