ESP Workload Automation

Does anyone know what causes a WOBTRIG Failure?  Is there a way to prevent these from happening?  Any ideas on why some Wobtriggers will not reset (I have this in QA). 

I currently run a job 3 times a day (once a shift) where I dump the Events file to a flat file and search for any WOBTRIG Failures.  If any are found, I send an email to appropriate people to have the event reset (State monitored, status monitored for create).  I'd like to know what causes them.  Thanks!

Which scheduler are you using?  If it is ESP, there may be an eaiswer way to get your notifications out.  Also, can you give a few more details about the WOBTRIGs?  Are they watching files on distributed agents? If so, what OS?

Hi Barb, 

The biggest reason that a WOBTRIG fails is due to a "Scan Failed" error. This is usually caused by the PATH to the file not being available. This is different than the file not being available. In the case where the PATH is not available, the agent can't even get to folder to start looking for the file. This "Scan Failed" error can be caused by various things like brief network outages or short NAS issues. 

It is possible to add a MAILBOX to the event so it will send out an E-Mail to a group when the event goes into "Scan Failed".  Let me know if you need help with this. 

Another option is to add a retry and a retry period to the agent. The default is 1 so the agent only tries 1 time before failing. It is important to only keep the number of retries and the period to a minimum. I think it is relatively safe to add the retries to the agent especially if the agent does not have many file triggers but there are potential issues. Below are my notes on this topic. 

filemonplugin.scan.retries=3                       (default is 1)                                      

filemonplugin.scan.retries.sleep=30000      (30 Seconds)         

ISSUES:

Let's say the agent has 20 file triggers.

The scan retries is 1 time per minute for 5 minutes.

PROBLEM 1:

If the file trigger is not able to reach the directory it will go into retries mode. It will fail after 5 minutes

The next file trigger scan will wait for the above step to complete.

They will single thread through.  If there are many file triggers some could end up waiting a long time. 

PROBLEM 2:

Someone changes the password for the userid in ESP and in the NAS

The file triggers do not get passed the new password.

The first one fails and retries for 5 minutes.

The next file triggers  scans wait in single file for the first one to exhaust the 5 minutes of retries

 If there are several / many file triggers on the agent the retries for all of them will take a long time.

 Then the file triggers can be retriggered. They will then get the new password and they will be successful .

2¢

Don Powell

I'm verifying things right now and will let you know shortly.

Broest,

For the record, we were having this exact problem over the summer.  We opened a ticket to CA, and the advice that Don has given has definitely worked for us. 

Also, for the record, you can add email notifications to WOBTRIG events by adding MAILBOX(NAME) to the event.  Of course, you'll have to update your mail list and load it before it will work, but here's an example:

EVENT ID(ESPPROD.ZE450W01) MAILBOX(OPS)  CLASS(PRODJOBS) -             
       SYSTEM(MASTER)  REPLACE                                         
CALENDAR SYSTEM                                                        
INVOKE 'SS.CAI.ESP.PROD.APPLLIB(ZE450W01)'                             
WOBTRIG FILE_TRIGGER Agent(WP000577) -                                 
        FileName('\\BCFIL024\is\Claims-Systems\2-Operations\App5-Payme-
nts\EncounterPro\Resubmissions\*') CREATE State(Monitored) -           
        Status('Monitored for CREATE')                                 

I apologize for my delay in responding --- crazy day yesterday.

I will add the mailbox to my event.

If I add the retry:

filemonplugin.scan.retries=3                       (default is 1)                                      

filemonplugin.scan.retries.sleep=30000      (30 Seconds)   this is added to the agent, not somewhere in the event?  If so, then the retry is done on any WOB trigger in any event referencing that agent?

If a retry is done, does ESP automatically reset the  Status and State or does someone still need to manually do this?

Again, thanks in advance for your help - it is most appreciated!     

Hi, 

Those two statements go in the agentparm.txt file. I add them near the bottom in the section that refers to the plugins.  The agent will then need to be bounced. 

You are correct. The retry will be done for all file triggers for this agent. 

The event will not fail until the retries are exhausted. In the example earlier the event will try 3 times over a 90 second interval before failing. This will prevent the event from going into "Scan Failed" if the outages are brief. If 90 seconds is too short you could extend the time a little.

Let me know if you have any other questions. 

Feel free to contact me directly at 214-473-1191

Don

2¢ 

Many thanks!  I will talk with out server guys to see if they will set this up for me.  It will be next week though due to the holidays.  I will let you know how this turns out - again, thanks!

Hi Barb,

Have you got the answer you need? If yes, will you mark this as "Answered"?

Thank you,

Lucy