stifler
Contributor
Contributor

ESXi mounts datastore directly after umounting

So I'm kinda stuck in here, while trying to unmount a datastore via commandline that is presented through a VPLEX:

/var/log # esxcfg-scsidevs -m|grep naa.600014400000001070149ac10bcf3d46

naa.600014400000001070149ac10bcf3d46:1                           /vmfs/devices/disks/naa.600014400000001070149ac10bcf3d46:1 5481c98b-4912a697-0833-0017a477502c  0  VPLXTST01

/var/log # esxcli storage filesystem unmount -u 5481c98b-4912a697-0833-0017a477502c

/var/log # esxcfg-scsidevs -m|grep naa.600014400000001070149ac10bcf3d46

naa.600014400000001070149ac10bcf3d46:1                           /vmfs/devices/disks/naa.600014400000001070149ac10bcf3d46:1 5481c98b-4912a697-0833-0017a477502c  0  VPLXTST01

/var/log # tail vmkernel.log

2014-12-19T10:56:30.013Z cpu1:28988)LVM: 8329:   on-disk disk ID: <type 2, len 22, lun 181, devType 0, scsi 0, h(id) 2489811071063174045>

2014-12-19T10:56:30.426Z cpu1:28988)LVM: 8315: Device naa.600014400000001070149ac10bcf3d6e:1 detected to be a snapshot:

2014-12-19T10:56:30.426Z cpu1:28988)LVM: 8322:   queried disk ID: <type 2, len 22, lun 183, devType 0, scsi 0, h(id) 1622136697610455165>

2014-12-19T10:56:30.426Z cpu1:28988)LVM: 8329:   on-disk disk ID: <type 2, len 22, lun 183, devType 0, scsi 0, h(id) 816211612732412737>

2014-12-19T10:56:30.430Z cpu1:28988)LVM: 3180: Device naa.600014400000001070149ac10bcf3d6e:1 is detected as being in volume 5481c9b0-f6756c64-2959-0017a477502c (0x412304f1b658)

2014-12-19T10:56:30.845Z cpu1:28988)Vol3: 705: Couldn't read volume header from control: Not supported

2014-12-19T10:56:30.845Z cpu1:28988)Vol3: 705: Couldn't read volume header from control: Not supported

2014-12-19T10:56:30.845Z cpu1:28988)FSS: 4972: No FS driver claimed device 'control': Not supported

2014-12-19T10:56:31.224Z cpu1:28988)VC: 1591: Device rescan time 1173 msec (total number of devices 53)

2014-12-19T10:56:31.224Z cpu1:28988)VC: 1594: Filesystem probe time 185 msec (devices probed 11 of 53)

It happens that the datastore is immediatly mounted again. Only thing that keeps me wondering is the message in the vmkernel.log stating that the volume is detected to be snapshot.

I have also tried to set the device to state=off and then detach it, but then it also mounts over again:

/var/log # esxcli storage core device world list -d naa.600014400000001070149ac10bcf3d46

/var/log # esxcli storage core device set -d naa.600014400000001070149ac10bcf3d46 --state=off

/var/log # esxcli storage core device detached list

Device UID                            State

------------------------------------  -----

naa.600014400000001070149ac10bcf3d46  off 

/var/log # esxcli storage core device detached remove -d naa.600014400000001070149ac10bcf3d46

/var/log # esxcfg-scsidevs -m|grep naa.600014400000001070149ac10bcf3d46

naa.600014400000001070149ac10bcf3d46:1                           /vmfs/devices/disks/naa.600014400000001070149ac10bcf3d46:1 5481c98b-4912a697-0833-0017a477502c  0  VPLXTST01

So right now ran out if ideas...

Thanks

Mathias

0 Kudos
0 Replies