Hello!
I am checking the server manager log to help diagnose server issues. In my watching, I have noticed a couple Severe errors that I cannot find help with addressing.
We are on 10.8.
Message:
Exception in Put(): 31152 Put exception: the client id 31152 already has a value Attendance/MiddleAttendance_2021_2022.MapServer31152 checked out.
Code: 393939
Method Name:
DynamicServiceCache::GetService
I also receive the same code/method name with an Exception in Get().
Then it will follow with warnings and severe for the "System/DynamicMappingHost.MapServer"/Error code: 8001
Any help will be greatly appreciated!!
Heather.
Hi Everyone, coming across a similar error. Does this have to do with shared instances? In our case, the error is saying that map service A (tied to database A and in its own folder on ArcGIS Server) cannot be checked out because map service B (tied to database B and in its own folder on ArcGIS Server) is already checked out.
We have 254 map services, most of them published as shared instances.
Errors
Exception in Get(): _GetItemIterator exception: client id 19972 is trying to check out FolderA/MapServiceA.MapServer but it already checked out FolderB/MapServiceB.MapServer19972
Service instance activation failed in 'System/DynamicMappingHost.MapServer'. AutomationException: 0x80004005 - Unspecified error
I'm seeing this also. I wonder if it is tied to Shared Instances somehow. The 2 feature services mine are referencing have nothing to do with each other. I'll have to call Esri Tech support.
Ditto on 10.9.1.
This is causing Fieldmaps to crash repeatedly for field staff.
@Anonymous User did you learn anything from tech support?
@ahargreaves_FW sorry, I never called them back in August. I saw a lot of error messages and poor performance from Shared Instances so the feature services that had high demand I changed to Dedicated Instances and haven't had any problems since. Check how many cores you have and are licensed for with ArcGIS Server and how much your server can handle - that was also part of our problem.
yep - we're getting the 'but it already checked out' warning - plus a few crashes - we've just gone live with Pro generated services so they are defaulting into the Shared Pool - whilst I get the theory, I;m not sure of it in practice
A question: The Shared pool seems to have an Idle Time out defaulting (and unchangeable) to 30 minutes - if no hits are made to any of the services in the Shared Pool in that time does the whole pool go to sleep and therefore takes a long time to restart properly when a request is made?