Possible network problems detected - check tracefile and adjust the DNS settings

Which user did you start SAP with,
adm  or  ora

Are you on Windows or Unix?.
WARNING => DpNetCheck:

NiServToNo(sapdp12) failed (rc=-3)
ERROR => DpNetCheck:
NiServToNo(sapgw12) failed (rc=-3) [dpxxtool2.c  3733]

LOG GZZ=> 2 possible network problems
detected - check tracefile and adjust the
             
DNS settings [dpxxtool2.c  3873]
Do as advised and check dev_w0, there seems to be
some name service issue...

3733]
the
Do as advised and check dev_w0, there seems to be some name service issue...

I use IBM AIX v5.2
First of all i start lsnrctl with ora
Then i do startsap with adm

3733
adjust the
>> DNS settings [dpxxtool2.c 3873]
I have SAP in IBM HACMP environment. There are two nodes. One is CI-node, the
other is DB-node.
So, there can be three variants:
1. The R/3 is running on it`s home node (CI). Oracle db is also running on it`s
home node (DB)
2. The CI node has failed. All of it`s resources (including it`s service ip
label) are acquired by the DB node and R/3 is started there. Database is already
running on DB
3. The DB node has failed. All resources are on CI node. The database is
started. The R/3 is already running there.
In first case when R/3 starts, it says that ABAP database is not avaliable via
R3Trans. Oracle is running on DB. R/3 doesn`t start on CI normally (there are
few processes, like co.*, ms.*, but no dw.* processes).
In the second case, ABAP database is avaliable, but R/3 is not running normally
(again, no dw-processes). I try to "stopsap" and "startsap" manually, but there
is no result.
In the third case everything goes well. All processes are started. Lots of
dw-processes, and so on.
During takeover (DB to CI and vise versa) the takeover node logically has two
machines. Both it`s and other node`s file systems are mounted, two ip aliases
exist. The most srange thing is that when DB falls to CI everything goes ok, but
when CI falls to DB, dw processes do not start. The only difference is the
machine`s hostname. But i tried to change the DB`s hostname to CI`s. But it did
not help.
PS. I don`t use DNS. I use /etc/hosts instead. The name resolve and ip resolve
works good. The R/3 was installed on CI machine. Maybe this is the reason...
Maybe R/3 searches for some processor`s s/n...
Don`t know. Let me know, if you have any ideas.

Hi There,
I think I experienced this problem before
and we’re on AIX too.
Can I ask u if you experience this problem:  -  Unable to
“mount” filesystems from other host with Subsystems error when reboot
server.              
-  Found errors for Ethernet Down in command “errpt”
If so try this:-
1.stopsrc  -g nfsß   stop all Subsystem for NFS
2.lssrc -g  nfs       ß  should say all Subsystem
“Inoperative”
3.startsrc -g  nfs
4.lssrc -g  nfs   ß  should show “active”
5.smitty inetà Show/Change characteristics of a Network Interface à detach for (en1, et0, et1)
6.netstat -a ß  make sure ent1 not display
7.inetstat -i ß  Should show ONLY en0, lo0 after detach…we only use en0 node !
8.shutdown -Fr ß  reboot server
Hope this helps.  Please let me know
if this was the problem.
Derek Phung
  WARNING = DpNetCheck:
NiServToNo(sapdp12) failed (rc=-3)
  ERROR = DpNetCheck:
NiServToNo(sapgw12) failed (rc=-3) [dpxxtool2.c  3733]
 LOG GZZ= 2 possible network
problems detected - check tracefile and adjust the           
DNS settings [dpxxtool2.c  3873]
Do as advised and check dev_w0, there seems to
be some name service issue...
I`ll explain the situation:
I have SAP in IBM HACMP environment. There are two
nodes. One is CI-node, the other is DB-node.
So, there can be three variants:
1. The R/3 is running on it`s home node (CI).
oracle db is also running on it`s home node (DB)
2. The CI node has failed. All of it`s resources (including
it`s service ip label) are acquired by the DB node and R/3 is started there.
Database is already running on DB
3. The DB node has failed. All resources are on CI
node. The database is started. The R/3 is already running there.
In first case when R/3 starts, it says that ABAP
database is not avaliable via R3Trans. Oracle is running on DB. R/3 doesn`t
start on CI normally (there are few processes, like co., ms., but no dw.
processes).
In the second case, ABAP database is avaliable,
but R/3 is not running normally (again, no dw-processes). I try to
"stopsap" and "startsap" manually, but there is no result.
In the third case everything goes well. All
processes are started. Lots of dw-processes, and so on.
During takeover (DB to CI and vise versa) the
takeover node logically has two machines. Both it`s and other node`s file
systems are mounted, two ip aliases exist. The most srange thing is that when
DB falls to CI everything goes ok, but when CI falls to DB, dw processes do not
start. The only difference is the machine`s hostname. But i tried to change the
DB`s hostname to CI`s. But it did not help.
PS. I don`t use DNS. I use /etc/hosts instead. The
name resolve and ip resolve works good. The R/3 was installed on CI machine.
Maybe this is the reason... Maybe R/3 searches for some processor`s s/n...
Don`t know. Let me know, if you have any ideas.