MOSS NLB HTTP 500 Error

Hello.

I have two moss servers. server1 was the first one running moss. Then I installed server2 and joined this server to existing farm(on server1). Then I configured NLB (IGMP multicast).

Now I have the following problem_

when trying to access portal.domain.local on server2 it gives me HTTP 500 Internal server error. The site portal.domain.local shows as it should on server1 and also all other non-sharepoint sites shows correctly on both servers.

So, server1 in nlb has priority1 and server2 priority2. If I stop the server1 then the server2 takes over and shows HTTP 500 error. As soon as I start server1 again the portal is displayed correctly because server1 shows the portal.

Why am I getting the http 500 errors? Should I configure anything else (maybe something in CA) to tell sharepoint which server is now the "primary" ?

thank you.

LVL 1
SimpleJAsked:
Who is Participating?
 
SimpleJConnect With a Mentor Author Commented:
Ok, the cause for not displaying sites on server was loopback check. I have enabled DisableLoopbackCheck in registry and now site3 shows as it should.

However site1 still reports "Cannot make a cache safe URL for "init.js", file not found. Please verify that the file exists under the layouts directory".

The difference between site1 and site3 is that site1 is in different language than site3 (which is in en).

0
 
Ted BouskillSenior Software DeveloperCommented:
Did you set Affinity in the NLB settings?  If Sharepoint is storing sessions in memory then if page requests from one client are flipping between servers it causes a LOT of problems.

It's better to have a user stay with one server for their session.  If there is a fail over then they will flip safely to the other server upon a page reload.
0
 
SimpleJAuthor Commented:
I have Filtering mode set to Single host therefore I can not set the affinity !? Right now, during production hours, I can not test the "failover" but I can try to browse sharepoint site on priority2 server and it gives me http 500.

I set the filtering mode to single host because I want to set which server should serve the site first and which should in case of failover.
0
Cloud Class® Course: Ruby Fundamentals

This course will introduce you to Ruby, as well as teach you about classes, methods, variables, data structures, loops, enumerable methods, and finishing touches.

 
Ted BouskillSenior Software DeveloperCommented:
By doing that you don't receive the benefit of load balancing.  Why have one server sitting idle without any load?
0
 
SimpleJAuthor Commented:
the problem was that when I set the filtering mode to multiple and set the affinity to single sometimes the site was reachable and somethimes the http 500 error showed up....
0
 
Ted BouskillSenior Software DeveloperCommented:
Hmm, with Affinity on my farm has had no problems.  I'm not at work so I can't relay my settings right now.  However, I don't remember using multicast.
0
 
SimpleJAuthor Commented:
i`ve added one simple html site to nlb and is working great! so just sharepoint sites are problematic.

I`ve used multicast because i have one NIC. I have tried to open another sharepoint site and it asks me for credentials. So I enter domain username and password....and it prompts three times and then shows blank - white page....no error :S

the site which shows http 500 error right away is configured to allow anonymous access.
0
 
webmaisterCommented:
Is SharePoint working properly on server 2?   Can you browse the site successfully on server 2 testing it using the hosts file?

500 error sounds like a permission error on the App Pool idenity.  What do the event logs say on server 2?
0
 
SimpleJAuthor Commented:
no, server1 is working properly. sites on server2 are not working it prompts three times and then shows blank - white page
0
 
SimpleJAuthor Commented:
event log does not report anything
0
 
webmaisterConnect With a Mentor Commented:
Eliminate server 1 from the picture by shutting it off or mapping the urls to server 2 using the hosts file on the machine that you are using IE on.  Forget about the NLB for a minute and verify that server 2 is working properly. If not, need to troubleshoot server 2 for different issues.
0
 
Ted BouskillConnect With a Mentor Senior Software DeveloperCommented:
I agree.  I assumed both servers were working correctly.

You can use 'Alternate Access Mapping' to set a valid route to Server2 or even override the DNS entry for it using a local host file to route directly to it.
0
 
SimpleJAuthor Commented:
you were right. It seems to be a sharepoint problem on server2.

However, the three sites that I have on sharepoint server2 (site1 , site2, site3) all display different result !?

site1 asks for user auth. 3 times and then shows blank (white bg) page
site2 shows error: Cannot make a cache safe URL for "init.js", file not found. which I think is for language. I have verified that the file exists in layouts directory and that all language files are installed.
site3 keeps stopping application pool. The identity of the application pool is the same as on all the others sites. Identity is also a member of the IIS_IUSRS group.

so far the site 4 works as expected on both servers. I will try to add site 4 to NLB cluster and see what will happen with the accessibillity of the site.

any thoughts on above described issue ?
0
 
SimpleJAuthor Commented:
ok, I managed to fix the problem on site3. So, the application pool is now ok.

The problem on site1 and site3 now is that it asks me for credentials three times and then shows blank (white bg) page. Any ideas what could be wrong here ?

0
 
SimpleJAuthor Commented:
ok, solved my problems by copying files from healthy sharepoint server to this one causing issues.

In a couple of hours I will test the nlb conf. again to see if now everything is working as it should.
 
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.