The Hobbyist
Cadet
- Joined
- Jun 19, 2017
- Messages
- 9
Hi.
I have googled and searched forums, I need to get a second opinion on a situation I am troubleshooting:
1) I have a raid z2 setup with 2 spares. Truenas Core is in a vm on Qemu with hardware passthrough. I am skipping posting hardware specs as my question is not related to that but to procedure. I have plenty of backups and a replicated copy on another server, so while I would like to avoid data loss it won't be the end of the world.
2) One drive failed, I replaced it and resilvering started but it never gets above 20% complete before it hangs the Truenas Core vm and I have to restart it. I have no direct solution to the issue, I cannot keep the Truenas vm active long enough to find out which of the other drives might be causing connection or other issues.
3) WHAT I WOULD LIKE TO DO: Image my Truenas Core to an ssd and boot directly into a bare metal install, import the ZFS pool and see if it will complete the resilver. I need to take the virtualization layer out of the equation to complete the troubleshooting.
4) QUESTION: If I were to import a zfs raid z2 array into a bare metal Truenas Core install, would the resilver continue? If yes, how does this work, is the resilver triggered by metadata kept around by zfs? Or is there a config setting / file that I have to make sure is around when I do the import? If this is not recommended, is it not recommended or is it an absolute DO NOT DO because the resilver will not be triggered and the array will be permanently damaged?
I have the array shut off for now, there is no point in leaving the Truenas vm trying to resilver and eventually hang.
Any feedback is appreciated, thanks.
Robert
I have googled and searched forums, I need to get a second opinion on a situation I am troubleshooting:
1) I have a raid z2 setup with 2 spares. Truenas Core is in a vm on Qemu with hardware passthrough. I am skipping posting hardware specs as my question is not related to that but to procedure. I have plenty of backups and a replicated copy on another server, so while I would like to avoid data loss it won't be the end of the world.
2) One drive failed, I replaced it and resilvering started but it never gets above 20% complete before it hangs the Truenas Core vm and I have to restart it. I have no direct solution to the issue, I cannot keep the Truenas vm active long enough to find out which of the other drives might be causing connection or other issues.
3) WHAT I WOULD LIKE TO DO: Image my Truenas Core to an ssd and boot directly into a bare metal install, import the ZFS pool and see if it will complete the resilver. I need to take the virtualization layer out of the equation to complete the troubleshooting.
4) QUESTION: If I were to import a zfs raid z2 array into a bare metal Truenas Core install, would the resilver continue? If yes, how does this work, is the resilver triggered by metadata kept around by zfs? Or is there a config setting / file that I have to make sure is around when I do the import? If this is not recommended, is it not recommended or is it an absolute DO NOT DO because the resilver will not be triggered and the array will be permanently damaged?
I have the array shut off for now, there is no point in leaving the Truenas vm trying to resilver and eventually hang.
Any feedback is appreciated, thanks.
Robert
