guermantes
Patron
- Joined
- Sep 27, 2017
- Messages
- 213
Hi,
I did not find anything on this subject in the documentation so I turn to the forum. Will I be shooting myself in the foot if I unmount a secondary pool each day? Will Freenas complain about it not being available or something or might it even affect my main pool which is up and running all the time? Will Freenas perhaps try to mount the pool again repeatedly? (I know there is an issue with not being able to unmount a pool while it is active, but I was thinking I could try my luck and see if this becomes a problem.)
This below is why I am asking:
I have a single drive constituting a pool of its own (bad English but I think you see what I mean). The purpose of this drive is to accept zfs send/recv issued by a bash script that runs every night, as a local backup in addition to my offsite backup (which can only be reached by taking the bus and fetching the drive in person). The drive in question is a WD Gold 7200 rpm and as such brings a bit more heat to the overall system, and runs 3-4 degrees C higher than the six WD Reds that constitute my main pool. The Gold (by sales pitch) should survive longer than my Reds if spinning all the time, but my concern is with the temperatures. I had found a sweet spot as regards fan noise and drive temps, but the Gold thwarts this a tad, and come summertime the WD Gold risks hitting 40 C even before the peak of summer. I would like to stay at the sweet spot if I can.
So I figured I could let the bash script unmount the drive after each successful backup and let the drive and rest before the script mounts it 23 hours later before zfs sending occurs again. My thinking is that the integrity of the data is checked by zfs in regards to my main pool where zfs is allowed to "do it's thing" all the time. I am also thinking that any possible corruption on the receiving drive would be alerted to me by error messages from zfs because if that happens incremental sending does not succeed due to said corruption. In this scenario the Gold drive would still be allowed to for be running when scrubs and SMART tests are to be performed.
Are there reasons why I should stop thinking about this 'solution'?
I did not find anything on this subject in the documentation so I turn to the forum. Will I be shooting myself in the foot if I unmount a secondary pool each day? Will Freenas complain about it not being available or something or might it even affect my main pool which is up and running all the time? Will Freenas perhaps try to mount the pool again repeatedly? (I know there is an issue with not being able to unmount a pool while it is active, but I was thinking I could try my luck and see if this becomes a problem.)
This below is why I am asking:
I have a single drive constituting a pool of its own (bad English but I think you see what I mean). The purpose of this drive is to accept zfs send/recv issued by a bash script that runs every night, as a local backup in addition to my offsite backup (which can only be reached by taking the bus and fetching the drive in person). The drive in question is a WD Gold 7200 rpm and as such brings a bit more heat to the overall system, and runs 3-4 degrees C higher than the six WD Reds that constitute my main pool. The Gold (by sales pitch) should survive longer than my Reds if spinning all the time, but my concern is with the temperatures. I had found a sweet spot as regards fan noise and drive temps, but the Gold thwarts this a tad, and come summertime the WD Gold risks hitting 40 C even before the peak of summer. I would like to stay at the sweet spot if I can.
So I figured I could let the bash script unmount the drive after each successful backup and let the drive and rest before the script mounts it 23 hours later before zfs sending occurs again. My thinking is that the integrity of the data is checked by zfs in regards to my main pool where zfs is allowed to "do it's thing" all the time. I am also thinking that any possible corruption on the receiving drive would be alerted to me by error messages from zfs because if that happens incremental sending does not succeed due to said corruption. In this scenario the Gold drive would still be allowed to for be running when scrubs and SMART tests are to be performed.
Are there reasons why I should stop thinking about this 'solution'?