j7n Posted November 12, 2022 Share Posted November 12, 2022 Maybe because the disks have been running continuously with few power cycles. Link to comment Share on other sites More sharing options...
D.Draker Posted November 13, 2022 Share Posted November 13, 2022 Hmm, everything is possible. I had many disks with thousands of power cycles and all was fine. Older disks are more reliable. Because of lower capacity. Link to comment Share on other sites More sharing options...
j7n Posted November 13, 2022 Share Posted November 13, 2022 (edited) I had two ST 250 GB PATA drives fail after a few years, with either weak heads or debris inside. They are only slightly older model than the 320 G. Clearly anecdotal experience with one doesn't apply to the other. New drives since 2012 or so have a built in function to park their heads after a short period of idle time. This normally only happens when power is removed. The drive might accumulate tens of thosuands of head parking cycles during normal use (S.m.a.r.t C1). This could be disabled by changing the advanced power management (APM) value using a procedure that applies a DCO and inadvertently also saves the APM value. On the most recent models this doesn't work. One must use Seagate OpenSeaChest utility (requires NT 6.1 or making Linux boot disk). I am still puzzled by he hours counter. Would be good to hear if anyone has it at greater than 66,000. Edited November 13, 2022 by j7n Link to comment Share on other sites More sharing options...
D.Draker Posted November 13, 2022 Share Posted November 13, 2022 3 hours ago, j7n said: I had two ST 250 GB PATA drives fail after a few years, with either weak heads or debris inside. They are only slightly older model than the 320 G. Clearly anecdotal experience with one doesn't apply to the other. New drives since 2012 or so have a built in function to park their heads after a short period of idle time. This normally only happens when power is removed. The drive might accumulate tens of thosuands of head parking cycles during normal use (S.m.a.r.t C1). This could be disabled by changing the advanced power management (APM) value using a procedure that applies a DCO and inadvertently also saves the APM value. On the most recent models this doesn't work. One must use Seagate OpenSeaChest utility (requires NT 6.1 or making Linux boot disk). I am still puzzled by he hours counter. Would be good to hear if anyone has it at greater than 66,000. I usually run the WD tool to disable the stupid idle parking. But it won't work on 8TB, sadly. You know why? Cause 8tb is NOT WD anymore, they bought Hitachi and it's just a re-labeled Hitachi sold as WD, that's why the tool doesn't work. Seagate always failed me. I had only one good seagate 4tb (enterprise model 5400RPM). But it was so damn noisy ! 66k ? Nah, I usuallly ditch old drives, not safe to use. The oldest I found here with me is abot 33-34k on it, guess you won't be impressed. Link to comment Share on other sites More sharing options...
Tripredacus Posted November 14, 2022 Author Share Posted November 14, 2022 All the drives in my first post are still running, but the first 80 GB one has two SMART warnings now and really should replace it. 1 Link to comment Share on other sites More sharing options...
Tripredacus Posted March 8 Author Share Posted March 8 (edited) Drive C was in Caution for about a year before I ran into some slowdown and Bad Blocks, so it had to be replaced. The replacement disk is the same size, but it already has 128,673 hours (14 years) on it. My two PATA disks are at 177527 and 177682 (20 years) And an odd-ball situation, this disk was not in the original post but it has the wrong hours showing at just 1487 which is a couple of months. This disk is over 10 years old so not sure why the hours are showing at such a low number. EDIT: Mystery solved regarding the low hour drive. What ended up happening was that I have both the C and G disk in an enclosure, and both were WD800AAJS. I got the serial number of the C drive and marked it on the disk, but I didn't mark the G drive. I was counting on being able to tell which were the original disks by the amount of dust on them. BUT one of the potential replacement disks was also a dusty WD800AAJS. I ran into problems cloning the C drive because Clonezilla was not erasing the destination disk and had multiple partitions, so the clone was always failing. And then when it did work, the OS booted into startup repair because it had the bootloader from the replacement disk and the OS from my disk. Eventually I got it sorted but in the end what ended up happening is that I put one of the potential replacement disks into the enclosure and ended up imaging my C disk to my G disk. And the new G disk is low hours because it was the actual replacement disk I was supposed to be cloning to. I have the data backed up so it didn't turn out to be a problem in the end but I was sweating for a bit once I realised what happened. Moral of the story: mark all of your disks before doing this kind of work. Edited March 9 by Tripredacus Mystery solved 1 Link to comment Share on other sites More sharing options...
j7n Posted April 19 Share Posted April 19 My oldest disks now have 140,000 hours. The counters for ST2000VN and ST200VX have reset again! You can see that the current percentage rating is 86 and the worst it has ever been is 11. They don't seem to want to enter a leaderboard? The error count is from a bad cable (ST3320) and use of the HDAT2 tool (ST2000) where DMA was not correctly implemented. Link to comment Share on other sites More sharing options...
Tripredacus Posted April 19 Author Share Posted April 19 I have been using CrystalDiskInfo as a general disk testing tool and have found a couple disks that don't show power on hours info. It shows a dash symbol or is blank. At about what hour mark do the two 2 TB disks reset their time at? Link to comment Share on other sites More sharing options...
j7n Posted April 19 Share Posted April 19 (edited) I haven't caught that moment. If I did, I might have a clue what caused it. If the scale was linear (excluding that it stops at 1), then a rating of 11 would put the counter at 76-82 thousand, which is over the 16-bit mark. The other values of seeks/errors done are happily holding 34 bits, and the temperature is two fields packed with something in bits 37-33. High-fly writes (0xBD) also seems to top out at 1. It seems to be a strict statistic, but hasn't caused problems yet. Edited April 19 by j7n Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now