Agent cannot connect to the server

[Guest]

On 29-11-2012 at 8:33 AM, Marius said:

Hyper-V Server 2012 is supported, check out this post:

 

How about firewall rules for enterprise servers? I can't get it to work.

- Jaapyse

[Chris]

Hi Jaapyse,

You only need 443 TCP OUT to be allowed. What error message are you getting when you open click the "Validate Account" button in the Pulseway Manager app?

Chris

[Guest]

Hey Chris,

The message I get is the following:

There was no endpoint listening at https://ourenterpriseserver.com/Server.svc that could accept the message. This is often caused by an incorrect address or SOAP action. See InnerException, if present, for more details.

Even after adding the firewall rule it says so. It shows the error in a pop-up menu. Any ideas on how to fix this?

Thanks in advance,

Jaapyse

Update:

Another server in the network just went "offline", however I can reach it by RDP. This server now gives the same error when I ask it to validate the account. This means that the problem is not likely caused by Hyper-V, but by something else. I am now trying to find out what. At the time that the second server went offline I was not logged in to it, in fact I was not making any changes in the network at all.

Should I maybe start a new topic on this?

- Jaapyse

[Chris]

Hi Jaapyse,

That's alright, I've done it for you. The error message you've quoted is saying that the agent could not connect to the server. Check that the hostname resolves correctly in the DNS, that you can open the URL in the browser and that the date and time is correct.

Is the system that cannot connect in the same network with the Enterprise server?

Chris

[Guest]

Hey Chris,

The system is in a different network. Here's a strange thing I just noticed. While the Windows 7 "server" was able to connect, the Hyper-V wasn't. Then the Win7 all of a sudden stopped and I posted the update. The Hyper-V then started working. A few minutes ago the Win7 could connect again and... surprise!... the Hyper-V couldn't. Date & time are set correctly, and since both machines have been able to connect I assume that DNS is set correctly as well. I'm starting to think that the router only allows one outbound connection to the enterprise server at the same time or that the enterprise server has somehow reached some kind of limit. It can't be the amount of systems, because we're only halfway through with those.

Strange problem...

- Jaapyse

[Guest]

Update:

It turns out that this is the first (and currently the only) network with multiple agents in it. If it were just the one network, I wouldn't care, but we have various networks where monitoring still needs to be set up where two or three agents are necessary. Is this a known bug, a feature or is there a workaround?

Thanks in advance,

Jaapyse

[Paul]

Hi Jaapyse,

Chris has asked me to take a look at your case. You should be able to run as many agents as you'd like in the same network so this is not a known bug. Can you try to install two agents on a different network to see if the problem can be reproduced on a different environment?

-Paul

[Guest]

Hey Paul,

After an hour of research we are not sure whether we can recreate the problem or not. We have seen a similar problem before, when testing, but now we cannot recreate the problem in that particular network. For now we will try to implement Pulseway in some more networks and I will hauler when we see the problem again. Thank you very much for your support!

- Jaapyse

[Guest]

Update:

We now have multiple networks with multiple agents successfully running, one of them being a very similar system (one Hyper-V and a Windows server on it). Interesting fact: the system that is giving me trouble had also been used for testing before I got to set it up. I tried to give it a fresh install, but the settings remained. I now have uninstalled the agent, deleted the Pulseway folder in C:\Program Files and deleted all the MMSoft registry keys. After that, installed Pulseway again and the problem was not solved. I presume some of the information above will be useful.

- Jaapyse

[Chris]

Hi Jaapyse,

We need to have a look what's causing this issue.

Can you please send an email to support [at] pulseway [com] and we take it from there.

Chris

[Guest]

Hey forum,

The problem has been solved and I would like to share the solution with the community.

This was the problem: two systems within the same network were going online and offline (according to Pulseway) at random times. Usually for roughly 40 minutes, but sometimes half an hour or less and sometimes over an hour. Never were they both offline, but sometimes they were both online.

This was a / the solution. This network is unique relative to our other customers’ networks in that DNS is done by the DrayTek router.  All the other networks have either DrayTeks and a Windows server doing DNS or the default router the ISP provides. I noticed that clearing the DNS cache once or twice always resulted in the system coming back online within seconds. Therefore I added (more like enabled) a rule to clear the DNS cache whenever anything appears in it, and since then the systems have been online the whole time.

If anybody runs into this problem, I hope this topic / post can help you.

- Jaapyse