vicidial.org

Hi,

i am not sure if this has been reported but we are having some issues with the call length report under inbound/closer calls for this time period. We have a call that has 3062 secs compared to 448 secs if you look under recordings for this time period. We are getting this kind of issues every now and then. other calls seems to be normal enough. please see below screenshots

Vicibox 7.0
VERSION: 2.14-664a
BUILD: 180310-2321
© 2018 ViciDial Group

We've been looking into these strange random rare logging bugs over the last several months and here are some conclusions so far:
- We were able to write a series of SQL queries that helped us to detect where this type of logging issue exists. These queries were fairly complex and can lock the database for several minutes.
- These issues have happened on less than half of the hundreds of systems that we manage
- These issues are quite rare, happening on less than 0.01% of agent events per day in all affected systems.
- These issues happen most often on systems that are overloaded or where agents have poor internet connections(Lots of LAGGED sessions)
- These logging issues are nearly impossible to replicate, we are working on a way to correct the logs as a part of the AST_cleanup_agent_log.pl process when it is scheduled to run each night, but we do not have a firm timeline as to when this will be finished.

Are these issues with "walking away" from the workstations without logging out, or other related login/logout coinciding with the call in question? We found several of those recently.

The issues I was talking about are not related to logging in/out, although the original reporter on this topic might have had a login/out related logging issue, you can't really tell with the information they have provided.

Could this be related to asterisk also? I am using 11.22.0-vici

These are not a walking issue. Agents are always in ready mode for the whole shift and will take pause mode for breaks. Campaign also is an inbound campaign.