Has anyone else seen poor performance issues since upgrading to 6.1 or 6.1.1? We have user complaints coming in across the board (local and branch offices across all pools) of latency like issues (choppiness, slowness, lag etc...) that resemble PCoIP latency or loss type issues. I have used vROps to look at the environment but I can't see anything that stands out. View hosts resources, connection server, security server, persona repo, SQL server resources, SAN all look good. I can't seem to put my finger on it.
Have you tried using the Session Stats Viewer from Teradici to look at the session logs?
No I have not. I will give it a shot and let you know the results. Thank you.
I am running the PCoIP Session Stats Viewer in a VDI machine but I have no idea how to interpret this data.
I just wanted to let you know I just upgraded to 6.1.1 from 6.1 but none of my users or myself have noticed any latency issues. I hope you can figure out where it's coming from!
Are you using an apex offload card? We just implemented a completely new infrastructure based on hp dl380 g9, esxi 6.0 and view 6.1 and we are also experiencing this issues when activating the apex 2800 offload card.
No we are not. After talking to a VMware escalation engineer yesterday, he had me uncheck the option to secure tunnel via the connection servers in View admin. I personally think it has to do with a botched connection server install. More to come.
Can you upload the agent logs fro one of the affected desktops, a good indication is in the PCoIP server logs, you can search for packet loss.
Teradici state that anything over 0.01% loss will cause the type of symptoms you are seeing. See below:
Does packet loss impact the PCoIP protocol? (15134-739)
Another thing to check is if you have the tools and agent installed the correct way around, it is recommended to install tools first before the agent,
something got to do with tools overwriting the SVGA drivers that the agent installs, if done the wrong way. This can also cause issues...
Are you using the VMxnet 3 adapter and installed all the hotfixes from Microsoft? another thing to rule out i suppose.
Thanks Steinb3rg. I did look at the agent logs with VMware support and yes there was in fact PCoIP loss. I also did install the VMtools before installing the 6.1.1 agent as recommended by VMware as well as applying the MS KB for the VMXNET3 nic. Turns out we had PCoIP secure gateway checked off in the View Admin under one of our connection servers that was the cause of the issue. Apparently, this is a default option although VMware support could not tell me why other than "they have been talking to the dev team about this". So, everything internally was being securely channeled via the connection server which was slowing things down massively! All this time it was a single check box causing the issue!
Just to be sure... You unchecked use PCoIP Secure Gateway for PCoIP connections to desktop?
I'm currently running 5.1, about to upgrade to 6.1. Currently have Use Secure Tunnel and use PCoIP Secure Gateway UNCHECKED for internal connection servers.
Yes that's correct bjohn.
Thanks, I guess I will leave both options unchecked.
I definitely would! Also, if you're upgrading to 6.1.x I hear 6.1.2 is right around the corner FYI.
I want to make sure I have this correct. For the past 10 months we have been running View 6.0 and our environment has been running great, no slowness or latency. Just a few days ago we upgraded to View 6.1.1. Below are the settings for our single connection server. I am assuming these were the same settings we had under View 6.0 but of course the upgrade to View 6.1.1 could have checked the box "Use PCoIP Secure Gateway for PCoIP connections to machine". Is this the checkbox that VMWare support says should be unchecked?
Try to disable temporal image caching with an Apex, it helped for us..
1. Browse to HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Teradici\PCoIP\pcoip_admin.
(If you do not see the 'Teradici' folder and subsequent sub-folders, you will have to create them.)
2. Add new DWORD Value and name it pcoip.enable_temporal_image_caching, and modify the value to decimal 0.
Can you link the MS KB for the VMXnet3?
Thanks for your advise. We already tried that on the Teradici supports advise.
Unfortunately there is no improvement concerning the performance...
i am also having this issue. the connect using secure over pcoip is already unchecked. i started having this issue after upgrade from 5.3.2 to 6.1.1. i started a SR. will post back what corrects this.
related post? : Urgent Help!!! one of my client meets - "the view agent reports that this desktop source is unable...
i resolved this issue by removing agent 6.1.1/6.1.0 and using agent 6.0.2.
i was told that support would inform the users here, great support vmware.
the users are back to being their normal crabby selves.