Gemini Community Support Site

This Gemini community support site can be used to find solutions to product issues. You can log in using Open Id, Google Profile and even Facebook. Feel free to ask a question or browse FAQs and documentation. Product tour videos are also available along with how-to videos demonstrating key Gemini capabilities.




General network errors -- upgrade from 1.9.1 to 2.0.2

web-app

Back in January we upgraded from 1.9.1 to 2.0.2 and started getting General network errors. At the time I sent an email to countersoft support describing the issue, but they did not have any suggestions. I'm posting on the forum now in the hope that more may be known about this problem.

Originally, we had a single server running the Gemini web application (W2K Server, IIS 5.0, ASP.NET 1.1)which was located in the DMZ; the database was on a different server (W2K Server, SQL Server 2K) that was located on our LAN (inside the DMZ).

After upgrading to 2.0.2, everything seemed to be working fine except we started getting general network errors a few times each day. The place and timing of these errors seems random. We can go into Gemini one time and everything works fine, then the next time we get an error when moving between pages in the application. The errors do not correlate with a particular page, action, or user.

By far the most common error that shows up in the Gemini error log is:

Error Message = General network error. Check your network documentation.

Stack Trace = at CounterSoft.GeminiBiz.33c10d769b8bcf93.48e77a23dc9d1a1d(String 1894643d4df24f5c) at CounterSoft.GeminiBiz.33c10d769b8bcf93.ConstructCustomPrincipal(IIdentity id) at Gemini.Global.Application_AuthenticateRequest(Object sender, EventArgs e)

Usually, pressing Refresh in the browser reloads the page correctly, but needless to say it’s disconcerting to be getting these errors on a regular basis and we don't want our customers running into this kind of thing.

As a test, I set up another instance of Gemini 2.0.2 on a server inside the DMZ. If we use this instance we do not get any errors.

As another test, I set up another server (2K3, IIS 6.0) in the DMZ with Gemini 2.0.2 and we get the same errors with the same frequency.

I saw Harvey's FAQ note about General Network Errors, but in all cases I have Integrated Windows authentication enabled.

Any ideas?

Thanks,

Fred

Pawnder
· 1
Pawnder
Replies (5)
helpful
0
not helpful

[quote user="Pawnder"]

As a test, I set up another instance of Gemini 2.0.2 on a server inside the DMZ. If we use this instance we do not get any errors.

As another test, I set up another server (2K3, IIS 6.0) in the DMZ with Gemini 2.0.2 and we get the same errors with the same frequency.

[/quote]

 

What is the difference between the 2 setups above?


Saar Cohen
· 5000
Saar Cohen
helpful
0
not helpful

I have three installations:

1) Gemini 2.02, W2K Server, IIS 5.0, in DMZ  -- gets general network errors

2) Gemini 2.02, W2K Server, IIS 5.0, inside DMZ -- does not get errors

3) Gemini 2.02, W2K3 Server, IIS 6.0, in DMZ -- gets general network errors (same symptoms as #1)

The Web.configs for all three are the same with the following exceptions:

- the Gemini URLs are different for each installation

- the database connection for #2 uses an internal IP address since it is located in the LAN with the database server (connection strings for #1 and #3 are the same.)

 

- Fred


Pawnder
· 1
Pawnder
helpful
0
not helpful

This does not make sense really. All 3 servers are inside the DMZ. Surely they could all use the same connection string.

It seems like the connection string you are using (IP) in #1 and #3 is not being resolved properly sometimes.


Saar Cohen
· 5000
Saar Cohen
helpful
0
not helpful

I should have use clearer language...

#1 and #3 are "in DMZ" -- outside of our LAN (uses external routable 206... IP address)

#2 is "inside DMZ" -- in our LAN (uses internal 192... IP address).

Both connection strings use a specific IP address, so it shouldn't have to go to DNS to resolve the server name.

If no one else is having this problem, then it's probably something particular to our network... I just thought I would ask in case anyone else was having a similar issue.

Thanks,

Fred


Pawnder
· 1
Pawnder
helpful
0
not helpful

Thanks for clarifying this.

As far as we know, no one else has reported such issues.


Mark Wing
· 9108
Mark Wing