VMDK file grows and error message "There is no more space for the redo log"

[Michael1414]

Hi All, I created a VM several weeks ago, I have one back-up image I took when I got the VM online. I'm Getting the error "There is no more space for the redo log xxxxx-000001.vmdk." At the beginning of the new VM the VMDK file was at 66,560.00 KB. Now the same file is at 5,161,984.00 KB. That's a huge mark up since this is a test VM and its running a Cisco Call Manager application on it but, its in a test environment I don't have any users associated with the VM. I have the one snapshot which I could delete or revert to and probably get this up and running however, this is the second time this is happened and I'm assuming that this will happen in another couple of weeks after this file grows again.

Please any suggestions?

Thanks,

Mike

[Lightbulb]

Cant speak to why your snapshot grows at the rate that it does, but growing is what snapshots do. you should not use snapshots as long term backup. Snapshots should be used for short intervals to allow you a point in time recovery (After patching for example).

[Michael1414]

Hi Tom, is "vmdk" my snapshot file I thought extension "vmsn" is the snapshot file. I'm attaching a jpeg that shows the datastore files with the file that is growing highlighted.

thanks,

Mike

[athlon_crazy]

yes .vmsn is the snapshot file for the vm's

while xxxxx.vmdk is the changes from the last state you had taken the snapshot. Meaning, the size will grow the longer you keep the snapshot.

You could understand the snapshot better here : http://www.petri.co.il/virtual_vmware_snapshot.htm

System Engineer

Zen Systems Sdn Bhd

Malaysia

[AntonVZhbankov]

Michael, vmsn - is a memory snapshot, and vmdk have their snapshots too. Your vmdk snapshot can grow up to size of the original disk.

---

http://blog.vadmin.ru

[malaysiavm]

Here to explain how the snapshot behave in ESX. When you have a vm with 20GB, once you create a snapshot, it will start generate a seperate vmdk and virtual disk to keep the changes after the snapshot. If the changes after the snapshot you took is 10 GB, then the entire VM will consume 10 GB + 20 GB + swap and etc.

Although from OS level, your vm only show 20 GB, but it does consume 30 GB in this case. You should delete the snapshot from the snapshot manager to allow the system to commit the changes after the last snapshot and merge with the original virtual disk and vmdk. Once you deleted the snapshot, everything should back to normal.

Craig

Malaysia VMware Communities -

[steelburn]

I suggest you get your snapshots removed as soon as possible. If there is no space at all in the datastore during removal process, you would just be getting more issues.

As per earlier posts, VMware snapshots are never meant to be kept for a long duration. It's just for short term use, e.g: patching.

Say, if you are already at a point whereby removing the snapshot is no more possible for you (like the entire ESX stop responding, etc), and you have other spare datastore. Arrange for a downtime of the VM, clone it (so that you can get a fresh copy without snapshots), and run your new clone. Delete the original afterwards.

[karemabdou]

HI Craig

when i try to delet the snapshot the operation time out and but when i restat the esx services the snapshot manager deleted the snapshot but i can found it in the data store. any comment with that