Take the 2-minute tour ×
Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

Not sure, is this right site for this Q, but let me try

Last time i have problem with hard disk. Sometimes its do strange sound, and i get it from logs:

$dmesg | grep ata4

[29409.945516] ata4.00: exception Emask 0x10 SAct 0xf SErr 0x90202 action 0xe frozen

[29409.945529] ata4.00: irq_stat 0x00400000, PHY RDY changed

[29409.945538] ata4: SError: { RecovComm Persist PHYRdyChg 10B8B }

[29409.945546] ata4.00: failed command: READ FPDMA QUEUED

[29409.945562] ata4.00: cmd 60/30:00:56:22:5f/00:00:00:00:00/40 tag 0 ncq 24576 
in
[29409.945573] ata4.00: status: { DRDY }

[29409.945580] ata4.00: failed command: READ FPDMA QUEUED

[29409.945594] ata4.00: cmd 60/18:08:8e:22:5f/00:00:00:00:00/40 tag 1 ncq 12288 
in
[29409.945605] ata4.00: status: { DRDY }

[29409.945611] ata4.00: failed command: READ FPDMA QUEUED

[29409.945625] ata4.00: cmd 60/08:10:46:02:66/00:00:00:00:00/40 tag 2 ncq 4096 
in
[29409.945635] ata4.00: status: { DRDY }

[29409.945641] ata4.00: failed command: READ FPDMA QUEUED

[29409.945656] ata4.00: cmd 60/80:18:ee:04:66/00:00:00:00:00/40 tag 3 ncq 65536 
in
[29409.945666] ata4.00: status: { DRDY }

[29409.945679] ata4: hard resetting link

[29413.976083] ata4: softreset failed (device not ready)

[29413.976097] ata4: applying SB600 PMP SRST workaround and retrying

[29414.148070] ata4: SATA link up 3.0 Gbps (SStatus 123 SControl 300)

[29414.184986] ata4.00: SB600 AHCI: limiting to 255 sectors per cmd

[29414.243280] ata4.00: SB600 AHCI: limiting to 255 sectors per cmd

[29414.243292] ata4.00: configured for UDMA/133

[29414.243324] ata4: EH complete

[680674.804563] ata4: exception Emask 0x50 SAct 0x0 SErr 0x90a02 action 0xe 
frozen
[680674.804575] ata4: irq_stat 0x00400000, PHY RDY changed

[680674.804584] ata4: SError: { RecovComm Persist HostInt PHYRdyChg 10B8B }

[680674.804603] ata4: hard resetting link

[680678.840561] ata4: softreset failed (device not ready)

Is this ata4 sata hard drive dead? Must i change it ASAP ? Need I specify more info?

share|improve this question
    
If you do not have a backup and the disk is readable then get a backup right now. As for should you replace the disk - yes, as Janne says. Apart from anything else, drives are usually cheap these days, certainly in comparison to the cost of an engineer's time (I'm assuming this isn't a home computer disk if you're asking on server fault) and very definitely in comparison to the cost of losing any data or availability of a business service. –  RobM Nov 14 '11 at 8:56
1  
thank you all. Disk replaced, data saved (md raid here, so i just rebuild array with new disk) –  Korjavin Ivan Nov 15 '11 at 5:24

3 Answers 3

up vote 11 down vote accepted

Replace your drive immediately, especially if that 'strange sound' is a clicking noise.

share|improve this answer
1  
Rule 0 - if you are even slightly unsure about a disk, and it is supposed to house anything you want to keep, replace it. If you can't easily replace it (no backup?) you need to get yourself in a situation where you can ASAP. Remember Disks are cheaper than data. –  Tom Newton Jul 9 '12 at 13:24

I would suggest you to change your hard drive. Noise seems to be dangerous.

share|improve this answer

THe clicking is an issue of course, but I found this question investigating a similar error on our server, and in our case the issue was NOT the drive, but a problem with WD (Western Digital) and faulty NCQ implementation!

You can read about it on these links to see if it fits your issue:

But the error (specifically this part) looks similar enough that I found this question :)

ata6.00: exception Emask 0x10 SAct 0x0 SErr 0x1810000 action 0xe frozen
ata6.00: irq_stat 0x00400000, PHY RDY changed
ata6: SError: { PHYRdyChg LinkSeq TrStaTrns }

For us, short term fix/test was this:

echo 1 > /sys/block/sd{a,b,c,d}/device/queue_depth

Long term would be to add it to the blacklist as you can read in the references. I have NO experience with that, but link 2 says a patch would look like this:

--- a/drivers/ata/libata-core.c 2010-05-20 20:39:08.000000000 +0200
+++ b/drivers/ata/libata-core.c 2010-05-20 20:43:54.000000000 +0200
@@ -3924,6 +3924,7 @@
        { "Maxtor 7V300F0",     "VA111630",     ATA_HORKAGE_NONCQ },
        { "ST380817AS",         "3.42",         ATA_HORKAGE_NONCQ },
        { "ST3160023AS",        "3.42",         ATA_HORKAGE_NONCQ },
+       { "WDC WD2502ABYS-5*",  NULL,           ATA_HORKAGE_NONCQ },
share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.