Ciao!
I have another question: it happened twice that in a multi disk hot swap cabinet from Quanta was doing the following: ada0 offline ada0 online, immediately after (before resilvering finishes) ada1 offline ada1 online. Those two disks where on MIRRORING. I supposed weird hardware or cable failure but it turned out that the machine was still in testing and was making a super-cool noise when you arm the release leverage of the sata housings, so a TLC Phd student was having fun playing with the array (also to make sure ZFS was invincible.) All this happened during writes so part of the newst added data were on ada0 and another part on ada1.
After another scrub i was having chksum error, after clear and scrub same errors. No way to fix it automatically. I was able to mount the mirror in degraded mode: before only ada0 and after only ada1 and retreive all data. Is that the expected behavior? I was able to replicate it on another machine... but what would happen in the case of a raidz??
Cheers, Alessandro
I have another question: it happened twice that in a multi disk hot swap cabinet from Quanta was doing the following: ada0 offline ada0 online, immediately after (before resilvering finishes) ada1 offline ada1 online. Those two disks where on MIRRORING. I supposed weird hardware or cable failure but it turned out that the machine was still in testing and was making a super-cool noise when you arm the release leverage of the sata housings, so a TLC Phd student was having fun playing with the array (also to make sure ZFS was invincible.) All this happened during writes so part of the newst added data were on ada0 and another part on ada1.
After another scrub i was having chksum error, after clear and scrub same errors. No way to fix it automatically. I was able to mount the mirror in degraded mode: before only ada0 and after only ada1 and retreive all data. Is that the expected behavior? I was able to replicate it on another machine... but what would happen in the case of a raidz??
Cheers, Alessandro
