1

Screenshot of Lock Dump Error Message

About a week ago we started experiencing these LkMgr BEGIN Long Held Lock Dump error messages at the Domino server concole. Now we see that this is causing the http server to hang/crash.. It only takes minutes from we restart the server to the http hangs.. I have located the NoteID it's complaining about, and it's always a view design element. I have the tried to delete this view, and create a new one from scratch, but the very next day I get the same LkMgr BEGIN Long Held Lock Dump error message complaining about the new view design element..

Does anyone know what might be causing these Locks ? What can be done to eliminate them ?

Any information about this would be greatly appreciated ! Thanks !

Best regards, Petter Kjeilen

user1674918
  • 11
  • 1
  • 4

1 Answers1

0

Long held locks usually indicate an issue with disk access and not an issue with the document itself. From your screenshot, it doesn't say if there's an update of the view index is being performed, but that's usually what I see causing these. Is that view one that holds a large # of documents? Look at the design of the view and see if there's a way to reduce the number of docs in it if that's the case.

Look at the OS stats (on the OS or via Domino admin) and see if the disk queue length is >1. That would indicate that there's an I/O issue and you'll want to see if there's a way to resolve that (spread across more disk, move to SSD, etc).

You could try to run an updall of that view with HTTP down and see how long it takes to complete - 

load updall db.nsf -T viewname

Monitor the disk queue stats while that's going on and if it takes a long time to complete.

Also, can you open that view using a Notes client? Is there something in the design that forces a refresh of the view each time, maybe an @Now or @Today in a column. That would be "bad" and there are better methods to calculate that that don't cause a performance hit like this.

thardenb
  • 11
  • 1