V7 - VMware Issue - False alerts

[carlos.luna]

We are also experiencing this.
See picture below:

I will be checking this post for some solution.
Carlos Luna

[Vitaliy S.]

Carlos, not sure there will be a solution form VMware for this problem in the short term, however we will workaround this VMware issue in the next update, that should be available this month.

[athome]

Hi!

I now see that Veeam ONE R2 has been released, but I dont see any notes about this issue in the release notes, has there been any fixes/workaround for this?

[Vitaliy S.]

Yes, we have implemented a workaround for this issue. Now all alerts with abnormally high values should be automatically suppressed. Do you see a different behavior?

[athome]

I haven't tested yet..

[Vitaliy S.]

Well...keep us updated!

[guillermo.lozano]

Hello, I have tested the patch released for this issue along with other fixes for two days now without any latency spikes on Veeam ONE or ESXi Servers.

I've received this patch through the support engineer from my case id N° 00470178.

[Vitaliy S.]

Hi Guillermo, thanks for confirming that everything worked well for you.

[ddayton]

Since I update One from V7 to V7 R2 I have not received any of the errors of read/write latency alarms that appeared to span days.

[Vitaliy S.]

Great, thanks!

[mberry]

We are having the same issue. Happens about 3-5 times every 24 hours. Does not seem to be caused by the backups as it occurs randomly throughout the day.
We are on Veeam ONE Monitor Version 7.0 (7.0.0.876)

[Vitaliy S.]

Please upgrade to our latest version to workaround this VMware issue > http://www.veeam.com/kb1841

[joshlott]

I know this post is old but I'm out of idea's.... Our VMWare environment is running 6.5. We are not using Veeam One however we are running B&R 9.5U2. We are experiencing the same issues of extreme datastore latency spikes. I have created tickets with VMWARE and Hardware vendors and everyone is at loss. Any ideas?

[Vitaliy S.]

My recommendation would be to keep working with VMware support team, as vSphere Client reports latency spikes regardless of you have Veeam ONE or not. Our workaround was to just suppress these false values reported from vSphere.

[ChrisGundry]

Hi. We are having the same issue as Josnlott, after upgrading from vSphere 5.5 to 6.5 we are now getting high latency showing in vSphere perfmon. Problem is VMware are not able to explain it.

9.5yrs latency is obviously an error in VMware but Veeam is alerting on it. It seems the supression of extreme/abnormal latency that was in place in older versions of VeeamOne is no longer in place in 9.5U2 as we have a lot of alerts from this.

Thanks

[Vitaliy S.]

The same checks are still there, but maybe vSphere now is generating these alerts via different methods thus our checks no longer work. Can you contact our team and have them review your log files?

[ChrisGundry]

OK, I have logged the case Vitaliy. Although for some reason my VeeamOne is not exporting logs at the moment, it gets stuck on 'saving 6/6' and won't save the logs! Grr

[stevehughes]

I'm interested in this topic and I'm surprised there isn't more chatter about it. I trialed Veeam One in our environment back in 2015 or 2016 and found the latency alerts to be a deal breaker for me, not because of Veeam but because ESXi was generating random huge values (days or weeks) which VeeamOne was reporting on. I really want reliable alerts on increased datastore latency but the alerts I was getting were meaningless because of this issue with ESX.

I have just upgraded one of my hosts to 6.7 and I note that it is still producing these nonsense latency spikes. I really want to give Veeam One another go though. Surely I'm not the only one seeing these spikes? Is there a current fix or workaround for this?

[Vitaliy S.]

The fix can only be produced if there are certain amount of cases opened for the issue. The previous issue has been fixed, but it seems like VMware has changed something in their APIs, which brought this issue back. I would recommend to contact our support team and ask for the fix. Meanwhile, I will ask our QC team to try to reproduce this issue.

[stevehughes]

Thanks Vitaliy. Will do. I’m puzzled by two things. (A) why more Veeam users aren’t talking about this, and (B) why VMware don’t just fix it after all this time.

[Vitaliy S.]

Good questions! Maybe there are some specifics for this configuration, as our dev team has confirmed that we are still suppressing these alarms starting from when this issue was first introduced.

[stevehughes]

I think Veeam One is actually suppressing these glitches. Sorry for the false alarm. I get a much more useful latency graph from Veeam One that I do from vSphere because you are filtering these out.

[Vitaliy S.]

Ok, thanks for the follow up.

[rossi123]

I'm interested in this topic and I'm surprised there isn't more chatter about it. I trialed Veeam One in our environment back in 2015 or 2016 and found the latency alerts to be a deal breaker for me, not because of Veeam but because ESXi was generating random huge values (days or weeks) which VeeamOne was reporting on. I really want reliable alerts on increased datastore latency but the alerts I was getting were meaningless because of this issue with ESX.

I have just upgraded one of my hosts to 6.7 and I note that it is still producing these nonsense latency spikes. I really want to give Veeam One another go though. Surely I'm not the only one seeing these spikes? Is there a current fix or workaround for this?

me too i am interested in this topic and i think all members if they are not interested in this topic i don't think they will be here

[Vitaliy S.]

We've suppressed these abnormal values in Veeam ONE, so you should not be seeing them with the latest version. However, if you see these spikes in the vSphere Client, then contacting the VMware support team together with our support team would be a recommended approach. Thanks!