Exchange 2010 some users cannot always send mail

The external DNS setting may be something you would want to check.  This article relate to Exchange 2013 but it can give you insight on what to troubleshoot.  There should be similar settings for Exchange 2010.

https://exchangemaster.wordpress.com/2014/06/10/mails-stuck-in-the-draft-folder/

http://thoughtsofanidlemind.com/2013/03/25/exchange-2013-dns-stuck-messages/

Also have a look at this:

https://social.technet.microsoft.com/Forums/exchange/en-US/e7d72510-3d86-41ea-ba1a-b742aeb4ccc8/error-not-able-send-mail-from-owa-2010-when-i-send-mail-the-mails-go-to-drafts-folder?forum=exchange2010

https://www.petri.com/email-stuck-in-drafts-folder

So according to your info it works internal but external it does not work? Do you also know if external mail is working

I would be interested in the message tracking logs
Make sure DNS is configured also use nslookup
Verify MX record

I would think DNS problems would affect all users and all clients. It might be corrupt Outlook profiles or OST. You can test both at the same time by creating a new outlook profile to see if the problem goes away.

On a second read of the OP, it sounds more like mailbox corruption. The easiest way to straighten out minor mailbox corruption is to move it to another store.

For more serious mailbox issues, use the mailbox repair request cmdlet: https://technet.microsoft.com/en-us/library/ff625221(v=exchg.141).aspx

I agree this does smell like a possible DNS problem (although the limited scope makes me suspicious); however, there's a new, fun little gotcha with Exchange 2013 called component states that I've personally witness cause email to queue in the drafts folder.  It may be worth a look to verify they're all active.  Basically you want them to look like the first screenshot of this Technet Blog:

http://blogs.technet.com/b/exchange/archive/2013/09/26/server-component-states-in-exchange-2013.aspx

Note the screenshot shows them all active, but out of the box ForwardSyncDaemon and ProvisioningRps are Inactive.

Another possibility is these mailboxes are hitting their throttling policy which CrApple iOS devices are very well known to do while connected to an Exchange server.  Check out Log Parser Studio 2.2, it's pretty freakin awesome for situations that require parsing any kind of log (read IIS, Message Tracking, Protocol, etc):

https://technet.microsoft.com/en-us/scriptcenter/dd919274.aspx

When you say 'our logs files don't show anything', which logs are you referring to?

As per the author, it has issue in webmail as well, so it should not be anything regrading on "DNS"

Please advise what type of storage you are using for the exchange server.

Mailbox corruption is probably the cause but what lead to that will be a question mark.

As per another expert, try to move the issue mailbox to other database and see if that fixed it

Is there an echo in here?

Hello everyone

Thank you very much for all the replys, and I am sorry that I have not been able to get back before now.

First of all, it is not DNS. I have testet trough MS testconnectivity and all fine.

I may think as some of you that this could be corrupted users, but the funny part is that restarting the store.exe makes it work for about 8-10 hours, and then it stops again :)

Also, when this happens users can still send from Windows Outlook! Only Apple devices and OWA are makeing the problems! And sending form OWA makes it go to draft, but then we can open it from drafts and send without problems...
One of my colleagues have installed a bunch of security updates on the 01-02-2016 and I have done some digging, wich indicates that it is after these updates they started to have problems. Please see attached images for update overview.
Mabye one or more is known to affect problems like these, but I have not been able to find anything. I will now read what exactly each of the updates does, but mabye one of you can remember something about one or more of them?

Thanks agaian, I will get back soon...
Server-2008-R2-upd1.jpg
Server-2008-R2-upd2.jpg

Okay, so we have now had some time to search for the error or symptoms.
We are down to this. If we have a user that cannot mail, then restarting Default Web Site in IIS make the user able to send 1 as in one! mail and then it does not work again???

I have never seen such behavior from an Exchange server...

The issue has been narrowed down after extensive research:
When Outlook 2013 for Windows is open in a remote Desktop session the same user can't send e-mails from OWA, iOS or Outlook for Mac. Closing Outlook for Windows immediately makes the troubled devices work until Outlook for Windows is opened again.
Please note this is only for some users. We do have users that have no trouble sending from OWA when Outlook or Windows is open on the same Remote Desktop Server. All users are on the same exchange server. Also the affected users consistently have this issue when Outlook for Windows 2013 is opened
Thanks in advance

We found the solution our selfe, no one could have forseen this issue without knowledge of the whole system