not certain if I need to replace drive

[ccssid]

Volume status: healthy
after scrub completed: errors: 0 repaired: 0

but, I get these notifications:

• 
  CRITICAL: Dec. 4, 2019, 4:23 a.m. - Device: /dev/da2 [SAT], Temperature 40 Celsius reached critical limit of 40 Celsius (Min/Max 36/40!)
• 
  
• CRITICAL: Dec. 3, 2019, 8:21 p.m. - Device: /dev/ada4, 2 Currently unreadable (pending) sectors
• 
  
• CRITICAL: Dec. 3, 2019, 8:51 p.m. - Device: /dev/ada4, new Self-Test Log error at hour timestamp 25629
• 
  
• CRITICAL: Dec. 4, 2019, 4:23 a.m. - Device: /dev/da1 [SAT], Temperature 40 Celsius reached critical limit of 40 Celsius (Min/Max 36/40!)
• 
  
• CRITICAL: Dec. 4, 2019, 5:23 a.m. - Device: /dev/ada0, new Self-Test Log error at hour timestamp 25866

OK: Dec. 3, 2019, 8:21 p.m. - There is a new update available! Apply it in System -> Update tab.

I have read through many posts and can not quite figure this one out. I built my system a couple of years ago and it has been running flawlessly. Do I need to address ada4 and ada0 now or wait until I get a degraded posting.? Also, since I built the system, I have been away from all the technical stuff regarding FreeNas.

 

[PhiloEpisteme]

Hi @ccssid. In the future for readability you may consider using code tags to make code look cleaner.

[code]
some code here
[/code]

Do I need to address ada4 and ada0 now or wait until I get a degraded posting.?

Very likely, yes. Unreadable sectors and SMART test failures are indications that a disk is on its way out.

I have been away from all the technical stuff regarding FreeNas.

Hopefully during that time the system has been running regularly scheduled SMART tests and scrubs at the very least and someone gets those emails. A system can work flawlessly for years; but without some sort of monitoring you may not know about issues until it is too late.

Would you mind posting your system specs? I assume it is the system in your signature but signatures can change over time so posting in the thread helps keep the post relevant as your build changes. One important bit is what vdev type(s) you're using.

 

[ccssid]

Hi @ccssid. In the future for readability you may consider using code tags to make code look cleaner.

[code]
some code here
[/code]



Very likely, yes. Unreadable sectors and SMART test failures are indications that a disk is on its way out.


Hopefully during that time the system has been running regularly scheduled SMART tests and scrubs at the very least and someone gets those emails. A system can work flawlessly for years; but without some sort of monitoring you may not know about issues until it is too late.

Would you mind posting your system specs? I assume it is the system in your signature but signatures can change over time so posting in the thread helps keep the post relevant as your build changes. One important bit is what vdev type(s) you're using.

Yes , I monitor the sytem periodically. I use this as a home server. My system hardware has not changed. Other than the removal of dust from time to time...I have left it untouched.

A stupid question I know, but, why would it state "0" errors and a healthy main volume if there are problems picked up during a smart test , ureadable sectors..etc?

Freenas: FreeNAS-9.10.2-U2 (e1497f2)
CPU: Xeon e3 1230 v5
MB : Asrock c236 ws
RAM: 32 GB, unbuffered ecc
HDD: 10 @ 3TB, wd red (z2)
BOOT DRIVE: Transcend 32 GB ssd370s
PSU: Seasonic G series ssr 550 rm (550 watts)
HBA: LSI 9211-8I, flashed to IT mode
CyberPower 1500AVR

 

[ccssid]

@PhiloEpisteme
Thank you

 

[PhiloEpisteme]

A stupid question I know, but, why would it state "0" errors and a healthy main volume if there are problems picked up during a smart test , ureadable sectors..etc?

SMART tests indicate potentially impending failure. The system may still be running just fine between the drive reallocating sectors etc but if the drive fails entirely the pool will become degraded. If enough drives outright fail the pool will fail completely.

 

[ccssid]

Sounds good. I figured this may be the case. Just ordered a couple of WD Red ,this morning...just in case.
The 2 drives in question are 2 months out of warranty.....go figure
Thanks again

 

[PhiloEpisteme]

ha, of course, I bet those drives planned it that way! Anyway; yeah go ahead and burn in those drives and follow the User Guide for replacement procedures and you should be all set. Since you're experiencing issues with two drives and it is a RAIDZ2 system you'll want to prioritize this work since if these two fishy drives outright die you're out of redundancy and any unrecoverable read errors or drive failures could result in file corruption or outright data loss.

 

[ccssid]

In the meantime, I am backing up all my files that are residing on my FreeNas server! As well as my FreeNas configuration file.