Q: 9
An administrator recently deployed a new three-node VMware vSAN Express Storage Architecture
(ESA) cluster to an existing workload domain. After creating a number of Virtual Machines (VMs), the
administrator discovers that storage is being consumed a lot quicker than expected.
While investigating the issue, the administrator discovers that the datastore default policy has been
set to RAID-1 by Auto-Policy Management rather than the expected RAID-5.
What is a possible cause?
Options
Discussion
B. since RAID-5 on vSAN ESA needs at least 4 hosts. Seen similar in practice exams.
B imo. RAID-5 needs at least four hosts in vSAN ESA for erasure coding to be available. Since this is a three-node cluster, Auto-Policy Management drops back to RAID-1 by default, which eats up storage way faster. Pretty sure that's it, but open if anyone else sees something I'm missing.
C looks like a possible cause since Force Provisioning can let you bypass some requirements, but with ESA clusters it's tricky. In practice, official guide and hands-on labs suggest host count is the main factor for RAID-5 eligibility but I could see C confusing some. Thoughts?
Why would D be the cause here if three-node ESA clusters don't even allow RAID-5?
D imo. If Host Rebuild Reserved is enabled, the cluster reserves capacity for rebuilds which can skew storage usage stats and might trick you into thinking storage is getting eaten up. Pretty sure B is the common cause, but if they've tweaked cluster reserves this could explain it too. Anyone disagree?
Gotta be B, since a 3-node ESA cluster doesn't meet the minimum for RAID-5. Anyone think C makes sense here?
Maybe B, but I wonder if some policies could override the default if admin forced it. Just not sure if three hosts would ever let you pick RAID-5 even with override. Open to being corrected here.
A is wrong, B. You need at least four hosts for RAID-5 on ESA clusters, so a three-node setup can't apply that policy. Pretty common gotcha if you rush through. Unless VMware changed something lately, that's still the rule.
C isnt right here, B is. With only three nodes in ESA, you cant use RAID-5, so the cluster defaults to RAID-1. Option A is a bit misleading since RAID-5 does work on ESA, just needs 4 hosts. Let me know if theres another angle.
Not A, definitely B. RAID-5 in vSAN ESA requires at least four hosts, so auto-management drops to RAID-1 if there are only three. Seen this in similar questions, easy to miss if you forget the host requirement. Anyone think D is close? I’m pretty sure it’s a trick option.
Be respectful. No spam.
