Change of Status from Available to offline within minutes in agent workspace in AWA
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
3 weeks ago
Hi Everyone,
When the user changes his status to Available and goes to work in other tab within minutes his status is getting changed to offline state even though he is working.
Please note that this behavior is observed only in PROD instance , but not in Sub PROD instance.
What can be done to rectify this behavior?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
3 weeks ago
Hi Satyanarayana,
The fact that it only happens in PROD is the important bit - AWA presence in workspace is driven by the AMB (WebSocket/Comet) heartbeat from the browser, so when that channel drops the agent gets flipped to Offline even though they're clearly still working. Sub-prod usually isn't sitting behind the same network path, which is why it looks fine there. A few things to confirm:
1. Do the agents reach PROD through a corporate proxy, VPN or load balancer that isn't in front of sub-prod? Anything that terminates idle WebSocket connections after a minute or two would explain this exactly.
2. Is the status going to Offline, or to a custom "Away"-type presence state? And is it the same for all agents or only some (e.g. one office / one network)?
3. Roughly how many minutes - is it consistent (say always ~2-3 min) or random?
4. Have you compared the AWA / presence related system properties between PROD and sub-prod, in particular the presence inactivity / heartbeat timeouts, and any recent PROD-only plugin or patch update?
5. If you open the browser dev tools Network tab on an affected agent's session in PROD and filter on amb, do you see the long-poll / websocket requests failing or being closed?
Happy to help narrow it down further once you have those.
