[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference smaug::snagwy

Title:	SNA GATEWAY NOTEFILE
Notice:	Note 1.* -> kits and doc, 288.* -> obtaining product support
Moderator:	EDSCLU::GARROD

Created:	Fri Feb 07 1986
Last Modified:	Fri Jun 06 1997
Last Successful Update:	Fri Jun 06 1997
Number of topics:	7116
Total number of notes:	28576

7115.0. "LU6.2 V3.1 FUNCABORT defining local LU" by BARNA::DSMAIL () Wed Jun 04 1997 15:22

    Hi,
    
    I've installed LU6.2 V3.1 version copied from snads and I'mdoing some
    testing.
    
    We have a running configuration in a node with Unix 3.2C, Peer Server
    V1.3-2 and LU6.2 V3.0-7. We are trying to declare same definitions in
    another node with Unix 4.0B and LU6.2 V3.1. SNA gateway is the node
    with Peer Server.
    
    We run lu62_config, connect to lu62_server and define local LU. Data is
    as follows:
    
    
    SELECT_SERVER{
    	server_name=LU62_SRV,
    	server_node=morvedre,
    	server_transport=TCPIP
    	}
    DEFINE_LOCAL_LU{
    	fq_local_lu_name=NETCAV01.COPADND1,
    	local_lu_alias=copadnd1,
    	lu_session_limit=0,
    	session_activation=false,
    	max_sync_level=confirm,
    	gateway_node_name=alfaib,
    	gateway_transport=TCPIP,
    	lu_capability=primary_or_secondary
    	}
    
    
    At this moment we receive an error message:
    
    %SNALU62-E-UNSUC, LU6.2 verb did not execute successfully
    -SNALU62-E-COMERR, connection to server has failed
    -SNA-E-FUNCABORT, access routine function aborted
    
    
    and lu62_server dies...What does it mean?
    
    Any help will be appreciated. Regards,
    Herminia Codina.
    Barcelona

T.R	Title	User	Personal Name	Date	Lines
7115.1	May be a known problem	EDSCLU::GARROD	IBM Interconnect Engineering	`Wed Jun 04 1997 16:18`	203
	Are you sure that the LU6.2 server is dying. If so please find the core dump in /core and do a: dbx /usr/sbin/lu62_server /core on it. If the server is crashing then this isn't the problem reported below. But if it is not crashing check to see if the problem you are having is the one below. Dave <<< SMAUG::EDS1$:[NOTES$LIBRARY]IBM_IC_SUPPORT.NOTE;1 >>> -< IBM Interconnect Support >- ================================================================================ Note 408.0 lu6.2 V3.1 configuration failure 9 replies TAV02::SHEFTEL 26 lines 2-JUN-1997 09:37 -------------------------------------------------------------------------------- Hi, I'm trying to configure SNA/APPC LU6.2 V3.1 over SNA/Peer Server 1.3. The connection to the server is fine but when I try to define 14:24:33.73 snalu62_define_local_lu exiting with status %SNALU62-E-UNSUC, LU6.2 verb did not execute successfully Client logmask shows no additional info. Also server logmask doesn't seem to have anything to add. 14:24:33.73 snalu62_define_local_lu exiting with status %SNALU62-E-UNSUC, LU6.2 verb did not execute successfully I'll put server logmask, cloent logmask and ncl in the deposit area by the name NATA_LU62V31.bck I've done it in 2 sites . One was Unix 4.0 and Peer 1.4 . Second one was 3.2G and Peer server 1.3. Both experienced the same problem. Appreciate your help Thanks Anat ================================================================================ Note 408.1 lu6.2 V3.1 configuration failure 1 of 9 TAV02::SHEFTEL 0 lines 2-JUN-1997 09:40 -< Escalated to EDS >- -------------------------------------------------------------------------------- ================================================================================ Note 408.2 lu6.2 V3.1 configuration failure 2 of 9 TAV02::SHEFTEL 1 line 2-JUN-1997 09:42 -< misspelling, saveset name ANAT_LU62V31.BCK >- -------------------------------------------------------------------------------- ================================================================================ Note 408.3 lu6.2 V3.1 configuration failure 3 of 9 EDSCLU::GARROD "IBM Interconnect Engineering" 10 lines 2-JUN-1997 10:06 -< Set up the server security >- -------------------------------------------------------------------------------- I took a quick look at the logs. I think the reason for the failure is that you have not set up the security properly to allow you to manage the server. Refer to the documentation on how to properly set up the security. I don't know why there isn't a secondary error message saying "authentication failure" or something. Dave ================================================================================ Note 408.4 lu6.2 V3.1 configuration failure 4 of 9 EDSCLU::GARROD "IBM Interconnect Engineering" 13 lines 2-JUN-1997 12:15 -< Does your host have a name? >- -------------------------------------------------------------------------------- I just had a further look at the code. Security violations should have suberrors so I now doubt that is the reason. The only straight UNSUC (with no suberror) I could find in this area of the code is if the system doesn't have a hostname defined. Enter the command: hostname on the LU6.2 server system. What does it say? If it doesn't say anything then here is your problem. Dave ================================================================================ Note 408.5 lu6.2 V3.1 configuration failure 5 of 9 TAV02::SHEFTEL 20 lines 2-JUN-1997 12:18 -< I've created and edited the lu62_amd.access >- -------------------------------------------------------------------------------- Hi Dave, Thanks for the quick answer. I did what is mentioned in the docs. I've set up a file named /usr/lib/sna/lu62_amd.access(I hope I remeber the name correctly) and set it to be nodename::TCPIP::LU62_SRV::root (Just followed the example). There is nothing more about the security. I'be tried to rename the file to another name to see if there is any other code but it was the same. I did'nt find anything regarding security exept this file. If I missed anything, I'll be more then happy to know what. Thanks Anat ================================================================================ Note 408.6 lu6.2 V3.1 configuration failure 6 of 9 TAV02::SHEFTEL 9 lines 2-JUN-1997 12:20 -< hostname exists >- -------------------------------------------------------------------------------- Dave, You were quicker then me. I've checked that too. hostname gives the correct name which is dodo in this case. dodo aslo appears in the /etc/hosts. I have no domain name so just used simple name. Anat ================================================================================ Note 408.7 lu6.2 V3.1 configuration failure 7 of 9 EDSCLU::GARROD "IBM Interconnect Engineering" 17 lines 2-JUN-1997 13:18 -< I've found it >- -------------------------------------------------------------------------------- Re .-1 Well I thought about this over lunch and came back with a theory. Since proved correct. And your latest reply confirms it. You'll get the exact error you are seeing if your TCP hostname doesn't ave a "." in it ie it has no domain or sub-domain. It looks like oversite in the code. I'm going to log this preoblem in our bugs file as bug #266 for addressing in a future ECO and/or release. The only workaround id for you to give your host a name with a . in it for the moment. I don't think we'll be issing an ECO just for this problem. Dave ================================================================================ Note 408.8 lu6.2 V3.1 configuration failure 8 of 9 TAV02::SHEFTEL 14 lines 4-JUN-1997 09:14 -------------------------------------------------------------------------------- Hi Dave, This was it. Thanks Small question. In /usr/lib there is snatprm.a file that has entries of all the snalu62_routinexxx routine names but without the snalu62_ prefix. How is the connection done between the call in the program and this lib. Thanks Anat ================================================================================ Note 408.9 lu6.2 V3.1 configuration failure 9 of 9 EDSCLU::GARROD "IBM Interconnect Engineering" 29 lines 4-JUN-1997 09:30 -< You shouldn't be using that file >- -------------------------------------------------------------------------------- You really shouldn't be referencing this library. Your programs should be compiled: cc ... -lsnatprm which will case /usr/shlib/libsnatprm.so to be used. The reason you shouldn't be using the static library (libsnatprm.a) is because if newer versions of the APPC product are installed you won't automatically pick up the fixes without a relink. By default libsnatprm.so will be used unless you explicitly do cc -non_shared which will use libsnatprm.a (which I highly advise against). Unfortunately if you are forced to use the ld command the default there is to unfortunately to produce a static image so if you use ld you should use the command: ld -call_shared In regard to your question I can't see the symnols you are referring to. I can see all the symbols with the right names ie: snalu62_routinexxx Dave
7115.2	server crashes	IB004::CODINA		`Thu Jun 05 1997 12:44`	16
	Hi, server is crashing and generates a core. We did: # dbx /usr/sbin/lu62_server /core dbx version 3.11.10 Type 'help for help. Core file created by program "lu62_server" thread 0xd signal Segmentation fault at <*[strncpy, 0x3ff800d2610] ldq_u8, 0(r16) Any more testing? Regards, Herminia
7115.3	Post a tstack	EDSCLU::GARROD	IBM Interconnect Engineering	`Thu Jun 05 1997 16:47`	9
	You need to open a support call to take this further. But just in case it is obvious after doing the dbx type: tstack and post the output here. If I can see the problem right off the bat I'll let you know. Otherwise you must open a support call. Dave
7115.4	tstack	BARNA::DSMAIL		`Fri Jun 06 1997 05:05`	46
	(dbx) tstack Thread 0xd: > 0 strncpy(0x0, 0x0, 0x0, 0x0, 0x0) [0x3ff800d2610] 1 srv_validate_host_node_access(llcb = (nil), status_vec = (nil)) ["/usr/use] 2 srv_process_message(llcb = (nil), ccb = (nil)) ["/usr/users/snagwy/sna2/sn] Thread 0x6: > 0 msg_receive_trap(0x0, 0x0, 0x0, 0x0, 0x0) [0x3ff8053ea40] 1 msg_receive(0x3ffc01877e0, 0x3ffc0188c30, 0x7, 0x400, 0x3ffc018a100) [0x3f] 2 (unknown)() [0x3ff8056c80c] 3 thdBase(0x0, 0x0, 0x0, 0x1, 0x45586732) [0x3ff80574274] Thread 0x10: 0 (noname)() [0x120000000] Thread 0x9: > 0 __select(0x1400b9a00, 0x1400b9a10, 0x140034b60, 0x3, 0x140034b60) [0x3ff80] 1 rcv_gap_thread(dcb = 0x140041000) ["/usr/users/snagwy/sna2/snalu62v3_bl05/] More (n if no)?y 2 srv_outbound_thread(cp_parm = 0x14001c0d0) ["/usr/users/snagwy/sna2/snalu6] 3 thdBase(0x0, 0x0, 0x0, 0x1, 0x45586732) [0x3ff80574274] Thread 0xb: > 0 __select(0x140027bc0, 0x11ffff698, 0x140041440, 0x3ff805afd30, 0x12007ee3c] 1 srv_receive_connect(ecb = (nil), llcb = 0x11ffffab8) ["/usr/users/snagwy/s] 2 srv_rcv_conn_thread(ecb = 0x140041300) ["/usr/users/snagwy/sna2/snalu62v3_] 3 main(argc = 2, argv = 0x11ffffe28) ["/usr/users/snagwy/sna2/snalu62v3_bl05] Thread 0xf: > 0 nxm_idle(0x0, 0x0, 0x0, 0x0, 0x0) [0x3ff8053ea88] 1 (unknown)() [0x3ff8057bcf0] (dbx)