I have one host in our 23 host cluster that will not recognize a new set of HP Lefthand related LUNs. I have tried to reset the iSCSI initiator via disable/enable, but the same IPs remain even though the shares have been removed from the host at the HP Lefthand level. I get this error message when I try to remove either of the Dynamic IP addresses:
"Error during the configuration of the host:Remove Send target failure on HBA vmhb33, Target 10.10.11.200:3260 - iScsiLibException: status(80000000): unknown error; Message= IMA_RemoveDiscoveryAddress"
Aside from rebuilding the entire host, anyone have an ideas?
Rebuilding the ESX host should be the last option, since it might create more issues. Try connecting to the ESX host directly using VI Client and then try configuring iSCSI again since this would eliminate some locks. You may also try restarting the management service # service mgmt-vmware restart let us know if this helps.
I would recommend that you follow the KB below religiously and Im sure
it will help you resolve the issue.
Troubleshooting iSCSI array connectivity
Troubleshooting ESX and ESXi connectivity to iSCSI arrays using hardware
Troubleshooting ESX Server and ESXi connectivity to iSCSI arrays using
I would recommend that you save these link for future use.
Hope this information helps
If you found this or other information useful, please consider awarding
points for "Correct" or "Helpful".
Thanks Andre', I have restarted the host several times and remove and reinstalled it into the vcenter to no avail. There is nothing in the logs that explains the error message I still receive.
Any progress on this? I have the same problem.
I'm unable to add or remove targets, and rebooting, disabling/enable iscsi service, adding or removing via cli, nothings seems to help.
Is there another way to totally scratch the iscsi configuration on a host? I suspect the host has a static discovery to a target that doesn't exists, but all my other esx servers did not encounter this problem.
Not sure if this will help with your specific problem, but...
You can clear down the iscsi config by doing the following:
1. Disable the software iscsi initiator
2. On the ESX(i) host rename the all the files in the folder /etc/vmware/vmkiscsid - there should be 3 files in here (initiatorname.iscsi, iscsid.conf and vmkiscsid.db) - you could just delete them but i'd recommend renaming first...
3. Reboot the host
4. Once rebooted, enable the software iscsi initiator and reconfigure as required.
I've tested this in my lab and it works fine, but i'd advise caution in a live environment... it's most like completely unsupported by VMware, so may be best to use as a last resort before re-building... 😃
Hope this helps.
Hi and thanks!
That worked like a charm. When the server rebooted after moving the files to a temp folder it came up without targets and with a default initiator id. Great
Thanks a lot again!
We had a similar scenario when one of our ESX4 cluster nodes had a memory problem. We couldn't reconfigure the iSCSI Software Adapter, it somehow had lost the default initiator name - was set to "iSCSI Software Adapter" instead of original name "vmhba35", and wouldn't allow us to manually set the WWN for the initiator - received "Object reference not set to an instance of an object" error message.
After performing the steps outlined by WaffleSniffer the default vmhba name reappeared and allowed us to manually change the WWN back to the original. After a quick rescan it found all SAN LUNs and we were back in business!
Hope this helps someone else (and thanks WaffleSniffer)!!!
Oh, the error we were seeing was similar, which is how I came upon this post to begin with.
Error accessing iscsi interface: iScsiLibException: status(80000000): unknown error; Message= IMA_VS_GetDiscoveryStatus