Host-based backup of VMware vSphere VMs.
Post Reply
d.bonini
Novice
Posts: 3
Liked: never
Joined: Nov 23, 2016 11:23 am
Full Name: Dario
Contact:

Network Flapping and very low performance

Post by d.bonini »

Case ID: 02308965
My conf:
Nutanix Cluster 3 node all SSD version 4.7.5.2
Esxi 6.0
Backup Server (dedicated) Hp DL160 G9 16 Gb ram - Win2016 - 10Gb Eth
Veeam 9.5 U2
Switch Supermicro SSE-X24S (all 10Gb) used only for Nutanix cluster, veeam server and repository QNAP
Targhet: Qnap TS-1231XU-RP 7 Disk Sata 6 TB raid 5 - 10Gb
Data size 7 TB
obviously being a physical machine the type of backup is NDB.
the option to install a vm proxy is not workable for customer specifications.

Hello,
my backups start at very high speeds at around 450 Mb / s.
after a few minutes of running (about 20-30minutes) the speed begins to drop rapidly until it reaches about 5-10 MB / s.
during the "slow" phase you can see a "flapping" on the Veeam server's network card with an alternate download and upload usage. you can also notice brief moments when no traffic jams are being made.

from what I was able to observe in the switch or in the Qnap (target) or in the Nutanix (source), they have usable values that can justify any kind of slowdown.

with such slow speed resolved it can not even run the first full backup.
d.bonini
Novice
Posts: 3
Liked: never
Joined: Nov 23, 2016 11:23 am
Full Name: Dario
Contact:

Re: Network Flapping and very low performance

Post by d.bonini »

following some test runs even with the support I have the strong doubt of being in this situation (if you need to specify the tests I have performed).
.
.
(https://www.veeam.com/kb2014​)
Q: Active full starts fast but then gradually becomes slower and slower. Target agent runs on Windows 2008 (not R2).
A: This is a known (yet undocumented) performance degradation issue on Windows 2008. Upgrade to or deploy 2008 R2 (or newer). In case of a CIFS repository you can force some other 2008 R2 server to be used as gateway/proxying. The issue itself can be demonstrated with diskspd if you set it to run for several hours, then observe how the write rate slowly goes down in Resource Monitor.

vmware-vsphere-f24/full-backup-speed-de ... 04-60.html

.
I tried hard but did not find documentation how to check this error / memory management bug.
on all Veeam forums the concept is "I've changed operating system and it works better"
since I seem to understand that for windows 2016 no one has yet reported similar problems as I can proceed to confirm or deny that problematic?
Post Reply

Who is online

Users browsing this forum: pdimarco and 103 guests