DBCleanupThread not running in my Schedulix instance

[Alonso Gómez]

Hi,

We have an instance of schedulix server in a Ubuntu server. Schedulix have a lot of master jobs and there are about 5 master jobs that are running every minute along all the day every day. 

Schedulix is sending master jobs and batchs all time... Is that situation the reason to DBCleanupThread is not working?... If it is the reason, how can I force DBCleanupThread work?

Thank you in advance.

[Dieter Stubler]

Hi Alonso,

The job load should not influence the DBCleanupThread.

Is the DBCleanupThread not working at all or just slow ?

To have a look at that I need some more information.

Which version of schedulix do you use ?

Which database system do you use for the schedulix repository ?

Please attach your server.conf (PLEASE REMOVE YOUR SYSTEM PASSWORD AND OTHER CRITICAL INFORMATION !!!)

Regards
Dieter

[Alonso Gómez]

Hello Dieter,

Is the DBCleanupThread not working at all or just slow ?

I think that DBCleanupThread is not working al all, but I am not sure. How can I verify that? At moment I only look for the total of rows in SUBMITTED_ENTITY table and this value is increasing for moments.

• select count(*) from SUBMITTED_ENTITY: 3094054 (2019/06/25)
  
• select count(*) from SUBMITTED_ENTITY: 3099754 (2019/06/26)
  
• select count(*) from SUBMITTED_ENTITY: 3112225 (2019/06/27)
  
• select count(*) from SUBMITTED_ENTITY: 3119677 (2019/06/27)
  
• select count(*) from SUBMITTED_ENTITY: 3127688 (2019/06/28)
  
• select count(*) from SUBMITTED_ENTITY: 3160402 (2019/07/01)
  
• select count(*) from SUBMITTED_ENTITY: 3174484 (2019/07/02)
  
• select count(*) from SUBMITTED_ENTITY: 3200849 (2019/07/04)
  

Which version of schedulix do you use ?

Schedulix 2.8

Which database system do you use for the schedulix repository ?

MySql 5.6

Thanks Dieter.

[Dieter Stubler]

Hi,

when restarting the scheduling server, about 15 minutes later the DbCleanupThread will start doing its work.
So you can have a look in the server logfiles whether there is some error message related to the DbCleanupThread.
Try to set the server trace level to 2 so there is more information logged.

In addition please check the data using this query:

SELECT MIN(FINAL_TS) FROM SCI_SUBMITTED_ENTITY WHERE ID = MASTER_ID and STATE IN ('FINAL','CANCELLED')

Please do this a few days and compare to see if any data has been removed by the DbCleanupThread.

Please let me know the results and any possible error messages ion the logfiles regarding DbCleanupThread.

Regards
DIeter

[Alonso Gómez]

Hi Dleter,

Thank you so much for your help...

We changed the level log and when we restart the schedulix server we saw in logs this lines:

WARNING [main]          08 Jul 2019 09:06:46 GMT Invalid number format in 270000 #0

WARNING [main]          08 Jul 2019 09:06:46 GMT Invalid number format in 3 #0

WARNING [main]          08 Jul 2019 09:06:46 GMT Invalid number format in 50 #0

Can this lines affect to DbCleanThread?

In logs, appear also this lines:

MESSAGE [2,SchedulingThread(Worker0)]   08 Jul 2019 09:57:43 GMT ---------------> Start Cleanup LifeTables   <-------------------

StartTime = 0

MESSAGE [2,SchedulingThread(Worker0)]   08 Jul 2019 09:57:43 GMT purgeLow = 82131185

MESSAGE [2,SchedulingThread(Worker0)]   08 Jul 2019 09:57:43 GMT purgeSetSize = 12

MESSAGE [2,SchedulingThread(Worker0)]   08 Jul 2019 09:57:43 GMT ---------------> End Resource Scheduling   <-------------------

EndTime = 0

Are this lines show that DbCleanThread is running?

Regards.

[Dieter Stubler]

Hi,

Please Have a look in your server.conf file and fix it.

See the warnings in your last post.
Comments after settings in the same line are not supported.

So 27000 # 0 is invalid, that's why DbCleanupThread is doing nothing.
Regards

Dieter