a linux virtual machine hosted in vshere 4.1 not start and in the event i view this error
Message on linux: There is no more space for
might be able to continue this session by freeing
disk space on the relevant volume, and clicking
Retry. Click Cancel to terminate this session.
this machine is started when i have deleted every snapshot
now i have seen that the space used for virtual machine is at 50%
what can have caused this problem ?
This might be because you might have given thin provisioning for the linux disk and actually you physical disk might be full. So as the data increase in the VM might not got the space to write data. Hence by deleting the snapshot consuming the same size have free your space.
Looks like the snapshots of the virtual machine grew big occupying major space from the datastore. When you deleted the snapshot this made the snapshots get consolidated to the original disk. The original disk was able to consolidate both the snapshots to itself and growing to 50% space on the datastore
You can execute this script from the command line of the ESXi to check what is consuming the space.
clear; for each in $(ls) ; do du -hs "$each" ; done
Use snapshots sparingly. don't use snapshots as backup strategy. Always make sure that 20% of disk space is available in the VMFS volume.
Instead of moving the snapshots to some other Lun , You can add a new Lun and attach as an extent to this existing Lun Online this will increase the space.
Thanks for reply
the free space on datastore in less than 20%
the free space is decreased from the first intallation
can i recover free space ?
i have a second datastore
can i set snapshot location in this datastore ?
there is nothing like recovering disk space. You need to add a new Lun to this VMFS volumes. first you need to create a new Lun , then use the add extent feature after selecting the properties of the Lun you want to expand.
Setting a different datastore for snapshot is not easy. For this you have to sutdown the VM , move all the snapshot to the other VMFS volume , edit the .vmx file of the VM to point to the new location for the snapshot file and start the machine.
You may also need to edit the .vmdk descriptor file of the snapshot which is the child of the parent snapshot to point to the new location of the parent disk as you have moved all the snapshots in a different Lun so they will not be able to access the parent disks as they are not in the same location.
Before you are going to add an extent (what I don't recommend) consider whether:
Please provide a screenshot of the datastore browser window for this VM to see how the files (sizes, ...) currently look like. Also provide the current size of the datastore and the free disk space in GB (MB?).
Placing snapshots to different datastores, is always a risky, there is more change to get the corrouption and management also tough.
The easy way to reclaim is to either do a v2v or a cloning of the affected vm to another datastore and delete the old one
the edatastore is 300 gb size
there 2 virtual machine
1 size 90 gb
other size 120 gb ( in this virtual machine i have deleted snapshot for the problem described)
the free space of the datasore is 5 gb
Can you provide the screen shots of the datastore browser window please? This should show what's going on. If possible provide a screenshot for both VMs and ensure all details are shown (names, extensions, sizes, time stamps, ...).
With the 300GB datastore and only two VM's (120GB and 90GB) you should have more free disk space than the 5GB you mentioned. That's why I would like to see which files there are in the datastore to determine what can be dons.