TM1 Instance Restart Error: “Attempting to allocate memory from an invalid pool. Cancelling operation.”

[dwise]

Hello Experts,

We are getting an intermittent error when restarting one of our TM1 instances.

We are using persistent feeders and occasionally on restart it will stop at one of the feeder files and return the error in the tm1server.log file:

Code: Select all

TM1.Server.Memory.Debug		Attempting to allocate memory from an invalid pool. Cancelling operation.

The only workaround is to delete one of the feeder files to force a full recalculation (which takes a long time!). It doesn’t always stop at the same feeder file, although it’s generally one of three where the restart process stops.

Has anyone seen this issue before? Any known or suggested solutions?

Many thanks,
David

[lotsaram]

We get this same error occasionally on Linux instances where there has been a shutdown without SaveDataAll and there's some corruption in the tm1s.log file. The only solution it to manually delete the malformed records from the transaction log or delete the whole log file and restart. (Doesn't help with your issue. I haven't seen this error before when reading feeder files.)