Delay in response time on vms

[Nisamudeen]

We are runnning high production stack environement for our application. We have a network of 5 dedicated servers and some of our meachines are using only 1% of server resource. Hence we have planned to move low resource consuming servers to vms for budget cut down. We had one haproxy, two api servers, two database servers.

We have installed and configured proxmox node for the purpose. We have setup two KVM vms for haproxy and redis server. We have tested haproxy yesterday. But we could see response three times bigger while compared to dedicated server. Meanwhile the server load was found to be very low.

our actual response time was below 8 seconds in dedicated server but after using haproxy new vm it went above 30 seconds. We suspect if there is something related to network bridge ?

 

[Nisamudeen]

root@prox:~# cat /etc/network/interfaces
# network interface settings; autogenerated
# Please do NOT modify this file directly, unless you know what
# you're doing.
#
# If you want to manage part of the network configuration manually,
# please utilize the 'source' or 'source-directory' directives to do
# so.
# PVE will preserve these directives, but will NOT its network
# configuration from sourced files, so do not attempt to move any of
# the PVE managed interfaces into external files!
auto lo
iface lo inet loopback
iface eno1 inet manual
iface eno2 inet manual
iface eno3 inet manual
iface eno4 inet manual
auto vmbr0
iface vmbr0 inet static
address ###publiv-ip ###
netmask ********
gateway *******
bridge_ports eno1
bridge_stp off
bridge_fd 0
auto vmbr1
iface vmbr1 inet static
address 172.16.2.15
netmask 255.255.255.0
bridge_ports eno2
bridge_stp off
bridge_fd 0
#local

 

[guletz]

I am using some setup like you (bridge, haproxy) but with zfs. I do not have this problem... lucky me . I would suggest to test directly your haproxy backends, and to see how is your latency. Maybe is a missaligned lvm or storage?

 

[manu]

I don't think this is related to the network bridges or even virtualization
for a timeout like this I would check if some DNS / reverse DNS is not working

you should start HA proxy in debug mode to see if this is really the time processing the request wich takes 30s

 

[byshivam]

I don't think this is related to the network bridges or even virtualization
for a timeout like this I would check if some DNS / reverse DNS is not working

you should start HA proxy in debug mode to see if this is really the time processing the request wich takes 30s

Its Working , When i change DNS , default host to Google DNS and CloudFlare, response time is suddenly increased. Thanks!