Hey All,
We have a horizon deployment on Vmware on AWS, we have 2 appvolumes v4 servers and we are currently load balancing but we are having some issues.
We have a Application Elastic load balancer setup in AWS. have a security group on the ELB to only accept 443 incoming from the internal networks using it. There is 1 listener on the ELB for 443, we are using a wildcard cert so all good there, we are forwarding to a "target group" with the 2 ip addresses of the Appvolumes servers, both show as healthy. We have the following attributes configured (stickyness : Enabled, Stickyness type: lb_cookie, Deregistration delay: 10 seconds, Stickyness duration:20 hours, slow start duration:0 seconds).
now it seems to work about 80% of the time, but every so often we get a user who gets the "connection error/Virtualization is disabled" message and no volumes attach.
The relevant part of the logs show this
[2021-02-10 19:15:31.033 UTC] [svservice:P1576:T6068] HttpSendRequest: WinHttpQueryHeaders returned status code 401
[2021-02-10 19:15:31.166 UTC] [svservice:P1576:T6068] ReadHTTPResponseHeader: WinHttpQueryHeaders failed (WWW-Authenticate) header not found
[2021-02-10 19:15:31.166 UTC] [svservice:P1576:T6068] HttpSendRequest: WWW-Authenticate header could not be read
[2021-02-10 19:15:31.166 UTC] [svservice:P1576:T6068] User login over HTTP failed
[2021-02-10 19:15:31.166 UTC] [svservice:P1576:T6068] HttpUserLogin: failed (user login)
Im wondering if anyone is using a ELB in AWS and what configuration you are using?. im currently just pointing the Cname for the server (which was previously the ELB) to just one server now till i get this fixed. ive pointed it at each server for a few weeks each and both are fine when not going via ELB.
Thanks,.
