Australian cloud for business software — built and run in Sydney since 2002

NetLeverage
NetLeverage

"Error 1053: The service did not respond to the start or control request in a timely fashion"

Technical article
Troubleshooting

About this article: This article is reviewed against the cited sources. This is source research, not lab validation. Check the stated versions and your own requirements before making changes.

Error 1053 is the Service Control Manager reporting that a service did not answer in time. Windows lists it as ERROR_SERVICE_REQUEST_TIMEOUT, 1053 (0x41D): "The service did not respond to the start or control request in a timely fashion." Microsoft's workaround is to raise the ServicesPipeTimeout value in the registry, and the same page tells you to "research this problem to determine whether it is a symptom of another problem."

Read the System event log and the current value first. A larger time-out helps a service that is only slow. It hides a service that is stuck, and it applies to every service on the server.

What the error means

Microsoft's page "A slow service does not start due to time-out error in Windows" says the cause: "The service control manager waits for the time that is specified by the ServicesPipeTimeout entry before logging event 7000 or 7011." The page names one case: "Services that depend on the Windows Trace Session Manager service may require more than 60 seconds to start." It gives no default value for the entry. It says the workaround is to "modify the registry to increase the default time-out value for the service control manager" and that if the entry does not exist you create it.

Other Microsoft pages show the same message in specific products. The AD FS 2.0 page lists the message "The service did not respond to the start or control request in a timely fashion." as a sign that the AD FS service "times out during startup", and gives the same registry change as its first step. The SQL Server Reporting Services (SSRS) page describes a start that ends in a time-out error, with event IDs 7000 and 7009 in the Application log. Its cause is outside the service's own code. If the system cannot reach Windows Update, "the network retrieval times out before the service can continue its startup procedure", and Microsoft adds: "If a service cannot report that startup completed after 30 seconds, the service control manager (SCM) stops the service." The 30-second figure is given for that SSRS case; Microsoft's service time-out page gives no default value. Raising the service time-out is one of four workarounds there, next to checking that firewalls and proxies allow the update traffic, changing Group Policy so the computer does not retrieve the lists, and a registry change that does the same.

So the message tells you which side gave up, not which side is at fault. The service might be slow, waiting on something else, or failing in its own start-up code.

Likely causes

Cause What you see How to confirm
A service that needs longer than the time-out to start Error 1053 on start; event 7000 or 7011 logged System log shows the event; the service starts when given longer
A dependency that is slow to start Microsoft names services that depend on the Windows Trace Session Manager service Start the dependency by hand first and see whether the service then starts; Microsoft's pages give no tool for listing dependencies
The service waits on a network resource Microsoft's SSRS page: certificate trust list retrieval blocked, events 7000 and 7009 The server has no internet access or a firewall blocks Windows Update
The service's start-up code fails A longer time-out changes nothing Microsoft's page: research the problem to see whether it is a symptom of another one
The service account or database is not usable AD FS example: a stopped configuration database, or a changed account password Microsoft's AD FS page checks both in the steps that follow its time-out step

Prove the cause before changing anything

All of these only read.

  1. Open Event Viewer on the server. Look for Service Control Manager events 7000 and 7011, which Microsoft's page says the manager logs when the time-out expires, and event 7009, which Microsoft's SSRS page shows in the Application log. Check the System log as well. Read the service name and the text.
  2. Read the current value. If the value is absent, the command reports it:
reg query "HKLM\SYSTEM\CurrentControlSet\Control" /v ServicesPipeTimeout
  1. If you know a service the failing one depends on, check in Services.msc that it is running. Microsoft's pages give no way to list a service's dependencies, so use the product's own documentation.
  2. If the service belongs to a product with its own Microsoft page, such as AD FS, read that page's checks as well: its AD FS 2.0 page lists the time-out first, then whether the configuration database service is running, then whether the service account and its password are current.

Write down the output of step 2. The record is the way back.

Example (illustrative): on app01 at contoso.com.au a line-of-business service called ContosoSync fails to start after every restart with error 1053. The System log shows event 7000 for it on each start. ContosoSync has a dependency, and starting that service by hand first makes ContosoSync start. The reg query command reports that ServicesPipeTimeout is not found. That is a slow dependency with no time-out set, so Microsoft's workaround fits.

Fix

Raise the value by a small amount. Microsoft's page uses 60 seconds and says: "Increase the number carefully. We recommend you increase the number with a small amount at a time untill the service can start." The value is in milliseconds, so 60 seconds is 60000.

Before you change it, make sure you have the reg query output from the checks above. If the value was absent, note that.

Risk: the next steps set the service time-out for every service on this server, not only the one that failed. A service that is stuck for good then holds up its start, and anything that depends on it, for the longer period. Microsoft's page says the change needs a restart of the computer, so choose a time when the server can be restarted.

Microsoft's steps, in Registry Editor:

  1. Click Start, click Run, type regedit, and then click OK.
  2. Locate and click the registry subkey HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control.
  3. In the right pane, locate the ServicesPipeTimeout entry. If it does not exist, on the Edit menu point to New, click DWORD Value, type ServicesPipeTimeout, and press Enter.
  4. Right-click ServicesPipeTimeout and click Modify.
  5. Click Decimal, type 60000, and click OK.
  6. Restart the computer.

To undo it: if your earlier reg query reported that the value was not found, delete the value:

reg delete "HKLM\SYSTEM\CurrentControlSet\Control" /v ServicesPipeTimeout /f

If the value existed, set it back to the data your earlier reg query showed. Replace the placeholder with that number, and keep REG_DWORD only if the earlier output showed that type:

reg add "HKLM\SYSTEM\CurrentControlSet\Control" /v ServicesPipeTimeout /t REG_DWORD /d <recorded value> /f

Run only the one command that matches your earlier output, and restart the computer afterwards.

If 60000 is not enough, Microsoft's page still asks for small steps and for research into the cause. The AD FS page uses the same value, 60000, in its first step, and its next steps check whether the database service is running and whether the service account is valid after a password change. Those are checks in the AD FS product, but they show the pattern Microsoft follows: time-out first, then what the service waits on, then the account it runs as.

The SSRS page adds a caution that applies to the registry change itself. It says "serious problems might occur if you modify the registry incorrectly" and "back up the registry before you modify it", which is why the value is recorded above and the back-out is stated for both cases.

Confirm the fix

  • Restart the server and watch the service start. It should reach Running without the 1053 message.
  • Read the System log again. Events 7000 and 7011 for that service should be absent after the restart.
  • Run the reg query command again and confirm the value shows what you set.
  • Time the start. If the service only starts after nearly the whole time-out, the underlying slowness is still there.

What to watch afterwards

  • Start time of the service after later updates, restarts and patching. A time-out that was enough last month may not be enough after a change to its dependencies.
  • Other services that now start late. The time-out is global, so a service that used to fail quickly now takes the whole period to fail.
  • The cause itself. Microsoft's Windows Trace Session Manager example and its SSRS example (certificate trust lists unreachable) are both causes outside the service's own code. If the service is your own application, Microsoft's "Debugging a Service" page says "To debug the initialization code of an auto-start service, you will have to temporarily install and run the service as a demand-start service."

Risks and limits

These commands have not been run or lab-tested.

Microsoft's service time-out page gives no default for ServicesPipeTimeout and no upper limit, and does not say which Windows Server versions it was tested on beyond "Supported versions of Windows Server". It does not explain how to find a slow dependency. The AD FS and SSRS pages are product-specific: they show that the same message and the same value appear in those products, and they do not claim the same cause for other services.

Sources