Event_pool_size

HCL Domino Domino Forum
1

Hello,

Since I upgraded to Domino 6.0.1, I get this message on the server console :

Warning: Cannot record event - cannot keep up with event occurrence rate!

The EVENT_POOL_SIZE param in NOTES.INI can be set up to 5Mb. It defaults to 3Mb.

How can I set this parameter to 5MB to avoid this error message ?

Many thanks in advance

2

Subject: Solution: EVENT_POOL_SIZE

Lotus states, that the setting for “EVENT_POOL_SIZE” is set in MBytes as of version 6.x.

Through our tests we have established, that this is not correct (at least on Linux and AIX vers. 6.5.1/2). The setting should still be set in Bytes!

To check this, one can simply call a “tell event dump” and will be returned the size of the event pool. This will be maximum 65536 Bytes when the size is set in MBytes.

To correct this:

  1. Stop the Domino server

  2. Change the Notes.ini setting from MB to Bytes (or in config doc, but this then needs to be checked if it has been written into the Notes.ini)

  3. restart the Domino Server

To check if working correctly:

  • call the “tell event dump” - you will see that a greater size has been allocated (than 64Kb).

  • check if errors concerning event monitor log still occur. If so, increase the size (in Bytes)

I’m going to write Lotus Support as well.

Hope that helps.

Andreas Hechler

Magirus International GmbH

Stuttgart - Germany

3

Subject: EVENT_POOL_SIZE

Try EVENT_POOL_SIZE=7340032, which would set the size to 7Meg (710241024). The message is also incorrect. The default value is 5Mb and currently there is no enforced maximum value.

You might also want to check to see what debugging parameters have been enabled. Excessive debug logging can sometimes cause the referenced message.

4

Subject: RE: EVENT_POOL_SIZE

Hi

Hmm, I’ve seen this error appear after upgrading an R5 internet mail routing server to R6.02CF1.

We’ve tried setting the value of event_pool_size up to 15Mb but we still get this error. Does the server need to be restarted for this to take effect or is restarting the Event task sufficient enough?

With the Event task shutdown the only console messages we see are the usual “message transferred to …” routing messages.

The same Events4.nsf database is used on a mix of R5 and R6 servers and we only see the errors on the busy R6 servers.

It appears that R6 is no longer able to keep up with the normal event messages that appear on the consoles or else the event_pool_size config variable is being ignored.

With debug_event=1 set I see the following values no matter what event_pool_size is set to;

Event pool size = 5036262 bytes, used = 4811572 bytes

I’d be interested to hear if someone else has been able to fix this problem.

Regards

Scoot