Local storage data was removed by unkonw reason @ 24.10.beta and RC1

TrueNAS General
, ,
1

1st time@24.10.beta:
data at both local storage and remote server lost(remote server was mounted at storage pool by nfs with read/write, after that i mount remote server with readonly)
2nd time@24.10.beta:
data lost at local storage only.
3rd time@24.10.RC1 updated from BETA:
data lost again at local storage pool.
The data in the docker pool is not affected. I don’t konw the actual time and actual action caused the data was removed since i found the data missed later on. everytime it’s not all data in storage pool missing. from the time of photos taken, the lost photos was taken before sep and the photos in sep and oct is kept well.
if it happend by wrong operation, it’s difficult to remove too many files in different folders. if not find the reason, it should happen again and again.

2

Data lost again@rc2.
Today power on truenas and try to rsync to backup remote files into truenas and find the data at truenas storage lost again.
:joy::joy::joy::joy::joy::joy::joy:
After confirmed the photos，only the photos after 28/sep remained!!!

3

…at least you take it well. Hopefully all these things happening in BETA will be collected and analyzed. Probably fixed in RC versions. Hopefully.

4

I’ll point out that nothing in this report is actionable. Not only do we have 2x record numbers of users running RC1/2, but we aren’t seeing reports of “lost data” like this. We treat these reports seriously, but the fact is historically the vast majority of the time we find out it is some sort of user error doing something “cute” from the command-line.

If you have a valid reproduction case, please write up very clear instructions on how you are inducing data loss so that we and others can review to see if this an issue in the code or your setup.

1 Like
5

don’t mean to steal a thread but since you’re here… .
I removed 2 drives from that 24 drive box of mine. Those 2 drives were not in use, just plugged in. They are bad/have problems:

New alerts:

  * Device: /dev/sdr [SAT], Self-Test Log error count increased from 11 to 12.

Current alerts:

  * Device: /dev/sdr [SAT], Self-Test Log error count increased from 9 to 10.
  * Device: /dev/sdv [SAT], Self-Test Log error count increased from 3 to 4.
  * Device: /dev/sdv [SAT], Self-Test Log error count increased from 4 to 5.
  * Device: /dev/sdr [SAT], 1462 Offline uncorrectable sectors.
  * Device: /dev/sdr [SAT], 1462 Currently unreadable (pending) sectors.
  * Device: /dev/sdr [SAT], Self-Test Log error count increased from 11 to 12.

but there are no reports of HDD removed and I wanted to request it as a feature. Would it make sense to have a report of the event ? like:

  * Device: /dev/sdr [SAT], Was there but not anymore.

Thanks ( please advise ) :slight_smile:

6

While not your cause, if you haven’t already you should setup snapshots so that you can recover from mistakes like this

I made a video which demonstrates my tiered snapshot approach and many ways of recovering data using snapshots

1 Like