Fix RPC_E_CALL_COMPLETE (0x80010117) on Server & Cloud
This COM error means your call context expired. Fix it in 30 seconds by restarting the service, or dig deeper with registry tweaks and DCOM config changes.
What causes 0x80010117?
This one's a COM+ runtime error – the system tried to use a call context that already finished. I've seen it most often on Windows Server 2016 and 2019 running Exchange 2016 or SharePoint, where a client (usually Outlook) connects, the server does something, and when it comes back to report, the context is gone. It's not a broken cable – it's a timing problem in the component object model.
Other culprits: third-party COM components that hold references too long, or stale DCOM permissions. Had a client last month whose entire print queue died because of this – a custom VB6 app was calling a printer driver that returned before the context was released.
The 30-Second Fix: Restart the Hung Service
First thing I do – don't overthink it. Open Services.msc, look for the COM+ Event System service. Right-click, Restart. While you're there, also restart System Event Notification Service. If the error's tied to a specific app (like Exchange RPC Client Access), restart that service too.
Why this works: it clears the COM call context pool. The error shows because a prior call didn't finish cleanly. Restarting forces a fresh pool. It sounds basic – but I've fixed maybe 30% of these calls with just that.
The 5-Minute Fix: Registry Tweak for Call Timeout
If restarting didn't help, the real issue is that the system's default CallTimeout is too short for your workload. Open Regedit and go to:
HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\OleCreate or edit a DWORD named CallTimeout. Set it to 300000 (that's 5 minutes in milliseconds). That gives COM components more time to finish before the context expires. I use 300000 as a starting point – if you're still getting errors, bump it to 600000.
On the same key, also set EnableDCOM to Y (value 1) if it's not already. I've seen admins disable DCOM to “secure” the server, then wonder why integrated apps break. Don't be that guy.
After the registry change, reboot the server or restart the COM+ Event System again.
The 15+ Minute Fix: Deep DCOM Configuration Reset
When the simple stuff fails, it's time to reset the DCOM security and permissions. Open Component Services (dcomcnfg), navigate to Component Services > Computers > My Computer. Right-click My Computer, choose Properties.
Under the COM Security tab, click Edit Default in both Access Permissions and Launch and Activation Permissions. Make sure the following are present with Allow:
SYSTEM – Local Access, Remote Access
Administrators – Local Access, Remote Access, Local Launch, Remote Launch
INTERACTIVE – Local Access, Local LaunchIf you see Everyone or ANONYMOUS LOGON, remove them – they're not needed and can cause conflicts.
Now go to the Default Properties tab. Set Default Authentication Level to Connect. If it's higher (like Packet), some components treat that as a security violation and drop the context. Set Default Impersonation Level to Impersonate. Click OK.
Still broken? Next step: clear the entire COM+ catalog. On a command prompt as admin, run:
regsvr32 /u /s comadmin.dll
regsvr32 /s comadmin.dllThen reboot. This re-registers the COM+ infrastructure without losing your custom COM components. I've had to do this exactly twice – both times it was the last resort that worked.
Final check
After any fix, test by reproducing the error. If it's Exchange, open Outlook, force a folder sync. If it's a custom app, run a normal operation. If the error's gone, you're done. If not, you may have a deeper code issue – like a component that's not releasing its context after an async call. In that case, you're looking at Application Event Log for the specific component CLSID and contacting the vendor.
Bottom line: start with the restart, move to registry timeout, then reset DCOM. I've never seen it require more than that.
Was this solution helpful?