S3 Server stopped working after LIF failover

bnies — Tue, 24 Mar 2026 10:41:23 GMT

Hi,

We have an old Netapp FAS2650 running with ONTAP 9.11.1P20 (yes, it is out of support) and a S3 object server running there for testing. It was running fine for couple of weeks until I noticed today that the LIF is not at home node. After reverting the LIF to home node, S3 service stopped working. Nothing is listening on port 443 anymore. I tried to disable/enable object-store-server, stop/start vserver, migrate LIF back to other node, adjust service-policy, all with no luck.

S3 object-store-server

mycluster::> object-store-server show (vserver object-store-server show) Vserver: mycluster-vs1 Object Store Server Name: mycluster-vs1.example.com Administrative State: up HTTP Enabled: false Listener Port For HTTP: 80 HTTPS Enabled: true Secure Listener Port For HTTPS: 443 Certificate for HTTPS Connections: mycluster-vs1.example.com Comment: Netapp S3 playground

Interfaces and service-policy

mycluster::> network interface show -vserver mycluster-vs1 Logical Status Network Current Current Is Vserver Interface Admin/Oper Address/Mask Node Port Home ----------- ---------- ---------- ------------------ ------------- ------- ---- mycluster-vs1 vs1data1 up/up 10.0.77.88/24 mycluster-01 a0a-123 true vs1data2 up/up 10.0.88.99/24 mycluster-01 a0a-456 true 2 entries were displayed. mycluster::> network interface show -vserver mycluster-vs1 -lif * -fields service-policy,services vserver lif service-policy services ---------- -------- ------------------ ----------------------------------------------------------------------------------------------------------------------------------------------------------------------- mycluster-vs1 vs1data1 default-data-files data-core,data-nfs,data-cifs,data-flexcache,data-fpolicy-client,management-dns-client,management-ad-client,management-ldap-client,management-nis-client,data-dns-server mycluster-vs1 vs1data2 my-s3-service data-core,data-s3-server 2 entries were displayed.

No TCP/s3 service listening on LIF vs1data2 port 443.

mycluster::> network connections listening show -vserver mycluster-vs1 Vserver Name Interface Name:Local Port Protocol/Service ---------------- ------------------------------------- ----------------------- Node: mycluster-05 mycluster-vs1 vs1data1:4049 UDP/rquota mycluster-vs1 vs1data1:2050 TCP/fcache mycluster-vs1 vs1data1:111 TCP/port-map mycluster-vs1 vs1data1:111 UDP/port-map mycluster-vs1 vs1data1:4046 TCP/sm mycluster-vs1 vs1data1:4046 UDP/sm mycluster-vs1 vs1data1:4045 TCP/nlm-v4 mycluster-vs1 vs1data1:4045 UDP/nlm-v4 mycluster-vs1 vs1data1:2049 TCP/nfs mycluster-vs1 vs1data1:2049 UDP/nfs mycluster-vs1 vs1data1:635 TCP/mount mycluster-vs1 vs1data1:635 UDP/mount 12 entries were displayed.

I presume it's a bug in this old ONTAP code. Any idea how to revive that without destroying the buckets and users?

topic S3 Server stopped working after LIF failover in Object Storage

S3 Server stopped working after LIF failover