max. resources jobs

[Ron Gates]

We have two automation engines running in our prod environment.  When an agent starts up it connects to one of those engines.

During the day we set the max. resources jobs setting for that agent to 0 so that nothing will run on that agent.

an hour later we bring down the engine that the agent is connected to and it fails successfully to the second automation engine and reconnects just fine... The issue is... when it reconnects it does so losing the max. resources jobs setting of 0.  It starts back up wide open allowing jobs to run on that agent which isn't what we want.  Can you set up the agent so that if if losses connectivity to one engine and reconnects to a second engine it will retain the values it was running with under the other engine? So.. can it retain the max. resources jobs setting of 0?

[Legacy User]

Hi Ron,

When you change the max resource the following message is received:
'U2000046 Settings for Agent 'TEST' have been changed. * These modication are valid until the next restart. *'

So that is the designed behavior and I am not sure of a way to get around this. If you group the jobs into a single queue you could limit the the max slots to 0 before the agent is taken offline.

[Anon Anon]

Maybe try using a script function instead of modifying through the System Overview. SET_UC_SETTING

:SET_UC_SETTING WORKLOAD_MAX, WIN01, 1000

[Luu_Le_185]

Hi RonGates608562

Did either suggestion provided (by Jenn and by Support via your ticket) last week resolved your issue?

Maybe try using a script function instead of modifying through the System Overview. SET_UC_SETTING

:SET_UC_SETTING WORKLOAD_MAX, WIN01, 1000

Change the WORKLOAD_MAX_JOB in the UC_HOSTCHAR_* variables for the Agent(s) in questions.

You can also make a custom UC_HOSTCHAR_* variable for the agent if there are only a few impacted agents. Just be sure to assign the new hostchar variable in the UC_EX_HOSTCHAR variable in client 0 if you elect to make new hostchar variables.

[Ron Gates]

all those work in a "controlled" failover.. but if for some reason the engine on p1 fails and everything fail automatically to p2 then the agent will restart and come back up wide open allowing jobs to run... There should be a setting "warm" start maybe?? that will restart the agent with the settings it had when it went down.. I opened a ticket and automic said to make a suggestion at ideas.automic.com... I have a hard time believing this hasn't bitten someone at some time?  I'll investigate a scripted way of doing this via the hostchar parms.. but still think it should be something provided by Automic by default...

 

[Anon Anon]

It looks like this did bite someone, you opened up a ticket for this same question in September 2012 asking if we had this exact same functionality.

The same answer was provided that the functionality does not exist and the same workarounds were provided.

I would highly suggest reaching out to ideas.automic.com and getting this on the roadmap as our Product Design team is really taking strides to ensure that use cases like these get added to the product.