Closing on lack of feedback.

@Rob Evers The messages you quoted and the ones I also see in the provided debug only say that ZFS can’t commit the transaction group during two minutes. It is obviously too long, since normally it should happen every few seconds. Blocked TXG sync thread will finally mean that some other ZFS operations beside of writes will be blocked too. Unfortunately they alone do not explain the cause. You’d have to look on the performance of your SSDs. For example, they are desktop-grade and may have quite long cache flush time, and one of the messages I see actually speaks about too long ZIL waiting (that may be a cause or another victim), that I guess means some of your VMs tries to send sync write or cache flush requests, that probably doesn’t help. You may try to set sync=disabled for some of the VM ZVOLs to see whether it help. It may cause VM state to roll back few seconds in case of system crash, but may prove the performance.

Hello @Rob Evers , thank you for supplying the requested information. This issue has been forwarded to an appropriate iXsystems Engineer, who will let you know if any further information is needed.

Regards,

Thank you for submitting this TrueNAS Bug Report! So that we can quickly investigate your issue, please attach a Debug file and any other information related to this issue through our secure and private upload service below. Debug files can be generated in the UI by navigating to System -> Advanced -> Save Debug.

https://ixsystems.atlassian.net/servicedesk/customer/portal/15/group/37/create/153