Please advise me about following.

The session "Initialize. Roll and Update Scenario(cprpd4200m000)"
show following message and ends incorrectly,
when we use it in the parallel processing mode.

But it does not occurs always.
We run above sessions every day. But it occurs about three times a month.

Message:
unable to receive message by mailbox, error 22

Contents of log "log.comdll0200"
07-01-23[20:11:09];31.089;20507;3;aaa0169;otccom0200(3);Timeout (300 sec) reached in server 3;client daemon reading queue 316937 and writing to this server's queue 102913:
07-01-23[20:11:09];31.089;20507;3;aaa0169;otccom0200(3);Last message received by server daemon: info:3:18578 at time 20:04:16
07-01-23[20:11:09];31.090;20507;3;aaa0169;otccom0200(3);Last exec command:
07-01-23[20:11:09];31.090;20507;3;aaa0169;otccom0200(3);# ackn send:16; # ackn recv:1; # recv:0; # send:0

parallel processing is nothing more than running the same session multiple times for single sets.

The separate sessions communicate by sending 'messages' to the parent-session. Messages are like 'busy' or 'ready'.
When a message is sent, the parent decides what to do. 'Ready' will make the parent sent a new set to process, 'busy' will result in some kind of suspend-action.

In this case the child process seems to have sent it's last message 300 secs ago (timeout). Probably a set is processed in that session that is in a loop or simply is not finished yet (although in that case it should send the 'busy' message)

What do you do when the session crashes ? Restart it or wait untill the next day ?
It might be data related......so when you start it the next day some user might have 'fixed' the problem and the session runs smoothly.

I would check the queue (qbox) to find the message (just a file with some itemdata) and check your system for 'strange' values.
When data is ok, you might want to extend the timeoutvalue....(In my time that was a parameter)

good luck

Thanks very much!!!

When above error occured, we retried it after a few minutes and ended normally.
So I think that threre were no 'strange' values and I should extend the timoutvalue.
But I don't know how to configure it.

Please tell me how to do it.

sorry,
No longer working with Baan... :D so I can't simply lookup the session for you
should be something like cprpd0100s000 (parameters usually are stored in the 000 tables)

Thanks.:)

Your hints made me solve that problems.
We set the parameter bigger COMDLL0200_ACKN_TIMEOUT.
Then, that problem never has occured.

Hello, we are having similar problem. Where did you find this parameter bigger COMDLL0200_ACKN_TIMEOUT ?

Thanks

"BAANPROD: Unable to start application servers 1, 2, 4, 6; timeout of 120 seconds means that " means exactly that Baan was not able to be started for the mentioned servers. In this case, Bshell 3 and 5 were successfully started while, for some reason, the other 4 were not.

Normally, if the error were 1,2,3,4,5,6 then it would for sure mean that in the BW configuration of the server the entry that is set "As Default" is not the correct one or that the password was not saved correctly. In this case, you have to check if the black bold entry is the one that should be run by the Daemon and that it has the password saved.


As in your case, it seems that some bshells were started while others not (I am supposing that your are stating thet error message exactly as it is, i.e. not all bshells), then perhaps the problem is that there were not enough licences left to be used by the other bshells to start.

So please, make sure that the BECs configuration is set correctly. This can be done by entering Baan from the server and running cprrp1210m000 in parallel processing and see if it is able to start 5 more bshells or not.

Now, if the reason that the rest of the Bshells were not started successfully is really the Ttime Out of 120sec(too short), then yes, you can make use of the parameter you mentioned.


How to set up this parameter?

You can add it in the Command line of the Default BECs entry as follows (you may start with a value of 900 for example)

-- -set COMDLL0200_ACKN_TIMEOUT=900

Note: if you have more settings already done in the command line for this entry then you have just to add:

-set COMDLL0200_ACKN_TIMEOUT=900

More Work-Arounds:

There is another variable that can also be used if the error is all about the time allowed before the servers timeout in running multi-bshell (too short). So setting a variable will make it ignore this time out value.

The variable that needs to be set is COMDLL0200_IGNORE_DAEMON and it needs to be set to 1.
So, you can set it up in the command line of your BW configuration as follows:

-- -set COMDLL0200_IGNORE_DAEMON=1

or

-set COMDLL0200_IGNORE_DAEMON=1

Would it be the env.bshell or db_resource in /baan/bse/lib/defaults or somewhere else.

Dave