Real-time performance monitoring and troubleshooting
Post Reply
brichardson1991_obk
Expert
Posts: 114
Liked: 8 times
Joined: Aug 20, 2015 2:20 pm
Full Name: Ben Richardson
Contact:
Contact brichardson1991_obk

Disk Latency Alarms

Post by brichardson1991_obk »

We have the following enviroment

ESX03 - HP Proliant DL360p G8 - Xeon E5-2640 with 65Gb RAM - ESXi 5.1
ESX04 - Dell PowerEdge R630 - Xeon E5-2643 with 128Gb RAM - ESXi 5.5
ESX08 - Dell PowerEdge R630 - Xeon E5-2687W with 64Gb RAM - ESXi 6.0

SAN1
EMC VNXe 3100
Hosts with access: ESX03, 04, 08

SAN2
EMC VNXe 3200
Hosts with access:
ESX04

We are currently using the default alarms from Veeam one.

Can you recommend any changes to these as we are getting a high number of vm disk latency alarms?
Top
Shestakov
Veteran
Posts: 7328
Liked: 781 times
Joined: May 21, 2014 11:03 am
Full Name: Nikita Shestakov
Location: Prague
Contact:
Contact Shestakov

Re: Disk Latency Alarms

Post by Shestakov »

Hello Ben,

Just to make sure, are we talking about "VM total disk latency" alarm?

Have you checkd the performance in VeeamONE Monitor? Could you provide us a graph, please?
Thanks!
Top
brichardson1991_obk
Expert
Posts: 114
Liked: 8 times
Joined: Aug 20, 2015 2:20 pm
Full Name: Ben Richardson
Contact:
Contact brichardson1991_obk

Re: Disk Latency Alarms

Post by brichardson1991_obk »

Hi Nikita,

How would i produce such a graph from Veeam One
Top
Shestakov
Veteran
Posts: 7328
Liked: 781 times
Joined: May 21, 2014 11:03 am
Full Name: Nikita Shestakov
Location: Prague
Contact:
Contact Shestakov

Re: Disk Latency Alarms

Post by Shestakov »

It can be seen in the Infrastructure View.
Since we are talking about "VM total disk latency" alarm, which uses "Datastore Highest Latency" counter, you need to choose the datastore -> "Disk I/O" tab and pick latency charts there.
Top
Vitaliy S.
VP, Product Management
Posts: 27377
Liked: 2800 times
Joined: Mar 30, 2009 9:13 am
Full Name: Vitaliy Safarov
Contact:
Contact Vitaliy S.

Re: Disk Latency Alarms

Post by Vitaliy S. » 1 person likes this post

brichardson1991_obk wrote:Can you recommend any changes to these as we are getting a high number of vm disk latency alarms?
If you're observing high latency all the time and the average value is above the configured threshold, then you can adjust alarm thresholds to match the performance of your existing infrastructure.
Top
brichardson1991_obk
Expert
Posts: 114
Liked: 8 times
Joined: Aug 20, 2015 2:20 pm
Full Name: Ben Richardson
Contact:
Contact brichardson1991_obk

Re: Disk Latency Alarms

Post by brichardson1991_obk »

Please see below graph of past day Datastore IOPS

Image
Top
Shestakov
Veteran
Posts: 7328
Liked: 781 times
Joined: May 21, 2014 11:03 am
Full Name: Nikita Shestakov
Location: Prague
Contact:
Contact Shestakov

Re: Disk Latency Alarms

Post by Shestakov »

Well, Datastore IOPS will not tell much since we were talking about latency metrics.
I`d also review it for a several days or a week rather than a day.
Thanks!
Top
brichardson1991_obk
Expert
Posts: 114
Liked: 8 times
Joined: Aug 20, 2015 2:20 pm
Full Name: Ben Richardson
Contact:
Contact brichardson1991_obk

Re: Disk Latency Alarms

Post by brichardson1991_obk »

Apologies here is a week worth of Datastore latency

Image
Top
Shestakov
Veteran
Posts: 7328
Liked: 781 times
Joined: May 21, 2014 11:03 am
Full Name: Nikita Shestakov
Location: Prague
Contact:
Contact Shestakov

Re: Disk Latency Alarms

Post by Shestakov »

As it can be seen from the graph, there are pikes exceeding the defined thresholds, means VeeamONE works as expected.
If you don`t experience any remarkable performance issues because of the latency alarms, you can increase the thresholds for 20-40%.
What are the existing thresholds of the triggered alarms?

Here is a latency troubleshooting article. Thanks!
Top
Post Reply

Return to “Monitoring”



Who is online

Users browsing this forum: No registered users and 2 guests