After rebooting bad sectors can be accessed again. Why?

[bigspace]

I have a Seagate HDD in a Linux system. I bought it last year. Last week I found there were many read and write errors for it. From the log, I thought there were many bad sectors on it. But after I reboot the system, I didn't find sector errors any more. And then I did a scan from the beginning of the drive to the end, and didn't find any errors again. It's strange. If some sectors are bad, then in theory I can't read or write them once more. Why I can access them after rebooting? Are there any problems with FW of HDD?

[fzabkar]

Does SMART show any reallocated sectors? Pending sectors?

[bigspace]

It doesn't show any reallocated sectors.

[fzabkar]

Maybe there were pending sectors, and these may have been retested and found to be OK.

[bigspace]

Maybe there were pending sectors, and these may have been retested and found to be OK.

Thank you very much. Maybe they are pending sectors. Are pending sectors unstable sectors that can be read but the FW tries to remap them?

[fzabkar]

If a drive has difficulty reading a particular sector, it marks it as "pending reallocation". At some future time, such as when the OS writes to it, this sector will be retested and returned to service if good, or replaced with a spare if confirmed bad.