Omnichannel Analytics vs Current Chats - info won't match

Description:

We compare information from Analytics and Current Chats. Total conversations matches in both cases, but the agents involved are different.

Steps to reproduce:

1. Go to Current Chats.
2. Filter chats by a date, example 1 december 2021.
3. Get info about the agents involved in the conversations.

1. Go to Analyticis.
2. Filter chats by a date, example 1 december 2021.
3. Get info about the agents involved in the conversations.

Expected behavior:

In this example, it's clear that the total number of conversations is 6 (six). So we expect that when converting % of Analytics:

That means that in Current Chats we will find agent "rpacab" with 3.6 conversations, "rsg" with 1.8 and "asesor1" with 0.6.

Actual behavior:

As we can read in Current Chats, real data shows agent "rpacab" with 3 conversations, "rsg" with 3 and "asesor1" with no conversations. Meaning that we can't trust in the % that Analytics is showing to us.

Server Setup Information:

• Version of Rocket.Chat Server: 4.3.1
• Operating System: Ubuntu
• Deployment Method: docker
• Number of Running Instances: 1
• DB Replicaset Oplog: yes
• NodeJS Version: v12.22.8
• MongoDB Version: 4.0.13

Additional context

Not pretty sure, but it version 4.2.1 this issue seemed not to be present.

Start by reproducing the discrepancy in Current Chats and Analytics using the reported date and agent data, then compare how each view counts conversations and assigns agents. Done means both views report consistent agent totals and percentages for the same filtered conversations, with the behavior checked against the supplied Rocket.Chat 4.3.1 setup.