I am running a Western Digital TSD-1000EADS DH hard drive. It is not my system drive. My PCChipps M830LR motherboard (circa 2001) has no SATA controller, so I am using a StarTech PCIESATA2I PCI Card. during periods of disk activity, I have been getting these errors in /var/log/syslog:
This problem does not seem to result in corrupt data on the drive; I have several ~12GB torrents that consistently pass their hash checks. The only identifiable problem resulting from these errors (other than the console being unusable because the errors continuously scroll the screen) is that file reading/writing consumes large amounts of system resources. Sometimes simply copying a file can result in load ratios of around 3.00.Code:Feb 19 09:05:35 thebone kernel: [50918.164220] ata1.00: exception Emask 0x12 SAct 0x0 SErr 0x1000500 action 0x2 Feb 19 09:05:35 thebone kernel: [50918.164265] ata1.00: BMDMA stat 0x5 Feb 19 09:05:35 thebone kernel: [50918.164283] ata1: SError: { UnrecovData Proto TrStaTrns } Feb 19 09:05:35 thebone kernel: [50918.164309] ata1.00: cmd 25/00:00:7f:71:9b/00:01:14:00:00/e0 tag 0 dma 131072 in Feb 19 09:05:35 thebone kernel: [50918.164311] res 51/84:af:7f:71:9b/84:00:14:00:00/e0 Emask 0x12 (ATA bus error) Feb 19 09:05:35 thebone kernel: [50918.164342] ata1.00: status: { DRDY ERR } Feb 19 09:05:35 thebone kernel: [50918.164358] ata1.00: error: { ICRC ABRT } Feb 19 09:05:35 thebone kernel: [50918.487372] ata1: soft resetting link Feb 19 09:05:35 thebone kernel: [50918.647235] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 310) Feb 19 09:05:35 thebone kernel: [50918.707356] ata1.00: configured for UDMA/33 Feb 19 09:05:35 thebone kernel: [50918.707385] ata1: EH complete Feb 19 09:05:35 thebone kernel: [50918.724057] sd 0:0:0:0: [sda] 1953525168 512-byte hardware sectors (1000205 MB) Feb 19 09:05:35 thebone kernel: [50918.724082] sd 0:0:0:0: [sda] Write Protect is off Feb 19 09:05:35 thebone kernel: [50918.724087] sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00 Feb 19 09:05:35 thebone kernel: [50918.724115] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Things I have tried so far, and other observations:
1. I have run a complete fsck, with the -f option. It said "File System Modified", but did not indicate that there were any problems otherwise.
2. I have looked at this page: http://ata.wiki.kernel.org/index.php...error_messages and I think what the error message are telling me is that there is a communications error between the PCI bus and the controller card, but getting down to that level is a little over my head.
3. I jumpered this down to 1.5 from 3, its default, when I was trying to install it in a different machine. I ultimately gave up on that because Ubuntu could not successfully start up with the drive and controller card in the computer. It is still jumpered down.
4. When if first got the drive, I put it in the same hardware and was using it under windows (again not the system drive) formatted as NTFS. It did not appear to have any problems, however I did not look at the Windows logs.
5. I have tried at least one other SATA cable.
6. I have tried a different hard drive, and that one does not experience these problems. It was ntfs, not ext3.
7. I have tried installing this drive and card in another computer, a Dell Dimension 3000 (circa 2005). In this new setup, this drive was still a data drive and another drive served as the system drive (similar to the previous machine). OS was fresh install of Hardy. Same problem. I upgraded the machine to Lucid, same problem.
8. My scenario for testing the drive involves copying a large file from one directory to another, or making a copy in the same directory. The errors show up when the copy starts, and if I abort the copy, the errors stop. If I repeat the same action, the errors do NOT show up right away. Some sort of cache?
Right now there is no irreplacable data on the drive. It's about half full.
Adv Reply
Bookmarks