I have installed weblogic 10.3.5 with os unix in the production enviromente.
The managed servers fail with the no health status, this behavior is ramdon and the workaround is restart the managed servers.
Do you know how to solve this problem?
Managed servers are plain ones with custom deployments OR were those configured using any of Oracle standard products e.g. FMW components like SOA, WCC?
Also, in such situation, did you try to check the server state using WLST?
Check if you Managed Server was configured to restart automatically when a stuck thread was detected.
Login wls->Environment->Servers-><Manage Server Name>->Monitoring->health
check "restart with stuck thread".