- Posts: 181
- Liked: 24 times
- Joined: Aug 26, 2015 2:56 pm
- Full Name: Chris Gundry
We have a scenario where we failed over to a replica some time ago, then failed back. Many months later, someone created a snapshot on the server and forgot to remove it. A week later we have a massive snapshot on the VM, which was only found because I investigated a performance/latency issue on the VM. I think if we didn't have such good storage the snapshot would have remained hidden for a lot longer...
We have alarms in VeeamONE to trigger on snapshot age and snapshot size. Both alarms should have triggered, but they did not. Looking into this with support, the reason is because the VM has "VeeamReplicaSummary" within its VMX file, so is considered a replica server... The logic says that it somehow this data got into the source VM when it was replicated back from the DR site many months ago.
So I have two issues:
1. How did this happen? Why did the replica failback process put this info into the source server VMX file?
2. How do I stop it happening again? Presumably Veeam are not aware this can happen and will hopefully raise a change to fix this? In the mean time or in the event they don't fix it, how can I prevent this issue occuring again, by somehow monitoring the fact we have a 'replica' server where we shouldn't do.
I created a business view based on the conditions 'IsReplica' and ''Name does NOT contain _replica''. This indeed shows the VM in question, which had the snapshot issue.
The problem is, I am not sure what I can then do with the business view. I really need it to generate some kind of alert for me. How can I do that?
I could use the business view to generate a tag, then schedule a PowerShell script to check for said tag in VMware. But I don't like that idea because it has too many moving parts. Is there any way to have VeeamONE alert me that there is a VM in a business view or something?
- Veeam Software
- Posts: 1857
- Liked: 214 times
- Joined: Aug 07, 2018 3:11 pm
- Full Name: Fedor Maslov
Thanks for reporting that behavior.
Could you please share your support case ID with us so we will be able to have a look at the whole picture?
Users browsing this forum: No registered users and 5 guests