Tomoyuki Kawata

I'm currently using versions 2106 and 2103 of Horizon. There's no particular reason for using Horizon Client 8.3, but even now, some thin client terminals come pre-installed with versions like 8.1 or 8.3. By the way, the issue mentioned above occurs in both 2103 and 2106 versions. According to the compatibility table, I don't see any problem whether it's 8.1, 8.3, or the latest version. I posted it not from the perspective of the BU but from the customer's perspective. Because there are actually devices with older Horizon Clients pre-installed, I just wanted to remind to be mindful if unintentionally connected with such protocols.

Hi Matthew It may be more correct to say that the connection changes to Blast when it is rebuilt, even though it is configured for PCoIP. When a thin client terminal is repaired or added to the pool with a version other than 8.3, the connection will be made using the explicit PCoIP protocol, and the terminal that changes to Blast will revert to the connection using the protocol explicitly specified in the pool when horizonClient is upgraded. There may be other causes, but currently the only trigger we have confirmed in the hospital is the Client version. Translated with DeepL.com (free version)

We have confirmed that the connection method from the pool settings is not reflected in some Horizon Clients. When deployed for the first time with Instant Clone, if PCoIP is explicitly specified in the pool settings, the virtual machine will be connected to PCoIP, but some versions of the Horizon Client force the connection to change to Blast connection from the time it is rebuilt. I am only aware of 8.3 and the same phenomenon occurs on different thin client terminals. The previous and following 8.1 and 8.4 versions do not have this problem. If you are using the latest version, you should be fine, but if you encounter this strange phenomenon, please check your Horizon Client version.