Maybe someone can suggest how to trace the issue I'm having.
I'm setting up a clean install of ESXi 6.5, backed by a "known good" external iSCSI store (which I can connect to using the MS Windows initiator, for example). The iSCSI store hasn't actually been used, it's not yet even been formatted, so there's no prior GPT or anything. The install itself is "clean", barring some basic config (IP addresses, NIC VIB added, MTU 9000 throughout, etc)
The ESXi host has a dedicated NIC connecting it to the file server for iSCSI use. The entire network uses Jumbo frames, and pinging from console --size 8972 works. When I login to the host using its internal web client in maintenance mode and enable iSCSI, then adapters -> iSCSI adapter also detects and lists the iSCSI targets correctly. In console:
esxcli network diag ping --df --interface vmk1 --ipv4 --size 8972 --host 10.2.0.1
returns 3 successes (and using --size 8973 fails "too long" exactly as one would expect)
The only log that seems to have iSCSI-related content is syslog.log, but that doesn't show any obvious errors:
2017-04-01T19:07:04Z iscsid: Login Target: iqn.2005-10.local.svr:esxi-target-readonly if=iscsi_vmk@vmk1 addr=10.2.0.1:3260 (TPGT:-1 ISID:0x1)
2017-04-01T19:07:04Z iscsid: Notice: Assigned (H64 T1 C0 session=2, target=2/2)
2017-04-01T19:07:04Z iscsid: Notice: Setting NODELACK for target=iqn.2005-10.local.svr:esxi-target-readonly (host=10.2.0.1)
2017-04-01T19:07:04Z iscsid: DISCOVERY: transport_name=iscsi_vmk Pending=2 Failed=0
2017-04-01T19:07:04Z iscsid: get_op_params_text_keys: Update target portal group tag, oldTpgt=-1, newTpgt=257 targetName=iqn.2005-10.local.svr:esxi-target
2017-04-01T19:07:04Z iscsid: Login Success: iqn.2005-10.local.svr:esxi-target if=iscsi_vmk@vmk1 addr=10.2.0.1:3260 (TPGT:257 ISID:0x1)
2017-04-01T19:07:04Z iscsid: connection 1:0 (iqn.2005-10.local.svr:esxi-target if=iscsi_vmk@vmk1 addr=10.2.0.1:3260 (TPGT:257 ISID:0x1) (T0 C0)) is operational
2017-04-01T19:07:04Z iscsid: get_op_params_text_keys: Update target portal group tag, oldTpgt=-1, newTpgt=257 targetName=iqn.2005-10.local.svr:esxi-target-readonly
2017-04-01T19:07:04Z iscsid: Login Success: iqn.2005-10.local.svr:esxi-target-readonly if=iscsi_vmk@vmk1 addr=10.2.0.1:3260 (TPGT:257 ISID:0x1)
2017-04-01T19:07:04Z iscsid: connection 2:0 (iqn.2005-10.local.svr:esxi-target-readonly if=iscsi_vmk@vmk1 addr=10.2.0.1:3260 (TPGT:257 ISID:0x1) (T1 C0)) is operational
2017-04-01T19:07:05Z iscsid: DISCOVERY: transport_name=iscsi_vmk Pending=0 Failed=0
2017-04-01T19:07:15Z iscsid: discovery_sendtargets::Running discovery on IFACE default(iscsi_vmk) target addr=10.2.0.1:3260 (drec.transport=iscsi_vmk)
2017-04-01T19:07:15Z iscsid: discovery_sendtargets::Completed discovery on IFACE default(iscsi_vmk) target addr=10.2.0.1:3260 transport=iscsi_vmk UniqueTgt=2 DuplicateTgt=0
2017-04-01T19:07:15Z iscsid: discovery_sendtargets::Running discovery on IFACE iscsi_vmk@vmk1(iscsi_vmk) target addr=10.2.0.1:3260 (drec.transport=iscsi_vmk)
2017-04-01T19:07:15Z iscsid: discovery_sendtargets::Completed discovery on IFACE iscsi_vmk@vmk1(iscsi_vmk) target addr=10.2.0.1:3260 transport=iscsi_vmk UniqueTgt=0 DuplicateTgt=2
2017-04-01T19:07:15Z iscsid: Login Target Skipped: iqn.2005-10.local.svr:esxi-target if=iscsi_vmk@vmk1 addr=10.2.0.1:3260 (TPGT:257 ISID:0x2) (Already Running)
2017-04-01T19:07:15Z iscsid: Login Target Skipped: iqn.2005-10.local.svr:esxi-target-readonly if=iscsi_vmk@vmk1 addr=10.2.0.1:3260 (TPGT:257 ISID:0x2) (Already Running)
2017-04-01T19:07:15Z iscsid: DISCOVERY: transport_name=iscsi_vmk Pending=0 Failed=0
So on the surface it should work. But when I try to format the storage as a VMFS datastore, it's not showing up under storage, or in the "new datastore" dialog.
How can I trace what's going on, and get this store usable?
