GU's "flakyness"

14 public posts in this discussion.

Post 1

GU has been unstable as hell since I installed the 2nd CPU and until I can get that fixed (only has issues under load) I'll have to go easy on game servers.

Personally, I think the box hates me - either that or loves toying with me.

Anyway - It's a cross I have to bear.

OD

Post 2

Good luck with that. Somebody seems to have cursed us! :(

Post 3

Quote from grandinferno;359537:
Good luck with that. Somebody seems to have cursed us! :(

Ill blame unseen :D

However thanks OD for getting it back up ;)

Post 4

Yeah cheers for all your hardwork Dan ;)

Post 5

Phew Christmas can go ahead now....

edit - On the plus side I didn't have to suffer rec and his 10 new topics a day

Post 6

Quote from Bad Crispy;359539:
Ill blame unseen :D

However thanks OD for getting it back up ;)

can only pick on the guy that rips your head time and time again ?? ;p

Post 7

did the lights out card work good to help solve this problem ?

Post 8

Re: The server crashes - you hardware guru's, does this look like a ram issue?

Got this from the MCE log.

MCE 0
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 0 4 northbridge TSC dbb88c6eb92d
ADDR 413916c0
Northbridge Chipkill ECC error
Chipkill ECC syndrome = 2254
bit46 = corrected ecc error
bit62 = error overflow (multiple errors)
bus error 'local node response, request didn't time out
generic read mem transaction
memory access, level generic'
STATUS d42a400022080a13 MCGSTATUS 0
MCE 1
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 0 4 northbridge TSC 1b1e956b6bc4d
ADDR 413918a0
Northbridge Chipkill ECC error
Chipkill ECC syndrome = b741
bit46 = corrected ecc error
bit62 = error overflow (multiple errors)
bus error 'local node response, request didn't time out
generic read mem transaction
memory access, level generic'
STATUS d420c000b7080a13 MCGSTATUS 0
MCE 2
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 0 4 northbridge TSC 1b2fb0be756ef
ADDR 41391f60
Northbridge Chipkill ECC error
Chipkill ECC syndrome = 2254
bit32 = err cpu0
bit46 = corrected ecc error
bit62 = error overflow (multiple errors)
bus error 'local node origin, request didn't time out
generic read mem transaction
memory access, level generic'
STATUS d42a400122080813 MCGSTATUS 0
MCE 3
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 0 4 northbridge TSC 1b356485048aa
ADDR 41393740
Northbridge Chipkill ECC error
Chipkill ECC syndrome = b741
bit46 = corrected ecc error
bit62 = error overflow (multiple errors)
bus error 'local node response, request didn't time out
generic read mem transaction
memory access, level generic'
STATUS d420c000b7080a13 MCGSTATUS 0
MCE 4
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 0 4 northbridge TSC 1b54c147b3d4f
ADDR 41393fa0
Northbridge Chipkill ECC error
Chipkill ECC syndrome = 3554
bit46 = corrected ecc error
bit62 = error overflow (multiple errors)
bus error 'local node response, request didn't time out
generic read mem transaction
memory access, level generic'
STATUS d42a400035080a13 MCGSTATUS 0
MCE 5
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 0 4 northbridge TSC 1bc11900370aa
ADDR 413918a0
Northbridge Chipkill ECC error
Chipkill ECC syndrome = b741
bit46 = corrected ecc error
bit62 = error overflow (multiple errors)
bus error 'local node response, request didn't time out
generic read mem transaction
memory access, level generic'
STATUS d420c000b7080a13 MCGSTATUS 0
MCE 6
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 0 4 northbridge TSC 1bc6ccc690629
ADDR 41393fa0
Northbridge Chipkill ECC error
Chipkill ECC syndrome = 3554
bit46 = corrected ecc error
bit62 = error overflow (multiple errors)
bus error 'local node response, request didn't time out
generic read mem transaction
memory access, level generic'
STATUS d42a400035080a13 MCGSTATUS 0
MCE 7
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 0 4 northbridge TSC 1bdac1fcc62df
ADDR 41391a20
Northbridge Chipkill ECC error
Chipkill ECC syndrome = b741
bit46 = corrected ecc error
bit62 = error overflow (multiple errors)
bus error 'local node response, request didn't time out
generic read mem transaction
memory access, level generic'
STATUS d420c000b7080a13 MCGSTATUS 0
MCE 8
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 0 4 northbridge TSC 1c197b828ca38
ADDR 413918a0
Northbridge Chipkill ECC error
Chipkill ECC syndrome = b741
bit32 = err cpu0
bit46 = corrected ecc error
bit62 = error overflow (multiple errors)
bus error 'local node origin, request didn't time out
generic read mem transaction
memory access, level generic'
STATUS d420c001b7080813 MCGSTATUS 0

Post 9

Probably, looks like ECC is going nuts. Can you run memtest on the box at all? Pretty sure any new version of memtest has the ability to look for ECC events. Might need to let it run for a while, obviously a little hard on a live server :p

Post 10

could u run up memtest in a vm and it still tests the ram?
i can imagine not but work investigating.

Post 11

404 Doodoo Error

Post 12

Well i had a look around on the linux geek server forums

Non-fatal mce's are usually ecc faults, and *USUALLY* track back
to bad memory, though it can also be overheating cpu, or a problematic cpu,
or rarely the MB could be the fault.

ECC/MCE counts will get worse under load, unless the problem is
really severe you won't see them at idle.

This is about all anyone had to say about it, most say memory is faulty, but can be other things as above

Post 13

Server time off or is it just me?

seems to be nearly 40 minutes fast :S

Post 14

Set to adelaide time and all fine - it has been crashing more often BUT has been auto restarting fine - just content to leave it atm until I can get the hardware done.