Downstream Throughput Vastly Diminished

We noticed this morning that downstream throughput for some (not all, oddly) of our ZIA users is vastly diminished to the point of being about 1/10 of what it is with the service turned off. Is anyone currently experiencing this issue or experienced it in the past?

Zscaler’s status page shows all green and no policy changes were made that could explain this (all ZIA users have the same policy).

We do have a ticket open, but thought I’d ask the community here.

I did not get it did you see if with ZIA client connector dissabled on the clients the issue is still there? Did you also check Zscaler Trust site during the issue?

Also from the Insights you can see how long it took for the sandbox to scan a file as maybe the issue was there? Also check the logs for policy procesing time or the client connector logs etc.

It is also good to use Load Balancing for your web traffic with a PAC file:

As you did not mention if you are using the Zscaler app or a gre tunnel:

Also ZDX (if you have it licensed) or using the Zscaler performance tool and Zscaler Analyzer may help you in the future:

https://zmtr.zscaler.com/

Thanks. The issue we had turned out to be our endpoint AV solution that was also performing web reputation checks, including SSL inspection. Once that was disabled, throughput and latency issues evaporated.

1 Like

It is good that the issue is gone. Be carefull with Endpoint systems as we had also slowness issues with malware scanning and dlp on the endpoint agents expecially on developer computers when compilying the code.