Both hard disk fault at same time

Question

Today one of the my computers have strange log.

At 8:00 both hard disks start to have errors like this:

Feb  9 07:24:48 bill2 kernel: ad4: WARNING - WRITE_DMA UDMA ICRC error (retrying request) LBA=62726607            
Feb  9 07:24:48 bill2 kernel: ad6: WARNING - WRITE_DMA UDMA ICRC error (retrying request) LBA=62726607            
Feb  9 07:25:46 bill2 kernel: ad6: WARNING - WRITE_DMA UDMA ICRC error (retrying request) LBA=62726191            
Feb  9 07:48:06 bill2 smartd[1526]: Device: /dev/ad4, ATA error count increased from 1 to 2                       
Feb  9 07:48:08 bill2 smartd[1526]: Device: /dev/ad6, ATA error count increased from 1 to 3

Full log file, smartctl output and other data I uploaded to gist. Here is a link: https://gist.github.com/korjavin/4745963

My question is: may i believe that two hard disk , without error , suddenly get problem in one day / one hour? Or I should find other reason of this? How?

Are they both part of an RAID array where one just failed and was the other heavily stressed while the array was reconstructed> (Using a hot spare). — Hennes, Feb 09 '13 at 17:01
yes, they both providers for gmirror raid1. So you suppose, that true, and both of them start to die? — Korjavin Ivan, Feb 09 '13 at 17:06
I think it is possible, but not very likely. Still, there is the bathtub curve for drives. — Hennes, Feb 09 '13 at 19:26

the-wabbit · Accepted Answer · 2013-02-10T09:28:04.930

WRITE_DMA UDMA ICRC error is a detected transmission error which occurred on a WRITE_DMA command.

I am not too familiar with FreeBSD, but this seems to be an often-documented issue with FreeBSD machines using certain controller types:

https://wiki.freebsd.org/JeremyChadwick/ATA_issues_and_troubleshooting

Also, a hardware failure of the controller or broken cabling is to be considered as the possible cause of the fault.

Both hard disk fault at same time

1 Answers1

Linked