smartctl makes kernel hang

[HDER]

run smartctl by cron task every day.

sometimes make my workstation in bad stat(can not do anything), and screen print(kernel 2.4.35)

ata1: command 0x35 timeout, stat 0x58 host_stat 0x21
ATA: abnormal status 0x58 on port 0x9F87

i have 2 workstation, kernel 2.4.35 and kernel 2.6.
i have replace a new WD drive and a new seagate drive, but no improve.

[HDER]

在 2012年11月5日星期一UTC+8上午11时37分13秒，HDER写道：

i had dump the kernel stack of smartctl

[<8011ebc0>] <wait_for_completion> (76) totalstack=76 at vmlinux-2.4.35.4-wano-7.5cn-64G.S +53245
[<802c32f8>] <scsi_insert_special_req> (16) totalstack=92 at vmlinux-2.4.35.4-wano-7.5cn-64G.S +559034
[<802bc5a4>] <scsi_wait_req> (56) totalstack=148 at vmlinux-2.4.35.4-wano-7.5cn-64G.S +550860
[<802bbe60>] <scsi_wait_done> (0) totalstack=148 at vmlinux-2.4.35.4-wano-7.5cn-64G.S +550317
[<802e7bda>] <ata_cmd_ioctl> (72) totalstack=220 at vmlinux-2.4.35.4-wano-7.5cn-64G.S +603096
[<802e7f4a>] <ata_scsi_ioctl> (24) totalstack=244 at vmlinux-2.4.35.4-wano-7.5cn-64G.S +603339
[<802bf150>] <scsi_ioctl> (44) totalstack=288 at vmlinux-2.4.35.4-wano-7.5cn-64G.S +554193
[<802f228e>] <sd_ioctl> (44) totalstack=332 at vmlinux-2.4.35.4-wano-7.5cn-64G.S +615860
[<80154b49>] <pipe_write> (56) totalstack=388 at vmlinux-2.4.35.4-wano-7.5cn-64G.S +120018
[<80151c5e>] <blkdev_ioctl> (16) totalstack=404 at vmlinux-2.4.35.4-wano-7.5cn-64G.S +116485
[<8015ac85>] <sys_ioctl> (40) totalstack=444 at vmlinux-2.4.35.4-wano-7.5cn-64G.S +127102
[<8010b087>] <system_call> (40) totalstack=484 at vmlinux-2.4.35.4-wano-7.5cn-64G.S +28505
totalstacksize = 484