Skip to Content

Archived discussions are read-only. Learn more about SAP Q&A

Cached /etc/services

Hi Guys,

Do anyone know how to reset the cached ports ( /etc/services) in SAP without restarting sap ?

RFC connection doesn't work from one of the application servers when we use load balancing method.

Issue :

1) in one of the application server , the entry was wrongly made in the past

sapmsBIP 3200/tcp ---> This is wrong

2) I had changed this to 3600

sapmsBIP 3600/tcp

3) It works perfectly from OS level

XXXXXX:bepadm 67> lgtst -H <hostname> -S sapmsBIP

using trcfile: dev_lg

list of reachable application servers


[XXXXX_BIP_00] [XXXX] [] [sapdp00] [3200] [DIA UPD BTC UP2 ICM ]

[XXXXX_BIP_00] [XXXX] [] [sapdp00] [3200] [DIA UPD BTC UP2 ICM ]

[XXXXX_BIP_00] [XXXX] [] [sapdp00] [3200] [DIA UPD BTC SPO UP2 ICM ]

list of selectable logon groups with favorites


[BI Production System] [XXXXXX] [3200] [730]

[BI_PARALLEL] [XXXXX] [3200] [730]

[SPACE] [XXXXX] [3200] [730]

nlsapapp29:bepadm 68>

4) Still SAP thinks , it has to reach 3200

M Wed Jan 14 12:10:19 2015

M ***LOG Q0I=> NiIRead: P=XXXXXXX:3200; L= recv (104: Connection reset by peer) [nixxi.cpp 5087]

Actions done:

SM51 - host name buffer reset  ( application serer , ICM and Entire system )

SMMS hostname buffer reset

Name cache daemon in linux is also not running ( to avoid the caching )

RFC test :

LogonConnection Error
Error DetailsError when opening an RFC connection (LB: Hostname or service of the message ser
Error DetailsERROR: internal error
Error DetailsLOCATION: SAP-Server XXXXX_BEP_21 on host bepas1 (wp 4)
Error DetailsCOMPONENT: MS (message handling interface)
Error DetailsCOUNTER: 267
Error DetailsMODULE:
Error DetailsLINE:
Error DetailsRETURN CODE: -101
Error DetailsSUBRC: 0
Error DetailsRELEASE: 720
Error DetailsTIME: Wed Jan 14 12:10:24 2015
Error Details


I am very positive the restart will solve as per forums and notes etc., But still wanted to try without SAP restart if possible:-)



Former Member
Former Member replied

Hi All,

Parameter settings  + SM51 Hostname buffer cleanup ( entire system ) + $sync  helped

In multiple app severs situations, run $sync in all app serves ( and CI ).

parameter settings : set to 0

  • is/hostbuffer_timeout_invalid_entry
  • is/hostbuffer_timeout_valid_entry

In worst case ( sometimes happens :-) ) , restart the affected app server. If you have a really bad day and if u have it on CI , Happy restarting the complete systems.


0 View this answer in context

Helpful Answer

Not what you were looking for? View more on this topic or Ask a question