we are using Veeam Backup & Replication V9 since day of release. In the past few days we have an increasing amount of jobs that fail due to connections erros
(wrong user or bad password) or random soap errors during connecting on the Vsphere-Server. Some jobs failing 6 times and the 7th attempt is successfull and
backup job is running clean.
We opend a support case #0171666 and loading up a bunch of logfiles. So far we have no feedback with a workaround.
What I need is to establish a stable Veeam-Setup that grants stable backups without clicking the retry-button over and over. Is there a procedure to rollback to version 8 in case of that our problems can not solved in a short time range ?.
We are using Veeam Backup & Replication over many years and it is the first time that we have such problems and we are full satisfied with this product.
My question for the rollback to V8 is only to have an solution if fixing the problems will need more time
There's definitely no in-place rollback, but of course you always have a flexibility to deploy a new v8 instance.
That said, it is unlikely that your issues are caused by v9 (or by v9 only). It is safe to say that you do have some issues with your environment which need fixing. Because if the product was acting this unreliably in every environment - we would be slammed by this issue in support with 90000+ downloads since release. Just an example of real product issue: our infamous leap day bug that only lasted for 24 hours only affecting new installs resulted in a 2000+ views topic and over 50 support cases opened in a single day.
So, I highly recommend you work with our support to find the reason for such an unreliable behavior, as this is not normal nor reported by other users.
sitting here nearly the whole night I found out that you are right.
It is not V9 to blame for the Job issues.
I checked my VSphere-values and found following:
On one of our ESXi-Hosts the ntp-client hang up and runs out time syncronisation.
In the runtime values for the vcenter-server the server-Ip-adress going lost.... (who now why ? )
With correct values all Jobs running 100% as they did before.
)