The problem you've never seen and never want to

[gaidem]

First off, sorry to the community members. I've been super busy these days and haven't been able to respond to posts as much as I would like. Here's an example of what's been keeping me busy.

Yesterday (Sunday) I get notified that our production instance crashed. For about 7 hours the instance was extremely slow and had to be rebooted almost 10 times. So what happened?

We received a spam ticket with CC's that ended up adding several donotreply addresses on the watch list and created a loop. This alone wasn't new to me. However, each of the bounce backs from the addresses were HUGE in size. For example 1 was 357 pages in length and over 1.5 million characters.

The Incident record became so big that any query related to it would deplete all the system resources. I don't mean opening the record, I mean referencing tables like sys_email that had a reference to this record.

Because of this I have developed a preventative solution for the future. This solution will prevent updates to a ticket once the specified journal field character count exceeds a new property I have created:

Create a new field on the task table: u_journal_count
Create a new system property:
Name: task.journal.limit
Type: integer
Value: Specify the amount of characters you want to limit (-1 equals no limit)

Now create a before insert/update business rule on the task table:
Condition: current.work_notes.changes()
Script:

var limit = gs.getProperty('task.journal.limit');
if(limit!='-1'){
var journalCount = current.u_journal_count;
var wrkNotes = current.work_notes;
var wrkNotesSPLT = wrkNotes.split('');
if((wrkNotesSPLT.length + journalCount)> limit){
current.setAbortAction(true);
gs.addInfoMessage("The Client Communication field has exceeded its journal limit");
}
else{
current.u_journal_count += wrkNotesSPLT.length;
}
}

Regards,