pegglseder
Contributor
Contributor

missing /vmfs/volumes/<uuid> after loss of power

Hi,

after loss of power (behind USV/UPS) one Virtual Machine in our ESXi 5.1 is listed as "Unknown".

Connecting via SSH and "cat /etc/vmware/hostd/vmInventory.xml" displayed the missing VM with the following config:

<vmxCfgPath>/vmfs/volumes/51c02e59-d89b8808-c478-20cf30f0ce6c/W2K12R2/W2K12R2.vmx

This UUID 51c02e59-d89b8808-c478-20cf30f0ce6c does not exist in folder /vmfs/volumes/.

How can I restore this folder???

Thanks for any suggestions.

Peter

PS: I´m speaking better german than english

0 Kudos
4 Replies
msripada
Virtuoso
Virtuoso

This UUID 51c02e59-d89b8808-c478-20cf30f0ce6c refers to a datastore. Possibly the datastore is inaccessible to the ESXi host which is leading to the problem.

Can you check which datastore the VM was residing on is visible to the esxi host or any other esxi hosts in the cluster?

Thanks,

MS

0 Kudos
pegglseder
Contributor
Contributor

It seems that the datastore was on a iSCSI Software Adapter and its location is (was) on a separate NAS, which was damaged for a short time after the loss of power due to an overvoltage.

This NAS is mounted via NFS and a search for my VM files on the NAS with user root returned no results!

Should I worry about lots of warnings in the logs like:

vmkernel.log:

2018-07-04T13:58:48.094Z cpu7:4661)WARNING: iscsi_vmk: iscsivmk_StartConnection: Sess [ISID: 00023d000002 TARGET: iqn.2000-01.com.synology:NAS.VM1 TPGT: 0 TSIH: 0]

2018-07-04T14:01:28.768Z cpu6:6125)WARNING: CBT: 986: Unsupported ioctl 58

2018-07-05T12:59:37.124Z cpu10:6291)WARNING: LinScsi: SCSILinuxQueueCommand:1193:queuecommand failed with status = 0x1056 Unknown status vmhba39:0:0:0 (driver name: ahci) - Message repeated 5 times

2018-07-05T13:29:37.614Z cpu2:5227)WARNING: LinScsi: SCSILinuxQueueCommand:1193:queuecommand failed with status = 0x1056 Unknown status vmhba37:0:0:0 (driver name: ahci) - Message repeated 2 times

2018-07-06T00:01:02.345Z cpu9:4171)WARNING: VFAT: 4346: Failed to flush file times: Stale file handle

hostsd.log

2018-07-04T14:01:17.068Z [3BB3AB90 warning 'Hostsvc.StorageSystem'] HandleCoreDumpConfigurationIssue: Failed to remove host config issue: esx.problem.coredump.unconfigured

2018-07-04T14:01:32.426Z [3B6D5B90 warning 'Default' opID=FD44FBC3-0000003E] NicInfo: Neither array nor xdr is set.

2018-07-06T08:01:09.323Z [3BABAB90 error 'SoapAdapter.HTTPService'] HTTP Transaction failed on stream TCP(error:Transport endpoint is not connected) with error N7Vmacore15SystemExceptionE(Connection reset by peer)

Best

Peter

0 Kudos
continuum
Immortal
Immortal

> It seems that the datastore was on a iSCSI Software Adapter and its location is (was) on a separate NAS, which was damaged for a short time after the loss of power due to an overvoltage.

> This NAS is mounted via NFS and a search for my VM files on the NAS with user root returned no results!
If your datastore was using VMFS via iSCSI before then the NFS-server will not display the VMs when you use it via NFS.
Try to connect it via iSCSI again.

Do you need support with a recovery problem ? - send a message via skype "sanbarrow"
pegglseder
Contributor
Contributor

Many thanks to continuum​.

A short skype call with a good leading to check my configuration fixed the problem.

SCSI LUN was not correctly assigned on the NAS.

0 Kudos