Looks like I'm using 1.1.1. I'll try downloading the newest version and see if 
that helps.

TC Tobin-Campbell | Technical Services | Willow | Epic  | (608) 271-9000

From: Karl Wright [mailto:[email protected]]
Sent: Friday, May 24, 2013 9:56 AM
To: [email protected]
Subject: Re: Jobs stuck in a Terminating status

Hmm.  An output connector is not supposed to report an error if the document it 
is deleting does not exist.  The JSON looks fine too.  What version of 
ManifoldCF are you using?  I believe I fixed just such a problem for 1.2.
Karl

On Fri, May 24, 2013 at 10:53 AM, TC Tobin-Campbell 
<[email protected]<mailto:[email protected]>> wrote:
In the Simple History, I'm seeing  a number of errors that look like this:
Start Time           Activity                Identifier            Result Code  
     Bytes     Time      Result Description
05-17-2013 17:02:34.683                Deletion (F Drive ES)      
file:/F:/CS/Client/HSHS/Tech/Good%20Maintenance/Backup%20of%2...
0HSHS%20GMP%20Checklist%205-3-2012.xlk
ERROR  0              2                
{"ok":true,"found":false,"_index":"docs","_type":"file","_id":"file:/F:/CS/Client/HSHS/Tech/Good%20Maintenance/Backup%20of%20HSHS%20GMP%20Checklist%205-3-2012.xlk","_version":19228}

This looks to me like we're unable to delete a document in Elasticsearch since 
it doesn't exist. I'm wondering if maybe creating this document in 
Elasticsearch will allow the job to then delete is cleanly and close?

TC Tobin-Campbell | Technical Services | Willow | Epic  | (608) 
271-9000<tel:%28608%29%20271-9000>

From: Karl Wright [mailto:[email protected]<mailto:[email protected]>]
Sent: Friday, May 24, 2013 9:27 AM
To: [email protected]<mailto:[email protected]>
Subject: Re: Jobs stuck in a Terminating status

The "Terminating" state is the part of the job cycle where ManifoldCF attempts 
to remove documents from the index that have been abandoned by the job.  If the 
output connector is not able to remove jobs from the index, the job will remain 
in the "Terminating" state until it *can* remove them.  (There are other 
situations where this can happen, such as when you uninstall the output 
connector during the Terminating phase.)
The best way to tell what is happening is to look at the Simple History.  That 
should give you a clue.  As for "killing" the job, how you do that will depend 
on what's wrong.

Karl

On Fri, May 24, 2013 at 10:16 AM, TC Tobin-Campbell 
<[email protected]<mailto:[email protected]>> wrote:
What's the best way to kill a job? I've got two jobs stuck in a status of 
Terminating that are clogging up the log file. I'm using the single process 
example that comes with the download.

[cid:[email protected]]

TC Tobin-Campbell | Technical Services | Willow | Epic  | (608) 
271-9000<tel:%28608%29%20271-9000>


<<inline: image001.png>>

Reply via email to