- HMC V7R220.127.116.11 with public IP in routed subnet A
- HMC ports of servers some in subnet A and some in other routed subnet B (all needed firewalls are opened for HMC management communication of servers in subnet B -- are "connected" on lssysconn output and can even install VIO server from "nimol" via hmc).
- some LPAR (AIX) IPs are in subnet A and some in subnet B (port 657 is allowed on firewalls beetween HMC IP and LPARs IP being in subnet B)
- lspartition -dlpar returns nothing on HMC (no server/LPAR is seen DLPARable)
- HMC status is OK (green)
But although all required ports are opened I always get below warning/error when I try to DLPAR from HMC:
"A RMC network connection to the partition is not present. Verify the network settings on the partition and the HMC. If you select OK you will have to restart the partition for the resource changes to take effect."
The problem exists also for servers with are configured in the same LAN HMC is.
restart of rsct_rm/rsct SRC subsystem group doesn't help.
I can locally telnet 657 on LPAR and 657 port on HMC from the LPAR.
Also, when I stop rsct* services on LPAR and start "sshd -p 657", then I can ssh access LPAR from HMC on port 657...
Servers microcode levels is up_to_date.
There are different AIX on LPARs (5.3, 6.1, 7.1).
None of tips described on https://www.ibm.com/developerworks/mydeveloperworks/blogs/cgaix/entry/february_24_2011_6_51_pm79?lang=en helpded.
i thought /usr/sbin/rsct/install/bin/recfgct should help but it doesn;t (I see ater it is run, rmc subsystems are restarted and /usr/sbin/rsct/bin/lsnodeid shows new noweid string...)
All the time "lssyscfg -r lpar -F name,rmc_state -m a_servername" shows for each listed LPAR rmc_state "none".
i there any known bug in HMC V7R18.104.22.168? or maybe the rsct services should also be reconfigured on HMC as well? if so, how to do it?