OWA gives HTTP 500 error for several users

Exchange 2013 SP1. Since last week we have the issue that some users are getting an HTTP 500 Internal Server error when they access OWA from the internet after logging on. For other users it works fine. When the users who are having problems access OWA from internally they can logon fine. OWA authentication is set to Windows Integrated.

For external access OWA is published using TMG. No errors are reported in the eventlog of TMG or the Exchange server. Any idea what might cause this? Is there a way to get some more detailed information about the HTTP error 500?

I've checked the issue with inheritance on the AD user object, but the userobjects have inheritance enabled. Also, since it's Exchange 2013 it doesn't have the Form based authentication service which also might cause this error.

Any idea?

Frank.

March 21st, 2014 10:37am

Hi Frank,

Are you connecting to 1 server, or do you have multiple servers configured via a load balancer?
If it works for other users maybe they are using another CAS?

Might be good to check with httpwatch to see what responses you get.

Do these users have problems with activesync too?


Free Windows Admin Tool Kit Click here and download it now
March 21st, 2014 12:27pm

Hi,

we only have one server. No loadbalancing. ActiveSync works fine, although I do notice that at random mobile devices are getting blocked by Exchange for no apparent reason and I have to manually unblock them. I've already opened a support case for that.

Frank.

March 21st, 2014 12:33pm

Hi,

Firstly, I would like to know the difference between the problematic users and good users, are they in different site?

Since the problematic users can login from Internal OWA, I think the issue is more related to the TMG settings.

Thanks,

Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com

Free Windows Admin Tool Kit Click here and download it now
March 24th, 2014 8:13am

It's a single domain. All users are in the same AD domain. When looking at the TMG logs, the response from exchange is http 500, this error can also be found in the exchange IIS logs itself. Even weirder is that, when the user get an HTTP 500 error and logs on from another PC from the internet then it works fine. Clearing temporary internet files and cookies won't help. Using another browser also won't help. Alhough, because of this behaviour, you might expect that this is a client issue. But it looks like it isn't because I get more and more reports from users that they can't use OWA, but again, not all users. A very strange issue....
March 24th, 2014 8:22am

It getting stranger.

With one account on the same PC I get the HTTP error 500 after logging on.

When I use another account on the same PC I can logon to OWA fine. No issues.

Then I disable protected mode or add the owa site to trusted websites and both accounts are able to logon to OWA. No http errors 500 anymore.

As soon as I remove the OWA site from trusted wbesites, or enable protect mode again. The original account fails again with HTTP error 500 and the other account still works.

Any idea what's going on here?

Frank.

Free Windows Admin Tool Kit Click here and download it now
March 24th, 2014 1:10pm

Hi,

Is TMG SSO enabled? If so, try disable it and see whether it works.

Thanks,

Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com

March 25th, 2014 2:00am

Hi, The problem is not with TMG. The error also occurs when I bring the client which has the issue to our internal network and connect to OWA thus bypassing TMG. Frank.
Free Windows Admin Tool Kit Click here and download it now
March 25th, 2014 2:06am

Exchange 2013 SP1. Since last week we have the issue that some users are getting an HTTP 500 Internal Server error when they access OWA from the internet after logging on. For other users it works fine. When the users who are having problems access OWA from internally they can logon fine. OWA authentication is set to Windows Integrated.

For external access OWA is published using TMG. No errors are reported in the eventlog of TMG or the Exchange server. Any idea what might cause this? Is there a way to get some more detailed information about the HTTP error 500?

I've checked the issue with inheritance on the AD user object, but the userobjects have inheritance enabled. Also, since it's Exchange 2013 it doesn't have the Form based authentication service which also might cause this error.

Any idea?

Frank.

I have different environment - 3 mailbox servers in DAG and 2 CAS in NLB, Exchange 2013 SP1. And the same problem. Since last week some users can't login to OWA and getting that HTTP Error 500. 
March 25th, 2014 10:10am

Make sure that the "Microsoft Exchange Forms-Based Authentication" service is running on your CAS server.
  • Edited by TheGuyFromIT Tuesday, March 25, 2014 10:24 AM Typo
Free Windows Admin Tool Kit Click here and download it now
March 25th, 2014 10:22am

It's doesn't work for Exchange 2013 - there are no "Microsoft Exchange Forms-Based Authentication" service
March 25th, 2014 10:25am

Is there any event about RPC Proxy failure? I had something similar and the problem was with rpc proxy...

Try do see that and look for ASP.NET events aswell... On my case, i had followed this http://social.technet.microsoft.com/Forums/en-US/52f3f6b7-0114-4056-85d7-e46c615b7fa9/2013-cas-server-rpc-aspnet-event-1309-code-3005-unhandled-exception

It solved my problem with blank pages and 500 error pages.


  • Edited by KelsonDaniel Wednesday, March 26, 2014 8:52 PM
  • Proposed as answer by KelsonDaniel Wednesday, October 15, 2014 2:30 PM
Free Windows Admin Tool Kit Click here and download it now
March 26th, 2014 8:52pm

No errors in the eventlog. Outlook over rpc works fine. Owa works for some users, but not for all.... It's very random....
March 27th, 2014 8:28am

I'm seeing a few cases of this cropping up following SP1 installation. I have had it myself in a test environment but thankfully not Live yet. The test environment was very clean, with no complications: - Single 2012 R2 server, setup as DC, Exch 2013 Mailbox and Client Access (not a best practise arrangement, granted). Test Scenario 1 - installed all Windows Updates - Got the HTTP500 error for OWA and ECP. Installed Exch 2013 SP1, still got HTTP500, but more intermittent. Test Scenario 2 - setup exactly the same scenario, but this time installed Exch 2013 SP1, before Windows Updates. Initially the HTTP500 was present but far more intermittent. Also, the Mailbox Transport Submission server fails to auto start intermittently, but setting DNS address under the server settings in ecp (when I could get in ecp) did appear to resolve the issue. So, where am I going with this... My thinking is whether a recent Update doesn't play nicely with Exch 2013 and/or SP1? I haven't had a chance to test this further, but perhaps looking at your update history might show something worth exploring. Be interested to hear how you get on. Thanks
Free Windows Admin Tool Kit Click here and download it now
March 27th, 2014 8:52am

I can't get an hold on it. With some people clearing the temporary internet files solves the issue. With other this didn't work. For some other users, adding the owa site to the Trusted Internet Zone in IE solves the issue. For others, this doesn't work.

I think we can troubleshoot a lot further if we know what error HTTP 500 exactly is, since there are no errors reported in the eventlog of the Exchange 2013 SP1 it's hard to track down. Is there some more logging we can enable so that we can see what the HTTP 500 error actually is? The 500 error does show up in the IIS logs on the Exchange server, so it is a server side issue.

Attached the update history. The time of installation of the updates seem to match the date when the first reports of the HTTP Error 500 came in. I'm going to uninstall these updates and see what happens.

Frank.

March 27th, 2014 10:59pm

Frank,

Good work on posting the update history! That is very interesting indeed.  I'll be very interested to hear how you get on.  I am planning to re-run the entire scenario on a test machine again but I'm going to struggle to get that done before the end of the weekend.  If I make any progress, I'll let you know.

Thanks for that

DS

Free Windows Admin Tool Kit Click here and download it now
March 27th, 2014 11:56pm

@Simon, my response wasn't an answer to the issue. The issue is still there.
March 31st, 2014 8:19am

Hi Frank,

Please check the IIS log and let us know the detailed error code about 500. I mean the sub-number after the 500. Or you can check the following article to understand more about the 500 error.

http://support.microsoft.com/kb/943891/en-us

Thanks,

Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com

Free Windows Admin Tool Kit Click here and download it now
March 31st, 2014 8:20am

Hi Simon,

the subcode is 0:

500.0 - Module or ISAPI error occurred.

Frank.

March 31st, 2014 8:29am

Frank, did you ever get to the bottom of this one? Others are still reporting similar issues, and it does seem to occur after Windows Updates although I've still not had time to tie down which update specifically. One suggestion doing the rounds is as follows: At an administrative/elevated command prompt, run the following commands: lodctr /r and then run IIS reset - Possibly a long shot... I've not had a chance to try it, but as soon as I reproduce the issues again, I will try it myself. Be interested in any news/progress. Thanks
Free Windows Admin Tool Kit Click here and download it now
April 3rd, 2014 12:26am

Unfortunately no. Still troubleshooting. Reports of users are still coming in though. Frank.
April 3rd, 2014 12:32am

Hi, have same issue here and a case with Microsoft trying to resolve it. We have moved from Exchange 2013 Server 2008 R2 to Exchange 2013 SP1 and Server 2012 R2. After mailbox migration to the new databases and clean uninstall of the "old" servers some of the user gets the http 500 error when they try to log in from OWA. Outlook works just fine for those users. Answer from Microsoft after a week of emailing:

"After doing the test in my lab, I find the same issue with yours. Then confirm this with our related team. It seems a known issue. Thanks for making us noticed. This issue is hit when Mailboxes are moved from one Database to another and old Database is deleted. Clients store old database guid in cookie, this guid is used to locate Database, which fails with the error. Issue will only occur if user attempts to login from client OWA which had logged-in successfully earlier before move. Now our Production team is working on this. But before any fix comes out, we can try a workaround to temporary make OWA work well:

1. Firstly delete cookies from clients (IE clear cache and delete cookies)

2. Recycle the application pool. (IIS recycle MSExchangeOWAAppPool and MSExchangeOWACalendarAppPool)"

I will try the workaround today cant confirm yet. Can someone else try and post the results also.



  • Edited by sc00ut Tuesday, April 08, 2014 6:17 AM
Free Windows Admin Tool Kit Click here and download it now
April 8th, 2014 5:57am

I don't think it's a cookie issue. Here's why: when a user reported the error, he gave me his credentials to troubleshoot. I logged on from my PC at home atnd got exactly the same error. If it was a cookie issue I shouldn't get this error because I never logged on before with this credentials. Frank.
April 8th, 2014 7:28am

Also, I've tried clearing the temporary internet files, but that didn't make any difference. Sometimes it works by adding the owa site to the trusted websites zone in internet explorer. After that it started working for several users, but not for all. Still a mystery.
Free Windows Admin Tool Kit Click here and download it now
April 8th, 2014 7:30am

We are having the exact same problems, and with regards to the microsoft description / reason for this occuring, we did indeed move users to new exchange 2013 databases / delete the original source databases after which we began experiencing the issue.  We also experienced seemingly random issues with activesync user devices no longer receiving email / receiving the error "Unable to connect to server" error's etc.

I can confirm that after deleting the cookies in my chrome browser that were specifically related to my exchange 2013 client access url (I believe there were 20+), closing the brower, reloading the browser and accessing the site, I was able to again get into the ECP.  It remains to be seen if this is a long term fix or whether the problem will resurface / grow two new heads.  Note that I did not have to recycle any of the application pools.

With regards to the activesync issue, I had the user clear his mobile device cache and cookies.  Retried to sync, no dice.  Delete the mobile device exchange profile, no dice still prompting with "unable to connect to server".  Then after deleting the activesync device from the user account properties in the exchange control panel.  Success!!   The user was able to again create a mobile device exchange profile and sync email.  Again remains to be seen if the fix is a permanent one or a temporary one.

I would very much appreciate it if you receive any additional information on this bug from the microsoft ticket / representative, such as permanent patch / fix development time frames or availability, or if this is associated with a particular windows update and if so which update / set of updates are responsible.  We have multiple customers that are on 2013 and some that have already performed mailbox database moves / removals, but have not yet been affected so I would like to keep it that way by avoiding the installation of any patches that are associated / responsible.

Thanks

April 8th, 2014 8:05pm

I will post information as soon as Microsoft have come up with permanent solution, hotfix or whatever it will be. Right now the answer from them is "We are working on a solution, this is just a workaround"

I can also confirm that clearing cookies in some cases does not work. I had affected user who logged in from "clean" computer and still no good. But its worth trying because it works in some cases.

Thank you for your information.

Free Windows Admin Tool Kit Click here and download it now
April 10th, 2014 7:03am

There's a fast publish KB article available, although I'm not sure it covers the entire issue since deleting cookies doesn't seem to always work.

http://support.microsoft.com/kb/2958434

Frank


April 12th, 2014 1:55pm

Still nothing new?

We still have problems with our users..... We can only delete user completely and recreate it with the same data. This is the only possible solution for us for now.

Free Windows Admin Tool Kit Click here and download it now
April 16th, 2014 2:44pm

The same problem, what about another solition from Microsoft?
May 7th, 2014 7:17am

Also same Issue. I'm in middle of migration from 2007 to 2010. Frist started with some test accounts and 90% do not encounter any problems but couple did. One of this account is Domain Admin and Exchange admin account. Since this account mailbox was not important I disable it and using secondary Exchange Admin account recreated. Now the weird part. It is still getting HTTP 500 error on both ECP and OWA for this account. And if I use different account on same computer there is no problem with login in to ECP. Extra strange. If any of you have any update pleas post as I'm stuck with migration. 


Free Windows Admin Tool Kit Click here and download it now
May 12th, 2014 7:37am

We're having the exact same issue.  Moved users to a new server, 2012 R2 with EX 2013 SP1.  Everything started acting up after removing the old databases.  Intermittent issues, OWA wont work for some users on computers they never used before.  Will be watching this thread for an update.
May 14th, 2014 9:19pm

We had this issue for a couple of weeks and it disappeared all of a sudden since about two weeks. Don't know why, but I suspect it might have to do with some cookie expiring which forces a renewal...
Free Windows Admin Tool Kit Click here and download it now
May 14th, 2014 9:23pm

We have solution about this problem. I was created another database and move there all mailboxes and over time the problem disappeared. Of course you must clear cache in your browser, before trying. And at this time don't delete old database at least one day :)

Good luck!



  • Edited by Navynine Thursday, May 22, 2014 8:52 AM
  • Proposed as answer by Navynine Monday, May 26, 2014 7:04 AM
May 22nd, 2014 8:51am

This issue was addressed recently in   

http://support.microsoft.com/kb/2936880/ Cumulative Update 5 for Exchange Server 2013

There is also a workaround in KB http://support.microsoft.com/kb/2958434.

Regards,

Jarrett Renshaw, MSFT

This posting is provided "AS IS" with no warranties, and confers no rights. 


Free Windows Admin Tool Kit Click here and download it now
August 25th, 2014 11:19pm

Thank you so much, that fixed my http 500 error... long day, new certificate.
April 9th, 2015 5:22pm

This topic is archived. No further replies will be accepted.

Other recent topics Other recent topics