Originally posted by: PontiacGeronimo
hello,
I have setup NIM server 7100-03-05-1524 but have one isse with mksysb of client in different subnet (no ports are blocked beetween NIM and that subnet, in both directions)
So, I can successfully add each new nim client using niminit on client but mksysb started from nim hangs now... and strange is that 1 (one) mksysb passed, but only one and don't know why it stopped works.
No resolving problem, systems are up-to-date with latest TL/SP.
After unsuccessfull mksysb, in nimlog I can see:
0042-001 nim: processing error encountered on "master":
0042-006 m_mkbosi: (From_Master) connect A remote host did not respond within the timeout period.
nconn: connect() failed, errno is 78
aclientname: A remote host did not respond within the timeout period.
What I observed on NIM server, is that nimsh does not work by default - is that ok?
# lssrc -a|grep nim
nimesis nim 3932296 active
nimsh nimclient inoperative
nimd inoperative
and the only what was autoconfigured in inittab is:
# grep nim /etc/inittab
nim:2:wait:/usr/bin/startsrc -g nim >/dev/console 2>&1
is that ok?
Another strange thing is that when I run command "nimclient -l -t lpp_source" on client in loop, it hangs after some try:
# while true;do echo $(date) :: $(nimclient -l -t lpp_source|grep lpp_AIX-6100-09-04);sleep 5;done
Wed Jul 1 08:02:16 CUT 2015 :: lpp_AIX-6100-09-04 resources lpp_source
Wed Jul 1 08:02:22 CUT 2015 :: lpp_AIX-6100-09-04 resources lpp_source
Wed Jul 1 08:02:27 CUT 2015 :: lpp_AIX-6100-09-04 resources lpp_source
Wed Jul 1 08:02:33 CUT 2015 :: lpp_AIX-6100-09-04 resources lpp_source
--- and I have to CTRL-C
0042-002 nimclient: error signal number 2 received -
processing aborted
# date
Wed Jul 1 08:03:37 CUT 2015
and tcpdump on client shows finally some reset:
08:10:58.324371 IP nimserver.1023 > client.3901: R 1:1(0) ack 1 win 65535
All these problems happens does no matter nimsh is running on NIM server or not.
and I don;t know why... because the port is listening on client:
# lsof -P -i tcp:3901
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
nimsh 4391120 root 5u IPv6 0xf1000e00002d0bb8 0t0 TCP *:3901 (LISTEN)
and even stopping nimsh temporaily there and starting test port on 3901 it works from nimserver:
on nim client:
# stopsrc -s nimsh
0513-044 The nimsh Subsystem was requested to stop.
# lsof -P -i tcp:3901
# nohup ./netcat_gnu -l -p 3901 &
[1] 786884
# lsof -P -i tcp:3901
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
netcat_gn 786884 root 5u IPv4 0xf1000e00005993b8 0t0 TCP *:3901 (LISTEN)
on NIM server:
# tn client 3901
Trying...
Connected to client.
Escape character is '^T'.
So why during mksysb conenction is reset?
By the way, - can nimsh be left stopped as it is by default on NIM server? if not, how it should be started during boot automativally, when the line is not added now to inittab? maybe not necessary?