Spinning Rust Attrition

[CwF]

A short Labor Day read

First off, I probably will buy another one and likely a Toshiba. I have many others that I'd prefer not to put into service again since they are too small for how I use old HD's.

Only 3 hard drives are in use among 8 ssd's and I still have not decided an old debate. They should be off when not in use, their use pattern should be a back up destination, powered up only for the operation, then off...On the other hand I firmly believe the power cycle is the worst thing for longevity = yes they last longer when you never turn them off! Old timers may remember stories of hard drives in use for thousands of hours with no evidence of failure at all failing a cold restart. It happens, it happened.

In an older post I commented on two instances where power poles where wiped out and forced a shutdown. The uptime in between was slightly over a year. No, I did not reboot this rig for over a year and no it is not an online computer. I've had hours of backup power so that these instances were always controlled shutdowns. It seems the recent growth around me has changed power reliability and I've had 5 outages in the last 2 months. The last two were back to back where my complicity did not foresee the second round outage being long enough to exhaust the already depleted battery capacity. Not good. Two hypervisors experienced plug-pull test.

Both survived, all ssd's are fine, all vms fine, 2 of the HD's didn't complain. This one HD was missing, no listing at all. It's also the only one internally mounted, of course. So I had to pull it apart and remove the drive. Everything else is in bay cages which is much better, but they still suck, stick, require an alarming amount of force to remove, sometimes some needle nose pliers. They piss me off...but it's better.

Trying the drive within the case I couldn't hear it. Now in one of the 3.5 external bays I could better watch and listen, and could do it live instead of only at boot. It was on a SCU controller without hot-plug support, the bay has that convenient feature. So stab it in the bay, be quiet please, apply power = bzzp,bzzp,bzzp,bzzp,bzzp,bzzp while the power led matched the rhythmic attempts to spin up the drive. Yep, it's stuck.

So I took it out, strategically angled the drive and with perfectly calibrated force rapped it on the wood floor...stuck it back in the bay, be quiet, apply power - the power led was solid, then magically the main HD activity LED (all drives) blinked and an icon appeared on the desktop - "DVR" BINGO!
I will let it spin a bit before I mount it and pull anything I need. I did do an immediate check

Code: Select all

ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate     0x000f   112   078   006    Pre-fail  Always       -       46064318
  3 Spin_Up_Time            0x0003   094   094   000    Pre-fail  Always       -       0
  4 Start_Stop_Count        0x0032   100   100   020    Old_age   Always       -       463
  5 Reallocated_Sector_Ct   0x0033   100   100   036    Pre-fail  Always       -       11
  7 Seek_Error_Rate         0x000f   074   060   030    Pre-fail  Always       -       29333210
  9 Power_On_Hours          0x0032   019   019   000    Old_age   Always       -       70957
 10 Spin_Retry_Count        0x0013   100   100   097    Pre-fail  Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   020    Old_age   Always       -       231
183 Runtime_Bad_Block       0x0032   100   100   000    Old_age   Always       -       0
184 End-to-End_Error        0x0032   100   100   099    Old_age   Always       -       0
187 Reported_Uncorrect      0x0032   001   001   000    Old_age   Always       -       17683
188 Command_Timeout         0x0032   100   100   000    Old_age   Always       -       0
189 High_Fly_Writes         0x003a   100   100   000    Old_age   Always       -       0
190 Airflow_Temperature_Cel 0x0022   071   052   045    Old_age   Always       -       29 (Min/Max 26/29)
194 Temperature_Celsius     0x0022   029   048   000    Old_age   Always       -       29 (0 18 0 0 0)
195 Hardware_ECC_Recovered  0x001a   034   013   000    Old_age   Always       -       46064318
197 Current_Pending_Sector  0x0012   100   097   000    Old_age   Always       -       16
198 Offline_Uncorrectable   0x0010   100   097   000    Old_age   Offline      -       16
199 UDMA_CRC_Error_Count    0x003e   200   200   000    Old_age   Always       -       0
240 Head_Flying_Hours       0x0000   100   253   000    Old_age   Offline      -       71488 (4 221 0)
241 Total_LBAs_Written      0x0000   100   253   000    Old_age   Offline      -       2469506867
242 Total_LBAs_Read         0x0000   100   253   000    Old_age   Offline      -       3539384632

I can't complain with 8.1 years of use!
I am somewhat more disciplined with the other HD's and they are off most of the time. A few years ago I started using RAM for the primary 'DVR' which is TV recordings that recycle. On the disc is archived stuff available to watch, and accumulated recent raw recordings I might want to watch. And even with 'discipline' I'm sure there are gigs and gigs of stuff I never backed up and forgot about, and really wouldn't miss. Everything actually important is many versions deep in multiple back ups.

Who wants to bet I'll tempt fate and continue to use it in the bay cage...

That's long enough, let's see

Code: Select all

$ systemd-mount /dev/disk/by-label/dvr ~/Videos/DVR
Started unit home-user-Videos-DVR.mount for mount point: /home/user/Videos/DVR

It's all there!
This other computer has a thumbnailer, annoying...

[sunrat]

I reckon 70957 hours is a long healthy life for a HDD. Congratulations!
I don't trust any drive whether HDD or SSD to not fail so I keep at least 2 backups of all important data on external drives. $HOME is also backed up to a separate internal drive weekly using KDE Plasma's default backup utility.

[CwF]

Indeed a good run!
It is running fine at the moment and I did pull gigs off of decent stuff nice to have. It is in an easy spot now, no more hard to get to drives anywhere. I wasn't using the bay very often, but it is in the 'wrong' computer.

The stats are not that bad really. In the scsi days I also had bays, sleds, backplanes for this purpose of live fixes. I seem to remember thousands of relocated sectors and very few mechanical issues. I think I still have a set of 6 80pin from an array that went over 100k hours (~99-2012) and I remember zeroing them out a few years after that looking for the quiet ones and they all still worked fine. They were built to a higher standard but with very low density.
During those same years nearly every consumer IDE drive failed <50k, 5 years seemed an upper limit. I think I warrantied a handful of sata's also within 5 years. Under 1TB they're not worth the space, noise, and power consumption even for frivolous dvr use. The next qualifier is the need to at least be able to max out network speeds, many can't. It's amazing how slow storage used to be.

This particular Seagate 7200.1 has a refurb sticker on it and I can't remember its exact history. It's quiet and just pushed 170MBs and averaged 111MBs across the network to a temp 2.5" smr drive that can barely keep up. They kinda suck btw, but for 'worm' use they are more convenient than a full 3.5".

For the record, the assorted SSD's are approaching 50k, jury's still out.

[CwF]

Update!
ssd, OS1

Code: Select all

 2407-4
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  5 Reallocated_Sector_Ct   0x0033   100   100   010    Pre-fail  Always       -       0
  9 Power_On_Hours          0x0032   087   087   000    Old_age   Always       -       61541
 12 Power_Cycle_Count       0x0032   099   099   000    Old_age   Always       -       618
177 Wear_Leveling_Count     0x0013   091   091   000    Pre-fail  Always       -       98
179 Used_Rsvd_Blk_Cnt_Tot   0x0013   100   100   010    Pre-fail  Always       -       0
181 Program_Fail_Cnt_Total  0x0032   100   100   010    Old_age   Always       -       0
182 Erase_Fail_Count_Total  0x0032   100   100   010    Old_age   Always       -       0
183 Runtime_Bad_Block       0x0013   100   100   010    Pre-fail  Always       -       0
187 Uncorrectable_Error_Cnt 0x0032   100   100   000    Old_age   Always       -       0
190 Airflow_Temperature_Cel 0x0032   071   043   000    Old_age   Always       -       29
195 ECC_Error_Rate          0x001a   200   200   000    Old_age   Always       -       0
199 CRC_Error_Count         0x003e   099   099   000    Old_age   Always       -       23
235 POR_Recovery_Count      0x0012   099   099   000    Old_age   Always       -       418
241 Total_LBAs_Written      0x0032   099   099   000    Old_age   Always       -       9988917860

ssd OS2

Code: Select all

2407-4
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  5 Reallocated_Sector_Ct   0x0033   100   100   010    Pre-fail  Always       -       0
  9 Power_On_Hours          0x0032   091   091   000    Old_age   Always       -       40708
 12 Power_Cycle_Count       0x0032   099   099   000    Old_age   Always       -       250
177 Wear_Leveling_Count     0x0013   001   001   000    Pre-fail  Always       -       504
179 Used_Rsvd_Blk_Cnt_Tot   0x0013   100   100   010    Pre-fail  Always       -       0
181 Program_Fail_Cnt_Total  0x0032   100   100   010    Old_age   Always       -       0
182 Erase_Fail_Count_Total  0x0032   100   100   010    Old_age   Always       -       0
183 Runtime_Bad_Block       0x0013   100   100   010    Pre-fail  Always       -       0
187 Uncorrectable_Error_Cnt 0x0032   100   100   000    Old_age   Always       -       0
190 Airflow_Temperature_Cel 0x0032   061   039   000    Old_age   Always       -       39
195 ECC_Error_Rate          0x001a   200   200   000    Old_age   Always       -       0
199 CRC_Error_Count         0x003e   099   099   000    Old_age   Always       -       1
235 POR_Recovery_Count      0x0012   099   099   000    Old_age   Always       -       135
241 Total_LBAs_Written      0x0032   099   099   000    Old_age   Always       -       15933509869

...and here's what bad weather on the horizon looks like, the persistent spinner!

Code: Select all

ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate     0x000f   105   078   006    Pre-fail  Always       -       140276726
  3 Spin_Up_Time            0x0003   096   094   000    Pre-fail  Always       -       0
  4 Start_Stop_Count        0x0032   100   100   020    Old_age   Always       -       483
  5 Reallocated_Sector_Ct   0x0033   070   070   036    Pre-fail  Always       -       1253
  7 Seek_Error_Rate         0x000f   075   060   030    Pre-fail  Always       -       29806754
  9 Power_On_Hours          0x0032   015   015   000    Old_age   Always       -       74895
 10 Spin_Retry_Count        0x0013   100   100   097    Pre-fail  Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   020    Old_age   Always       -       241
183 Runtime_Bad_Block       0x0032   091   091   000    Old_age   Always       -       9
184 End-to-End_Error        0x0032   100   100   099    Old_age   Always       -       0
187 Reported_Uncorrect      0x0032   001   001   000    Old_age   Always       -       18177
188 Command_Timeout         0x0032   100   093   000    Old_age   Always       -       15
189 High_Fly_Writes         0x003a   100   100   000    Old_age   Always       -       0
190 Airflow_Temperature_Cel 0x0022   060   051   045    Old_age   Always       -       40 (Min/Max 40/41)
194 Temperature_Celsius     0x0022   040   049   000    Old_age   Always       -       40 (0 18 0 0 0)
195 Hardware_ECC_Recovered  0x001a   021   013   000    Old_age   Always       -       140276726
197 Current_Pending_Sector  0x0012   046   046   000    Old_age   Always       -       2244
198 Offline_Uncorrectable   0x0010   046   046   000    Old_age   Offline      -       2244
199 UDMA_CRC_Error_Count    0x003e   200   200   000    Old_age   Always       -       0
240 Head_Flying_Hours       0x0000   100   253   000    Old_age   Offline      -       75448 (77 60 0)
241 Total_LBAs_Written      0x0000   100   253   000    Old_age   Offline      -       1077233897
242 Total_LBAs_Read         0x0000   100   253   000    Old_age   Offline      -       1999010746

[Uptorn]

Even a UPS is no guarantee of sustained power, as I've discovered. That is one area where I must concede that laptops are better.

[CwF]

Even a UPS is no guarantee of sustained power, as I've discovered. That is one area where I must concede that laptops are better.

Well I wouldn't get to that conclusion since the concept is configurable.
I mentioned, or didn't, that my current state of power backup is <15 minutes but that has not always been the case. I cost some efficiency (~20%) depending on how it's done. I have been on systems with two stages and more time than would be possible with a laptop on stage one, infinite on stage 2 depending only on the natural gas supply.

Recently I found some gadgets from old projects I might try to revive with debian tools if possible. From fusion brains to multichannel dc-dc regulators and multichannel DAC crossovers. Odd stuff, all usb devices, all support long gone. All rejects from prior projects. I'm looking to build a better power backup now that I have to, again.

Many of my hard drives in used inventory lived at 10,000'. That is very hard on drives. SSD's don't care.