SMART error (CurrentPendingSector) detected on host

[tgates]

I recently set up a bare metal cloud server with a hosting company. I have 2 ZFS pools, 1 with sda/sdb and the other with NVME drives. I have been getting these errors for both sda and sdb:

This message was generated by the smartd daemon running on:

host name: midwestern
DNS domain: pve-pbs.local

The following warning/error was logged by the smartd daemon:

Device: /dev/sdb [SAT], 1 Currently unreadable (pending) sectors

Device info:
CT4000MX500SSD1, S/N:2336E8735B42, WWN:5-00a075-1e8735b42, FW:M3CR046, 4.00 TB

For details see host's SYSLOG.

You can also use the smartctl utility for further investigation.
Another message will be sent in 24 hours if the problem persists.

Both drives are identical with less than 500 hours on each. I did a smart short test with no errors, both have 0 for Reallocated_Sector_Ct. This drive pool is used for my backup storage and only sees ~20 GB/day of writes with minimal reads.

Could this be an indication of bad drives? I find it unlikely that 2 brand-new 4 TB SSDs are dying after 500 hours with minimal reads/writes.

 

[deepcloud]

Hi

This "unreadable (pending) sectors" is a sign of a bad drive. You are using consumer drives and this is expected. Even I have seen this with consumer drives, I have tried using the Crucial MX500 series as boot drives and did not have a great experience with a mirrored setup, you keep them as single drives and they work. if you want to mirror pls revert to using enterprise drives only which is what we too ended up doing