StorageCurious
Explorer
- Joined
- Sep 28, 2022
- Messages
- 60
Using TrueNAS-13.0-U2
I have a pretty sweet rig running with two separate SAS3008 cards (same firmware, 16.00.12), running each a SAS cables, each to a separate controller on a Dell MD1400. This is all running over SAS3, but the disks themselves are SAS2.
When I booted up, I created a 5 wide strip of mirrors (i.e. RAID10-style) pool made of 10 "multipathing disks' (i.e. multipath/disk1, etc). and a hot spare, also "multipathed"
Then, for SCIENCE!, I started transferring files to it. In the middle of the transfer, I pulled one SAS cable. Then I reconnected it. Pulled the other cable. Reconnected it. At any point there was at least one SAS cable plugged in properly.
The transfer seems to have happened properly. BUT....
Went back to the UI. The current pool status (see screenshot at the bottom) is confusing me, and I'm not sure if it's expected to act like this.
Questions:
1) Shouldn't SAS multipathing have prevented these 2 disks (disk3 and disk9) having FAULTED states from a cable pull? I thought it was the entire point of having multipathing.
2) Why is disk11 both part of spare at the bottom, and part of a MIRROR/SPARE sub category (2nd mirror from the top) ? nothing seems to be resilvering as far as I know.
3) How do I bring this all back to "healthy" instead of degraded? All disks was healthy before and I imagine pulling SAS cables from the enclosures hasn't damaged them in any way.
4) Is multipathing in TrueNAS a marginal/not reliable affair and I should just forget about using it?
Note this is all SAS, no SATA involved in this multipathing.
I have a pretty sweet rig running with two separate SAS3008 cards (same firmware, 16.00.12), running each a SAS cables, each to a separate controller on a Dell MD1400. This is all running over SAS3, but the disks themselves are SAS2.
When I booted up, I created a 5 wide strip of mirrors (i.e. RAID10-style) pool made of 10 "multipathing disks' (i.e. multipath/disk1, etc). and a hot spare, also "multipathed"
Then, for SCIENCE!, I started transferring files to it. In the middle of the transfer, I pulled one SAS cable. Then I reconnected it. Pulled the other cable. Reconnected it. At any point there was at least one SAS cable plugged in properly.
The transfer seems to have happened properly. BUT....
Went back to the UI. The current pool status (see screenshot at the bottom) is confusing me, and I'm not sure if it's expected to act like this.
Questions:
1) Shouldn't SAS multipathing have prevented these 2 disks (disk3 and disk9) having FAULTED states from a cable pull? I thought it was the entire point of having multipathing.
2) Why is disk11 both part of spare at the bottom, and part of a MIRROR/SPARE sub category (2nd mirror from the top) ? nothing seems to be resilvering as far as I know.
3) How do I bring this all back to "healthy" instead of degraded? All disks was healthy before and I imagine pulling SAS cables from the enclosures hasn't damaged them in any way.
4) Is multipathing in TrueNAS a marginal/not reliable affair and I should just forget about using it?
Note this is all SAS, no SATA involved in this multipathing.
