ZFS ZFS Dataset Size is out of control, ZFS SEND/RECV hangs on random datasets

I reinstated my second remote backup of the smtp server and we're over a day with no hangs.

It has been a couple weeks without crashes (stopped after the SSD was replaced). I threw out the SSD that was removed today.

In future I will go to SSD quicker. I was irrationally convinced that SSD was fine and it had to be RAM, motherboard or something else.
 
I was referring to idprio/renice as a way to test loading the system in a way that would likely not cause increasing clock frequency for getting a better idea of when it adjusting vs not adjusting.

Used to be that CPUs were so reliable that we ignored them unless having reason to suspect them when I worked in computer repair but somewhere within the last 5 to 10 years I saw some problems that became common enough that I had to teach fellow technicians that it was no longer the case and they needed consideration too. I had both sticks of RAM fail in the same intermittent way together before to where I assumed it had to be a different component causing the issues. Between defects and incompatibilities everywhere its best to make no assumptions when troubleshooting. Glad to hear things are now stable.
 
Everything has been running great!

Which brings me back to my original problem of trying to get actionable information from the zfs list USED and REFER columns.

Code: 
# zfs list -o space -r jail/wfp/sun
NAME               AVAIL   USED  USEDSNAP  USEDDS  USEDREFRESERV  USEDCHILD
jail/wfp/sun        590G  18.9G     16.1G   2.83G             0B      1.39M
jail/wfp/sun/5433   590G   756K        0B    756K             0B         0B
jail/wfp/sun/5434   590G   672K        0B    672K             0B         0B

I thought there was 50G in jail/wfp/sun. This was wrong. Confusion over the mountpoint. It really is 2.83G.

Between a compression ratio of 2.59 and LBA Size of 4096 I think I understand everything now. Thank you all for the help!
 
You must log in or register to reply here.
Back
Top