Server won't start

Hi, Could you tell me why the server won't start, got the following error in the online.log file on one of our test box. Sat Apr 26 00:43:57 2003 00:43:57 Event alarms enabled. ALARMPROG = '/ids.9.3/etc/log_full.sh' 00:43:57 Booting Language <c> from module <> 00:43:57 Loading Module <CNULL> 00:43:57 Booting Language <builtin> from module <> 00:43:57 Loading Module <BUILTINNULL> 00:43:57 VP pid=25128 priority fixed at 60, former = 120 00:43:57 Affinitied VP 1 to phys proc 0 00:43:57 VP pid=28100 priority fixed at 60, former = 60 00:43:57 Affinitied VP 3 to phys proc 1 00:44:04 AIX MP latch code enabled 00:44:04 Requested shared memory segment size rounded from 236KB to 240KB 00:44:04 Informix Dynamic Server Version 9.30.UC3 Software Serial Number AA D#J341265 00:44:04 listener-thread: err = -27100: oserr = 9: errstr = : Internal Communic ations Error: NSF subsystem error. System error = 9. 00:44:04 Attempting to bring listener thread down. 00:44:04 Server stopped. Thanks in advance. Sushil... _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail

Hi, the problem is one of the communication configuration that you have in place for the IDS instance. Unfortunately (as I've written sometime ago already), the error message does not tell _which_ listener thread is having the problem (as you can have several different ones configured). If you've configured TCP/IP communication ("onsoctcp" or "ontlitcp" in your sqlhosts file), then it is likely a problem of network resources, e.g. the port that you configured is already in use by some other process. Check the port definition in your sqlhosts file and that it is correctly (unambiguously) defined in /etc/services. You may also want to use "netstat -an" to check for used ports ... If you've SHM communiction configured ("onipcshm" in sqlhosts file), then it can be a problem with the shared memory, maybe an old SHM segment still being around from another instance. Also it can happen that another instance has the same port definition in the sqlhosts. For SHM these may need to be unambiguous also. This is because the port name may be used to create a unique file name. It depends on the version of IDS and perhaps on the platform you're on ... For still other protocols ("onipcstr") there can be similar reasons why it doesn't work (sqlhosts port name entry). Last but not least - check your $ONCONFIG file. Especially check for the NETTYPE parameter. If it is defined for a protocol that you haven't configured in your sqlhosts with a proper server name, it may happen that IDS tries to start a listener thread for this protocol - using some rather unpredictable port number or filename (see above). Such constellations can have quite strange effects. Best is to remove all NETTYPE definitions from your $ONCONFIG unless you've properly defined them yourself, for a correct servername with proper sqlhosts entry. Regards, Martin -- Martin Fuerderer IBM Informix Development Munich Data Management Solutions "Sushil Shir...." <sushilps@hotmail.com> Sent by: forum.subscriber@iiug.org 26.04.2003 07:34 To: ids@iiug.org cc: Subject: Server won't start [1006] Hi, Could you tell me why the server won't start, got the following error in the online.log file on one of our test box. Sat Apr 26 00:43:57 2003 00:43:57 Event alarms enabled. ALARMPROG = '/ids.9.3/etc/log_full.sh' 00:43:57 Booting Language <c> from module <> 00:43:57 Loading Module <CNULL> 00:43:57 Booting Language <builtin> from module <> 00:43:57 Loading Module <BUILTINNULL> 00:43:57 VP pid=25128 priority fixed at 60, former = 120 00:43:57 Affinitied VP 1 to phys proc 0 00:43:57 VP pid=28100 priority fixed at 60, former = 60 00:43:57 Affinitied VP 3 to phys proc 1 00:44:04 AIX MP latch code enabled 00:44:04 Requested shared memory segment size rounded from 236KB to 240KB 00:44:04 Informix Dynamic Server Version 9.30.UC3 Software Serial Number AA D#J341265 00:44:04 listener-thread: err = -27100: oserr = 9: errstr = : Internal Communic ations Error: NSF subsystem error. System error = 9. 00:44:04 Attempting to bring listener thread down. 00:44:04 Server stopped. Thanks in advance. Sushil... _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail

Hi, Thanks for your email, we did the following but no postive results. -> Bounce the unix box. -> Checked the parameters, (parameters are fine because the same files are used during the failover). Ports numbers are fine. -> Did testing via shared memory and it works fine, when you try to use "onsoctcp" connection it doesn't work. -> Created another instance with different name, it works fine on the same server. -> When tried to use above settings for the orginal server, it didn't work, somehow the name of the instance is creating problem. Additional info: Everything works fine on Node -A (2 instances) but when we failover(hacmp cluster) to Node-B, one instance doesn't work, Informix-tech is pointing to DNS. Sushil... >From: Martin Fuerderer <MARTINFU@de.ibm.com> >To: "Sushil Shir...." <sushilps@hotmail.com> >CC: forum.subscriber@iiug.org, ids@iiug.org >Subject: Re: Server won't start [1006] >Date: Mon, 28 Apr 2003 01:40:32 -0600 >MIME-Version: 1.0 >Received: from e34.co.us.ibm.com ([32.97.110.132]) by >mc6-f4.law1.hotmail.com with Microsoft SMTPSVC(5.0.2195.5600); Mon, 28 Apr >2003 00:41:17 -0700 >Received: from westrelay04.boulder.ibm.com (westrelay04.boulder.ibm.com >[9.17.193.32])by e34.co.us.ibm.com (8.12.9/8.12.2) with ESMTP id >h3S7fHYp107324;Mon, 28 Apr 2003 03:41:17 -0400 >Received: from d03nm800.boulder.ibm.com (d03av02.boulder.ibm.com >[9.17.193.82])by westrelay04.boulder.ibm.com (8.12.9/NCO/VER6.5) with ESMTP >id h3S7fGtd072472;Mon, 28 Apr 2003 01:41:17 -0600 >X-Message-Info: JGTYoYF78jEHjJx36Oi8+Q1OJDRSDidP >X-Mailer: Lotus Notes Release 5.0.7 March 21, 2001 >Message-ID: <OF14B23712.98BEBBC6-ONC1256D16.002537E5@us.ibm.com> >X-MIMETrack: Serialize by Router on D03NM800/03/M/IBM(Release 6.0.1 >[IBM]|April 4, 2003) at 04/28/2003 01:40:33,Serialize complete at >04/28/2003 01:40:33 >Return-Path: MARTINFU@de.ibm.com >X-OriginalArrivalTime: 28 Apr 2003 07:41:17.0873 (UTC) >FILETIME=[8DCC4210:01C30D59] > >Hi, > >the problem is one of the communication configuration that you have >in place for the IDS instance. >Unfortunately (as I've written sometime ago already), the error message >does not tell _which_ listener thread is having the problem (as you can >have several different ones configured). > >If you've configured TCP/IP communication ("onsoctcp" or "ontlitcp" in >your sqlhosts file), then it is likely a problem of network resources, >e.g. >the port that you configured is already in use by some other process. >Check the port definition in your sqlhosts file and that it is correctly >(unambiguously) defined in /etc/services. >You may also want to use "netstat -an" to check for used ports ... > >If you've SHM communiction configured ("onipcshm" in sqlhosts file), >then it can be a problem with the shared memory, maybe an old SHM >segment still being around from another instance. Also it can happen that >another instance has the same port definition in the sqlhosts. For SHM >these may need to be unambiguous also. This is because the port name >may be used to create a unique file name. It depends on the version of >IDS and perhaps on the platform you're on ... > >For still other protocols ("onipcstr") there can be similar reasons why it >doesn't work (sqlhosts port name entry). > >Last but not least - check your $ONCONFIG file. Especially check for >the NETTYPE parameter. If it is defined for a protocol that you haven't >configured in your sqlhosts with a proper server name, it may happen >that IDS tries to start a listener thread for this protocol - using some >rather unpredictable port number or filename (see above). Such >constellations can have quite strange effects. Best is to remove all >NETTYPE definitions from your $ONCONFIG unless you've properly >defined them yourself, for a correct servername with proper sqlhosts >entry. > >Regards, >Martin >-- >Martin Fuerderer >IBM Informix Development Munich >Data Management Solutions > > > > > >"Sushil Shir...." <sushilps@hotmail.com> >Sent by: forum.subscriber@iiug.org >26.04.2003 07:34 > > > To: ids@iiug.org > cc: > Subject: Server won't start [1006] > > > >Hi, > >Could you tell me why the server won't start, got the following >error in the online.log file on one of our test box. > >Sat Apr 26 00:43:57 2003 > >00:43:57 Event alarms enabled. ALARMPROG = '/ids.9.3/etc/log_full.sh' >00:43:57 Booting Language <c> from module <> >00:43:57 Loading Module <CNULL> >00:43:57 Booting Language <builtin> from module <> >00:43:57 Loading Module <BUILTINNULL> >00:43:57 VP pid=25128 priority fixed at 60, former = 120 >00:43:57 Affinitied VP 1 to phys proc 0 >00:43:57 VP pid=28100 priority fixed at 60, former = 60 >00:43:57 Affinitied VP 3 to phys proc 1 >00:44:04 AIX MP latch code enabled >00:44:04 Requested shared memory segment size rounded from 236KB to 240KB >00:44:04 Informix Dynamic Server Version 9.30.UC3 Software Serial >Number AA >D#J341265 >00:44:04 listener-thread: err = -27100: oserr = 9: errstr = : Internal >Communic >ations Error: NSF subsystem error. >System error = 9. >00:44:04 Attempting to bring listener thread down. > >00:44:04 Server stopped. > > > >Thanks in advance. >Sushil... > > > > > > > >_________________________________________________________________ >Add photos to your e-mail with MSN 8. Get 2 months FREE*. >http://join.msn.com/?page=features/featuredemail > > > > > _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=features/junkmail

Hi, what you describe, i.e. that the problem seems to be "tied" to the server name -> I would think it still is a config problem. After all, if it would be a DNS problem, then other server instances on the same box should have the same problem, right ? You may want to check that (due to sqlhosts entries) your server on Node-B is not trying to listen at a socket that is on Node-A ... With the failover scenario I guess your server names are the same on Node-A and Node-B ? Make sure that the host column in sqlhosts file on Node-A lists Node-A as hosts for the server name, but on Node-B it needs to have Node-B as host in the host column of sqlhosts file. (This can be easily overlooked when copying the sqlhosts file from Node-A to Node-B ...). Apart from that [ and there's also dependency on the version, which you still heed as a secret as well as the platform ... :) ] there's a restriction as to how long a servername can be. So if you've two servernames in your sqlhosts that don't differ within the significant length of the name, you could have an effect that the name is truncated at a shorted length and therefore the wrong line is picked from the sqlhosts file ... That's what I can think of according to your descriptions. Regards, Martin -- Martin Fuerderer IBM Informix Development Munich Data Management Solutions "Sushil Shirodkar" <sushilps@hotmail.com> 28.04.2003 19:17 To: Martin Fuerderer/Germany/IBM@IBMDE cc: forum.subscriber@iiug.org, ids@iiug.org Subject: Re: Server won't start [1006] Hi, Thanks for your email, we did the following but no postive results. -> Bounce the unix box. -> Checked the parameters, (parameters are fine because the same files are used during the failover). Ports numbers are fine. -> Did testing via shared memory and it works fine, when you try to use "onsoctcp" connection it doesn't work. -> Created another instance with different name, it works fine on the same server. -> When tried to use above settings for the orginal server, it didn't work, somehow the name of the instance is creating problem. Additional info: Everything works fine on Node -A (2 instances) but when we failover(hacmp cluster) to Node-B, one instance doesn't work, Informix-tech is pointing to DNS. Sushil... >From: Martin Fuerderer <MARTINFU@de.ibm.com> >To: "Sushil Shir...." <sushilps@hotmail.com> >CC: forum.subscriber@iiug.org, ids@iiug.org >Subject: Re: Server won't start [1006] >Date: Mon, 28 Apr 2003 01:40:32 -0600 >MIME-Version: 1.0 >Received: from e34.co.us.ibm.com ([32.97.110.132]) by >mc6-f4.law1.hotmail.com with Microsoft SMTPSVC(5.0.2195.5600); Mon, 28 Apr >2003 00:41:17 -0700 >Received: from westrelay04.boulder.ibm.com (westrelay04.boulder.ibm.com >[9.17.193.32])by e34.co.us.ibm.com (8.12.9/8.12.2) with ESMTP id >h3S7fHYp107324;Mon, 28 Apr 2003 03:41:17 -0400 >Received: from d03nm800.boulder.ibm.com (d03av02.boulder.ibm.com >[9.17.193.82])by westrelay04.boulder.ibm.com (8.12.9/NCO/VER6.5) with ESMTP >id h3S7fGtd072472;Mon, 28 Apr 2003 01:41:17 -0600 >X-Message-Info: JGTYoYF78jEHjJx36Oi8+Q1OJDRSDidP >X-Mailer: Lotus Notes Release 5.0.7 March 21, 2001 >Message-ID: <OF14B23712.98BEBBC6-ONC1256D16.002537E5@us.ibm.com> >X-MIMETrack: Serialize by Router on D03NM800/03/M/IBM(Release 6.0.1 >[IBM]|April 4, 2003) at 04/28/2003 01:40:33,Serialize complete at >04/28/2003 01:40:33 >Return-Path: MARTINFU@de.ibm.com >X-OriginalArrivalTime: 28 Apr 2003 07:41:17.0873 (UTC) >FILETIME=[8DCC4210:01C30D59] > >Hi, > >the problem is one of the communication configuration that you have >in place for the IDS instance. >Unfortunately (as I've written sometime ago already), the error message >does not tell _which_ listener thread is having the problem (as you can >have several different ones configured). > >If you've configured TCP/IP communication ("onsoctcp" or "ontlitcp" in >your sqlhosts file), then it is likely a problem of network resources, >e.g. >the port that you configured is already in use by some other process. >Check the port definition in your sqlhosts file and that it is correctly >(unambiguously) defined in /etc/services. >You may also want to use "netstat -an" to check for used ports ... > >If you've SHM communiction configured ("onipcshm" in sqlhosts file), >then it can be a problem with the shared memory, maybe an old SHM >segment still being around from another instance. Also it can happen that >another instance has the same port definition in the sqlhosts. For SHM >these may need to be unambiguous also. This is because the port name >may be used to create a unique file name. It depends on the version of >IDS and perhaps on the platform you're on ... > >For still other protocols ("onipcstr") there can be similar reasons why it >doesn't work (sqlhosts port name entry). > >Last but not least - check your $ONCONFIG file. Especially check for >the NETTYPE parameter. If it is defined for a protocol that you haven't >configured in your sqlhosts with a proper server name, it may happen >that IDS tries to start a listener thread for this protocol - using some >rather unpredictable port number or filename (see above). Such >constellations can have quite strange effects. Best is to remove all >NETTYPE definitions from your $ONCONFIG unless you've properly >defined them yourself, for a correct servername with proper sqlhosts >entry. > >Regards, >Martin >-- >Martin Fuerderer >IBM Informix Development Munich >Data Management Solutions > > > > > >"Sushil Shir...." <sushilps@hotmail.com> >Sent by: forum.subscriber@iiug.org >26.04.2003 07:34 > > > To: ids@iiug.org > cc: > Subject: Server won't start [1006] > > > >Hi, > >Could you tell me why the server won't start, got the following >error in the online.log file on one of our test box. > >Sat Apr 26 00:43:57 2003 > >00:43:57 Event alarms enabled. ALARMPROG = '/ids.9.3/etc/log_full.sh' >00:43:57 Booting Language <c> from module <> >00:43:57 Loading Module <CNULL> >00:43:57 Booting Language <builtin> from module <> >00:43:57 Loading Module <BUILTINNULL> >00:43:57 VP pid=25128 priority fixed at 60, former = 120 >00:43:57 Affinitied VP 1 to phys proc 0 >00:43:57 VP pid=28100 priority fixed at 60, former = 60 >00:43:57 Affinitied VP 3 to phys proc 1 >00:44:04 AIX MP latch code enabled >00:44:04 Requested shared memory segment size rounded from 236KB to 240KB >00:44:04 Informix Dynamic Server Version 9.30.UC3 Software Serial >Number AA >D#J341265 >00:44:04 listener-thread: err = -27100: oserr = 9: errstr = : Internal >Communic >ations Error: NSF subsystem error. >System error = 9. >00:44:04 Attempting to bring listener thread down. > >00:44:04 Server stopped. > > > >Thanks in advance. >Sushil... > > > > > > > >_________________________________________________________________ >Add photos to your e-mail with MSN 8. Get 2 months FREE*. >http:

Hi, When we did the failover from NODE-A to NODE-B, the files under /INFORMIXTMP with the instance name did not get deleted for somereason. Tech. asked me to delete those files manually and server was back online. He got the clue when I said that we changed the name of the instance and everything was fine. Now I have modified failover script to take care of this problem. Thanks guys for giving your valuable suggestions. Regards Sushil.... >From: Martin Fuerderer <MARTINFU@de.ibm.com> >To: "Sushil Shirodkar" <sushilps@hotmail.com> >CC: forum.subscriber@iiug.org, ids@iiug.org >Subject: Re: Server won't start [1006] >Date: Tue, 29 Apr 2003 16:06:39 +0200 >MIME-Version: 1.0 >Received: from e31.co.us.ibm.com ([32.97.110.129]) by >mc6-f39.law1.hotmail.com with Microsoft SMTPSVC(5.0.2195.5600); Tue, 29 Apr >2003 07:07:29 -0700 >Received: from westrelay02.boulder.ibm.com (westrelay02.boulder.ibm.com >[9.17.195.11])by e31.co.us.ibm.com (8.12.9/8.12.2) with ESMTP id >h3TE7SQs095020;Tue, 29 Apr 2003 10:07:28 -0400 >Received: from d03nm800.boulder.ibm.com (d03av02.boulder.ibm.com >[9.17.193.82])by westrelay02.boulder.ibm.com (8.12.9/NCO/VER6.5) with ESMTP >id h3TE7RPS104506;Tue, 29 Apr 2003 08:07:27 -0600 >X-Message-Info: JGTYoYF78jEHjJx36Oi8+Q1OJDRSDidP >X-Mailer: Lotus Notes Release 5.0.7 March 21, 2001 >Message-ID: <OF28CFE003.CB368FD5-ONC1256D17.004D2AD0@us.ibm.com> >X-MIMETrack: Serialize by Router on D03NM800/03/M/IBM(Release 6.0.1 >[IBM]|April 4, 2003) at 04/29/2003 08:06:41,Serialize complete at >04/29/2003 08:06:41 >Return-Path: MARTINFU@de.ibm.com >X-OriginalArrivalTime: 29 Apr 2003 14:07:29.0369 (UTC) >FILETIME=[AB7FB490:01C30E58] > >Hi, > >hmm. The server names look reasonably short ... :) > >with that and your assurance that the sqlhosts and /etc/services >files are all correct I don't really know what the problem would be. >And I don't know of any specific names being "reserved words" >that would not work. >Don't know what else to say to this problem ... :-( > >Regards, >Martin >-- >Martin Fuerderer >IBM Informix Development Munich >Data Management Solutions > > > > > >"Sushil Shirodkar" <sushilps@hotmail.com> >29.04.2003 15:06 > > > To: Martin Fuerderer/Germany/IBM@IBMDE > cc: forum.subscriber@iiug.org, ids@iiug.org > Subject: Re: Server won't start [1006] > > > >Hi, > >IDS 9.3 UC3 AIX 5.1. >DBSERVERNAME "prodshm" >DBSERVERALIASES "prod">Hostname on NODE-A & B are different in sqlhosts file. > >Did additional test : stopped NODE-A completely but it didn't help so I >changed the instance name on NODE-B and it worked fine, now the question >is >why NODE-B is not recognising the orignial instance name from NODE -A. > >Sushil... > > > > >From: Martin Fuerderer <MARTINFU@de.ibm.com> > >To: "Sushil Shirodkar" <sushilps@hotmail.com> > >CC: forum.subscriber@iiug.org, ids@iiug.org > >Subject: Re: Server won't start [1006] > >Date: Tue, 29 Apr 2003 01:27:31 -0600 > >MIME-Version: 1.0 > >Received: from e31.co.us.ibm.com ([32.97.110.129]) by > >mc4-f5.law16.hotmail.com with Microsoft SMTPSVC(5.0.2195.5600); Tue, 29 >Apr > >2003 00:28:22 -0700 > >Received: from westrelay04.boulder.ibm.com (westrelay04.boulder.ibm.com > >[9.17.193.32])by e31.co.us.ibm.com (8.12.9/8.12.2) with ESMTP id > >h3T7SLQs265756;Tue, 29 Apr 2003 03:28:21 -0400 > >Received: from d03nm800.boulder.ibm.com (d03av02.boulder.ibm.com > >[9.17.193.82])by westrelay04.boulder.ibm.com (8.12.9/NCO/VER6.5) with >ESMTP > >id h3T7SJfQ082272;Tue, 29 Apr 2003 01:28:20 -0600 > >X-Message-Info: JGTYoYF78jEHjJx36Oi8+Q1OJDRSDidP > >X-Mailer: Lotus Notes Release 5.0.7 March 21, 2001 > >Message-ID: <OF6C63C3DB.06CAB4C0-ONC1256D17.0027A956@us.ibm.com> > >X-MIMETrack: Serialize by Router on D03NM800/03/M/IBM(Release 6.0.1 > >[IBM]|April 4, 2003) at 04/29/2003 01:27:34,Serialize complete at > >04/29/2003 01:27:34 > >Return-Path: MARTINFU@de.ibm.com > >X-OriginalArrivalTime: 29 Apr 2003 07:28:22.0382 (UTC) > >FILETIME=[E9FB78E0:01C30E20] > > > >Hi, > > > >what you describe, i.e. that the problem seems to be > >"tied" to the server name -> I would think it still is a config > >problem. > >After all, if it would be a DNS problem, then other server > >instances on the same box should have the same problem, > >right ? > > > >You may want to check that (due to sqlhosts entries) your > >server on Node-B is not trying to listen at a socket that is > >on Node-A ... > >With the failover scenario I guess your server names are > >the same on Node-A and Node-B ? > >Make sure that the host column in sqlhosts file on Node-A > >lists Node-A as hosts for the server name, but on Node-B > >it needs to have Node-B as host in the host column of > >sqlhosts file. (This can be easily overlooked when copying > >the sqlhosts file from Node-A to Node-B ...). > > > >Apart from that [ and there's also dependency on the version, > >which you still heed as a secret as well as the platform ... :) ] > >there's a restriction as to how long a servername can be. > >So if you've two servernames in your sqlhosts that don't differ > >within the significant length of the name, you could have an > >effect that the name is truncated at a shorted length and > >therefore the wrong line is picked from the sqlhosts file ... > > > >That's what I can think of according to your descriptions. > > > >Regards, > >Martin > >-- > >Martin Fuerderer > >IBM Informix Development Munich > >Data Management Solutions > > > > > > > > > > > >"Sushil Shirodkar" <sushilps@hotmail.com> > >28.04.2003 19:17 > > > > > > To: Martin Fuerderer/Germany/IBM@IBMDE > > cc: forum.subscriber@iiug.org, ids@iiug.org > > Subject: Re: Server won't start [1006] > > > > > > > >Hi, > > > >Thanks for your email, we did the following but no postive results. > > > >-> Bounce the unix box. > >-> Checked the parameters, (parameters are fine because the same files >are > > > >used during the failover). Ports numbers are fine. > >-> Did testing via shared memory and it works fine, when you try to use > >"onsoctcp" connection it doesn't work. > >-> Created another instance with different name, it works fine on the >same > > > >server. > >-> When tried to use above settings for the orginal server, it didn't > >work, > >somehow the name of the instance is creating problem. > > > >Additional info: Everything works fine on Node -A (2 instances) but when > >we > >failover(hacmp cluster) to Node-B, one instance doesn't work, > >Informix-tech is pointing to DNS. > > > >Sushil... > > > > > > >From: Martin Fuerderer <MARTINFU@de.ibm.com> > > >To: "Sushil Shir...." <sushilps@hotmail.com> > > >CC: forum.subscriber@iiug.org, ids@iiug.org > > >Subject: Re: Server won't start [1006] > > >Date: Mon, 28 Apr 2003 01:40:32 -0600 > > >MIME-Version: 1.0 > > >Received: from e34.co.us.ibm.com ([32.97.110.132]) by > > >mc6-f4.law1.hotmail.com wit

Hi, so you're running the instances on Node-A and Node-B from a shared (NFS) disk ? TIA, Martin -- "Sushil Shirodkar" <sushilps@hotmail.com> 30.04.2003 19:42 To: Martin Fuerderer/Germany/IBM@IBMDE cc: forum.subscriber@iiug.org, ids@iiug.org Subject: Re: Server won't start [1006] Hi, When we did the failover from NODE-A to NODE-B, the files under /INFORMIXTMP with the instance name did not get deleted for somereason. Tech. asked me to delete those files manually and server was back online. He got the clue when I said that we changed the name of the instance and everything was fine. Now I have modified failover script to take care of this problem. Thanks guys for giving your valuable suggestions. Regards Sushil.... >From: Martin Fuerderer <MARTINFU@de.ibm.com> >To: "Sushil Shirodkar" <sushilps@hotmail.com> >CC: forum.subscriber@iiug.org, ids@iiug.org >Subject: Re: Server won't start [1006] >Date: Tue, 29 Apr 2003 16:06:39 +0200 >MIME-Version: 1.0 >Received: from e31.co.us.ibm.com ([32.97.110.129]) by >mc6-f39.law1.hotmail.com with Microsoft SMTPSVC(5.0.2195.5600); Tue, 29 Apr >2003 07:07:29 -0700 >Received: from westrelay02.boulder.ibm.com (westrelay02.boulder.ibm.com >[9.17.195.11])by e31.co.us.ibm.com (8.12.9/8.12.2) with ESMTP id >h3TE7SQs095020;Tue, 29 Apr 2003 10:07:28 -0400 >Received: from d03nm800.boulder.ibm.com (d03av02.boulder.ibm.com >[9.17.193.82])by westrelay02.boulder.ibm.com (8.12.9/NCO/VER6.5) with ESMTP >id h3TE7RPS104506;Tue, 29 Apr 2003 08:07:27 -0600 >X-Message-Info: JGTYoYF78jEHjJx36Oi8+Q1OJDRSDidP >X-Mailer: Lotus Notes Release 5.0.7 March 21, 2001 >Message-ID: <OF28CFE003.CB368FD5-ONC1256D17.004D2AD0@us.ibm.com> >X-MIMETrack: Serialize by Router on D03NM800/03/M/IBM(Release 6.0.1 >[IBM]|April 4, 2003) at 04/29/2003 08:06:41,Serialize complete at >04/29/2003 08:06:41 >Return-Path: MARTINFU@de.ibm.com >X-OriginalArrivalTime: 29 Apr 2003 14:07:29.0369 (UTC) >FILETIME=[AB7FB490:01C30E58] > >Hi, > >hmm. The server names look reasonably short ... :) > >with that and your assurance that the sqlhosts and /etc/services >files are all correct I don't really know what the problem would be. >And I don't know of any specific names being "reserved words" >that would not work. >Don't know what else to say to this problem ... :-( > >Regards, >Martin >-- >Martin Fuerderer >IBM Informix Development Munich >Data Management Solutions > > > > > >"Sushil Shirodkar" <sushilps@hotmail.com> >29.04.2003 15:06 > > > To: Martin Fuerderer/Germany/IBM@IBMDE > cc: forum.subscriber@iiug.org, ids@iiug.org > Subject: Re: Server won't start [1006] > > > >Hi, > >IDS 9.3 UC3 AIX 5.1. >DBSERVERNAME "prodshm" >DBSERVERALIASES "prod">Hostname on NODE-A & B are different in sqlhosts file. > >Did additional test : stopped NODE-A completely but it didn't help so I >changed the instance name on NODE-B and it worked fine, now the question >is >why NODE-B is not recognising the orignial instance name from NODE -A. > >Sushil... > > > > >From: Martin Fuerderer <MARTINFU@de.ibm.com> > >To: "Sushil Shirodkar" <sushilps@hotmail.com> > >CC: forum.subscriber@iiug.org, ids@iiug.org > >Subject: Re: Server won't start [1006] > >Date: Tue, 29 Apr 2003 01:27:31 -0600 > >MIME-Version: 1.0 > >Received: from e31.co.us.ibm.com ([32.97.110.129]) by > >mc4-f5.law16.hotmail.com with Microsoft SMTPSVC(5.0.2195.5600); Tue, 29 >Apr > >2003 00:28:22 -0700 > >Received: from westrelay04.boulder.ibm.com (westrelay04.boulder.ibm.com > >[9.17.193.32])by e31.co.us.ibm.com (8.12.9/8.12.2) with ESMTP id > >h3T7SLQs265756;Tue, 29 Apr 2003 03:28:21 -0400 > >Received: from d03nm800.boulder.ibm.com (d03av02.boulder.ibm.com > >[9.17.193.82])by westrelay04.boulder.ibm.com (8.12.9/NCO/VER6.5) with >ESMTP > >id h3T7SJfQ082272;Tue, 29 Apr 2003 01:28:20 -0600 > >X-Message-Info: JGTYoYF78jEHjJx36Oi8+Q1OJDRSDidP > >X-Mailer: Lotus Notes Release 5.0.7 March 21, 2001 > >Message-ID: <OF6C63C3DB.06CAB4C0-ONC1256D17.0027A956@us.ibm.com> > >X-MIMETrack: Serialize by Router on D03NM800/03/M/IBM(Release 6.0.1 > >[IBM]|April 4, 2003) at 04/29/2003 01:27:34,Serialize complete at > >04/29/2003 01:27:34 > >Return-Path: MARTINFU@de.ibm.com > >X-OriginalArrivalTime: 29 Apr 2003 07:28:22.0382 (UTC) > >FILETIME=[E9FB78E0:01C30E20] > > > >Hi, > > > >what you describe, i.e. that the problem seems to be > >"tied" to the server name -> I would think it still is a config > >problem. > >After all, if it would be a DNS problem, then other server > >instances on the same box should have the same problem, > >right ? > > > >You may want to check that (due to sqlhosts entries) your > >server on Node-B is not trying to listen at a socket that is > >on Node-A ... > >With the failover scenario I guess your server names are > >the same on Node-A and Node-B ? > >Make sure that the host column in sqlhosts file on Node-A > >lists Node-A as hosts for the server name, but on Node-B > >it needs to have Node-B as host in the host column of > >sqlhosts file. (This can be easily overlooked when copying > >the sqlhosts file from Node-A to Node-B ...). > > > >Apart from that [ and there's also dependency on the version, > >which you still heed as a secret as well as the platform ... :) ] > >there's a restriction as to how long a servername can be. > >So if you've two servernames in your sqlhosts that don't differ > >within the significant length of the name, you could have an > >effect that the name is truncated at a shorted length and > >therefore the wrong line is picked from the sqlhosts file ... > > > >That's what I can think of according to your descriptions. > > > >Regards, > >Martin > >-- > >Martin Fuerderer > >IBM Informix Development Munich > >Data Management Solutions > > > > > > > > > > > >"Sushil Shirodkar" <sushilps@hotmail.com> > >28.04.2003 19:17 > > > > > > To: Martin Fuerderer/Germany/IBM@IBMDE > > cc: forum.subscriber@iiug.org, ids@iiug.org > > Subject: Re: Server won't start [1006] > > > > > > > >Hi, > > > >Thanks for your email, we did the following but no postive results. > > > >-> Bounce the unix box. > >-> Checked the parameters, (parameters are fine because the same files >are > > > >used during the failover). Ports numbers are fine. > >-> Did testing via shared memory and it works fine, when you try to use > >"onsoctcp" connection it doesn't work. > >-> Created another instance with different name, it works fine on the >same > > > >server. > >-> When tried to use above settings for the orginal server, it didn't > >work, > >somehow the name of the instance is creating problem. > > > >Additional info: Everything works fine on Node -A (2 instances) but when > >we > >failover(hacmp cluster) to Node-B, one instance doesn't work, > >Informix-tech is pointing to DNS. > > > >Sushil... > > > >