Irrelevant if mgmt sends (TCP) disconnect to the SSVM (which shoul not
affect regular OS pings from mgmt to SSVM) - if you have issues pinging
from mgmt to hypervisor node - than again I would bet on misbehaving switch
equipment.
On Thu, 25 Jul 2019 at 18:52, Rakesh v wrote:
> True, but I thoug
True, but I thought since mgt sends disconnect to ssvm, I thought it will reset
the interface on which is connecting
Sent from my iPhone
> On 25-Jul-2019, at 4:57 PM, Andrija Panic wrote:
>
> In your previous mail, I understond that you used the OS tool "ping" and
> NOT refering to internal AC
In your previous mail, I understond that you used the OS tool "ping" and
NOT refering to internal ACS pings?
"Out of all these, the ping drops were observed from MGT server to ssvm and
mgt server to nodes. Basically all nodes lost connection. Then it recovered
itself after 1 minute."
On Thu, 25 J
The ping between mgt server and ssvm fails because mgt sends disconnect message
to all nodes. If you look at the logs I pasted in first email, the mgt server
thinks ssvm is lagging behind on ping and sends a disconnect message without
investigation for all nodes. Also it happens at the beginning
since basic network connectivity (ping failures) was down between mgmts and
nodes (and SSVM on it) - I would point my finger to your networking
equipment - i.e. I expect zero problems with ACS (since pings fail).
Let us know how it goes.
Andrija
On Thu, 25 Jul 2019 at 16:04, Rakesh v wrote:
>
Yes I was monitoring it continuously. Below are the steps which I was doing
when issue happened
1. Ping from MGT server to ssvm
2. Ping from ssvm to secondary storage ip
3. Ping from ssvm to public IP like 8.8.8.8
4. Ping from MGT server to node in which ssvm was running
Out of all these, the
Can you observe the status of SSVM (is it UP/Connecting/Disconnected/Down)
while you have issues?
I would advise checking your Secondary Storage itself - and also running
the SSVM diagnose script /usr/local/cloud/systemvm/ssvm-check.sh - observe
if any errors with NFS or others.
Lastly - and don
Yes I have set the ip's of the three MGT servers in the "host" field
Sent from my iPhone
> On 25-Jul-2019, at 2:14 PM, Pierre-Luc Dion wrote:
>
> Do you have a load balancer in front of cloudstack? Did you set the global
> settings "host" to the ip of the mgmt server?
>
>
> Le jeu. 25 juill.
Do you have a load balancer in front of cloudstack? Did you set the global
settings "host" to the ip of the mgmt server?
Le jeu. 25 juill. 2019 03 h 24, Rakesh Venkatesh
a écrit :
> Hello People
>
>
> I have a strange issue where mgt server times out to send a command to
> secondary storage VM
