How to view effectiveness of vdev

[jgreco]

I would see some evidence of it being used

To see evidence of it being used, you probably want to look at the actual disk I/O stats for the SLOG device. There's probably some ZFS metric somewhere but I tend to be a bit of a luddite and look at the low level proof where I can.

 

[awasb]

Code:

gstat -p

as a start would suffice.

Code:

arc_summary

for statistics:

Code:

ZIL committed transactions:
        Commit requests:                        
        Flushes to stable storage:    
        Transactions to SLOG storage pool:    
        Transactions to non-SLOG storage pool: 

MacOS-Clients do sync writes to SMB (as well as SMB signing). (And that's why e.g. TimeMachine setups with SMB-shares on TrueNAS as backup targets slow down to a crawl. Without SLOG, that is.)

 

[NugentS]

A transaction group is five seconds. But you have the transaction group that is currently being built (in RAM), and the transaction group that is being committed to the pool, which is two transaction groups. I seem to recall discussion that suggested that the block free and TRIM activity was only done after a full txg commit, so that's three transaction groups worth of storage. Am I forgetting something? It's too early and I'm not sufficiently caffeinated. I would expect a "handful" of txg's worth of data to be hanging around the SLOG. Someone will come in and correct me as to where I've erred in some technical gotcha but the overall point is correct.

Good job I allowed 20GB then