truncate logical log

All My IDS10 has refused to come online throwing this error: Message Log File: /home/informix/online.log 15:27:47 IBM Informix Dynamic Server Version 10.00.UC4 15:27:47 Who: Session(11, informix@susesrvr, 0, 0xa54e8a2c) Thread(57, xchg_1.8, a54c32d8, 1) File: rsextlog.c Line: 1679 15:27:47 Results: Log record (OLDRSAM:HINSERT) in log 6448, offset 0xd0194 was not rolled back 15:27:47 Action: Use 'onlog' to view the transaction and repair manually. 15:27:47 stack trace for pid 6791 written to /home/informix/tmp/af.4211933 15:27:47 See Also: /home/informix/tmp/af.4211933 15:27:49 Rollback error 126 15:27:49 Assert Failed: Dynamic Server must abort 15:27:49 IBM Informix Dynamic Server Version 10.00.UC4 15:27:49 Who: Session(11, informix@susesrvr, 0, 0xa54e8a2c) Thread(59, xchg_2.0, a54c3d30, 1) File: rslog.c Line: 3589 15:27:49 Results: Dynamic Server must abort 15:27:49 Action: Reinitialize shared memory 15:27:49 stack trace for pid 6791 written to /home/informix/tmp/af.4231a44 15:27:49 See Also: /home/informix/tmp/af.4231a44, shmem.4231a44.0 15:27:58 rslog.c, line 3589, thread 59, proc id 6791, Dynamic Server must abort. 15:27:59 PANIC: Attempting to bring system down I have seen the ONLY way to resolve this is dail in for support with IBM, am just wondering is there anyone on this user group who knows how to do this trancation or has ever done this excersis and has some experience with process? And what are the couses for this problem? Paul.

ids-bounces@iiug.org wrote on 04/15/2007 03:01:38 PM: > All > My IDS10 has refused to come online throwing this error: > > Message Log File: /home/informix/online.log > 15:27:47 IBM Informix Dynamic Server Version 10.00.UC4 > 15:27:47 Who: Session(11, informix@susesrvr, 0, 0xa54e8a2c) > > Thread(57, xchg_1.8, a54c32d8, 1) > > File: rsextlog.c Line: 1679 > 15:27:47 Results: Log record (OLDRSAM:HINSERT) in log 6448, offset > 0xd0194 was > not rolled back > 15:27:47 Action: Use 'onlog' to view the transaction and repair manua= lly. > 15:27:47 stack trace for pid 6791 written to /home/informix/tmp/af.4211933 > 15:27:47 See Also: /home/informix/tmp/af.4211933 > 15:27:49 Rollback error 126 > 15:27:49 Assert Failed: Dynamic Server must abort > 15:27:49 IBM Informix Dynamic Server Version 10.00.UC4 > 15:27:49 Who: Session(11, informix@susesrvr, 0, 0xa54e8a2c) > > Thread(59, xchg_2.0, a54c3d30, 1) > > File: rslog.c Line: 3589 > 15:27:49 Results: Dynamic Server must abort > 15:27:49 Action: Reinitialize shared memory > 15:27:49 stack trace for pid 6791 written to /home/informix/tmp/af.4231a44 > 15:27:49 See Also: /home/informix/tmp/af.4231a44, shmem.4231a44.0 > 15:27:58 rslog.c, line 3589, thread 59, proc id 6791, Dynamic Server = must > abort. > 15:27:59 PANIC: Attempting to bring system down > > I have seen the ONLY way to resolve this is dail in for support with = IBM, am > just wondering is there anyone on this user group who knows how to do= this > trancation or has ever done this excersis and has some experience wit= h > process? > > And what are the couses for this problem? > > Paul. Hi, The message about manual repair is a legacy message - dates back to tho= se days, when it actually was possible to review the tX logs and repair maually. This is not so anymo= re since Online 5 (if I am correct) With IDS this is impossible . The only two ways to correct that is : - restore the affected dbspace or table (the latter in IDS 10 only) this is the recommended official way to proceed. - have Informix support to dial in and manipulate fast recovery this i an emergency solution with the drawback that it leaves the datab= ase online, but in an logically inconsitent state. You will have to review your applica= tion data for consistency and restore consitency manually from external data sources = e.g. Really only recommended when a restore is impossible for what ever reas= on. One would have to review the generated af.4211933 file to say more abou= t root cause. IDS was trying to roll back an INSERT, but the row was apparently not o= n the page where IDS was looking for it. Could be IDS read in a stale page image from disk. Howe= ver, other reasons are possible. I suggest to have IBM Informix Support engaged. Tilman Model-Bosch IBM SWG Premium Support ---------------------------------------------------------- IBM Deutschland GmbH Chairman of the Supervisory Board: Hans Ulrich M=E4rki, Board of Manage= ment: Martin Jetter (Chairman), Rudolf Bauer, Christian Diedrich, Christoph Grandpierre, Matthias Hartmann, Andreas Kerstan Corporate Seat: Stuttgart, Reg.-Gericht: Amtsgericht Stuttgart, HRB-Nr.= : 14 562 WEEE-Reg.-Nr. DE 99369940=

Tilman, Thanks for this info, but i still belive they should try document ways of fixing this issue or resolving? I have a very bad feeling about problems that have to resolved via support team, reason is this might take long and have a prolonged downtime, am just lucky becouse i was attempting to migrate my data to this new enviroment when all this happened! I see on your mail you have suggested to restore the given dbspace or table, but when this problem happens how would i know which dbspace or table is failing? Last how do i engage the IBM team for support? becouse i have a feeling this might happen again.

ids-bounces@iiug.org wrote on 04/16/2007 02:28:10 PM: > Tilman, > Thanks for this info, but i still belive they should try document ways of > fixing this issue or resolving? Documented way of fixing/resolving the issue is a restore . > I have a very bad feeling about problems that have to resolved via support > team, reason is this might take long and have a prolonged downtime, am just > lucky becouse i was attempting to migrate my data to this new enviroment when > all this happened! I agree that the need to invlove support is always unwanted - any problem you can fix yourselve is probably fixed much faster. How did you migrate your IDS DBspaces and chunks ? I assume you did copy all the chunks to the new environment? Are you positive IDS was offline or write-blocked at the time the copy was taken ? If not - that would explain the error . > I see on your mail you have suggested to restore the given dbspace or table, > but when this problem happens how would i know which dbspace or table is > failing? The af.file should give you a hint which table/dbspace is affected. You could also set ONDBSPACEDOWN to 0 , IDS should come online and the affected DBspace will be offline. (provided it was not a critical dbspace) (review IDS documentation for ONDBSPACEDOWN at http://publib.boulder.ibm.com/infocenter/idshelp/v10/index.jsp ) > Last how do i engage the IBM team for support? becouse i have a feeling this > might happen again. You may open an Electronic Service Request via http://www-306.ibm.com/software/support/probsub.html This requires you do have a valid maintenance contract. More info on Informix Support can be found here: http://www-306.ibm.com/software/data/informix/support/ Mit freundlichen Grüssen – With kind regards - Cordialement Tilman Model-Bosch IBM SWG Premium Support ---------------------------------------------------------- IBM Deutschland GmbH Chairman of the Supervisory Board: Hans Ulrich Märki, Board of Management: Martin Jetter (Chairman), Rudolf Bauer, Christian Diedrich, Christoph Grandpierre, Matthias Hartmann, Andreas Kerstan Corporate Seat: Stuttgart, Reg.-Gericht: Amtsgericht Stuttgart, HRB-Nr.: 14 562 WEEE-Reg.-Nr. DE 99369940

PLEASE do not post MIME. This is a plain text forum all we see is gobbledegook! Art S. Kagel ----- Original Message ----- From: Tilman Model-Bosch <ids@iiug.org> At: 4/16 9:04:35 ids-bounces@iiug.org wrote on 04/16/2007 02:28:10 PM: > Tilman, > Thanks for this info, but i still belive they should try document ways of > fixing this issue or resolving? Documented way of fixing/resolving the iss <SNIP>

I thought it was just Monday morning fuzzy eye syndrome. -----Original Message----- Subject: Re: truncate logical log [8907] PLEASE do not post MIME. This is a plain text forum all we see is gobbledegook! Art S. Kagel ----- Original Message ----- From: Tilman Model-Bosch <ids@iiug.org> At: 4/16 9:04:35 ids-bounces@iiug.org wrote on 04/16/2007 02:28:10 PM: > Tilman, > Thanks for this info, but i still beliv >e they should try document ways of > fixing this issuehey should try document ways of > or resolving? Documented way of fixing/resolving the iss <SNIP> ************************************************************************ ******* Forum Note: Use "Reply" to post a response in the discussion forum.

Second attempt : sorry for the MIME repsonse - don't know what happend, Seems Notes did cream my previous reply somehow :-( I hope this comes through as expected > ids-bounces@iiug.org wrote on 04/16/2007 02:28:10 PM: > > > Tilman, > > Thanks for this info, but i still belive they should try document ways of > > fixing this issue or resolving? > Documented way of fixing/resolving the issue is a restore . > > > I have a very bad feeling about problems that have to resolved via support > > team, reason is this might take long and have a prolonged downtime, am just > > lucky becouse i was attempting to migrate my data to this new > enviroment when > > all this happened! > > I agree that the need to invlove support is always unwanted - any > problem you can > fix yourselve is probably fixed much faster. > > How did you migrate your IDS DBspaces and chunks ? > I assume you did copy all the chunks to the new environment? > Are you positive IDS was offline or write-blocked at the time the > copy was taken ? > If not - that would explain the error . > > > I see on your mail you have suggested to restore the given dbspaceor table, > > but when this problem happens how would i know which dbspace or table is > > failing? > > The af.file should give you a hint which table/dbspace is affected. > You could also set ONDBSPACEDOWN to 0 , IDS should come online and > the affected DBspace will > be offline. (provided it was not a critical dbspace) > (review IDS documentation for ONDBSPACEDOWN at > URL: publib.boulder.ibm.com/infocenter/idshelp/v10/index.jsp > > > > Last how do i engage the IBM team for support? becouse i have a > feeling this > > might happen again. > You may open an Electronic Service Request via > URL: www-306.ibm.com/software/support/probsub.html > > This requires you do have a valid maintenance contract. > > More info on Informix Support can be found here: > URL: www-306.ibm.com/software/data/informix/support/ > > Mit freundlichen Grüssen – With kind regards - Cordialement > Tilman Model-Bosch > IBM SWG Premium Support > ---------------------------------------------------------- > IBM Deutschland GmbH > Chairman of the Supervisory Board: Hans Ulrich Märki, Board of > Management: Martin Jetter (Chairman), Rudolf Bauer, Christian > Diedrich, Christoph Grandpierre, Matthias Hartmann, Andreas Kerstan > Corporate Seat: Stuttgart, Reg.-Gericht: Amtsgericht Stuttgart, HRB- > Nr.: 14 562 WEEE-Reg.-Nr. DE 99369940

Ok thrid attempt - now via WEB interface : sorry for the MIME repsonses - don't know what happend, Seems Notes did cream my previous replies somehow :-( Don't knoew what happend > Thanks for this info, but i still belive they should try document ways of > fixing this issue or resolving? Documented way of fixing/resolving the issue is a restore . > I have a very bad feeling about problems that have to resolved via support > team, reason is this might take long and have a prolonged downtime, am just > lucky becouse i was attempting to migrate my data to this new > enviroment when > all this happened! I agree that the need to invlove support is always unwanted - any problem you can fix yourselve is probably fixed much faster. How did you migrate your IDS DBspaces and chunks ? I assume you did copy all the chunks to the new environment? Are you positive IDS was offline or write-blocked at the time the copy was taken ? If not - that would explain the error . > I see on your mail you have suggested to restore the given dbspaceor table, > but when this problem happens how would i know which dbspace or table is > failing? The af.file should give you a hint which table/dbspace is affected. You could also set ONDBSPACEDOWN to 0 , IDS should come online and the affected DBspace will be offline. (provided it was not a critical dbspace) (review IDS documentation for ONDBSPACEDOWN at http://publib.boulder.ibm.com/infocenter/idshelp/v10/index.jsp > Last how do i engage the IBM team for support? becouse i have a > feeling this > might happen again. You may open an Electronic Service Request via http://www-306.ibm.com/software/support/probsub.html This requires you do have a valid maintenance contract. More info on Informix Support can be found here: http://www-306.ibm.com/software/data/informix/support/ HTH Tilman

On 4/15/07, PAUL GATHOGO <pgathogo@gmail.com> wrote: > My IDS10 has refused to come online throwing this error: > > Message Log File: /home/informix/online.log > 15:27:47 IBM Informix Dynamic Server Version 10.00.UC4 > 15:27:47 Who: Session(11, informix@susesrvr, 0, 0xa54e8a2c) > > Thread(57, xchg_1.8, a54c32d8, 1) > > File: rsextlog.c Line: 1679 > 15:27:47 Results: Log record (OLDRSAM:HINSERT) in log 6448, offset 0xd0194 was > not rolled back > 15:27:47 Action: Use 'onlog' to view the transaction and repair manually. > 15:27:47 stack trace for pid 6791 written to /home/informix/tmp/af.4211933 > 15:27:47 See Also: /home/informix/tmp/af.4211933 > 15:27:49 Rollback error 126 > 15:27:49 Assert Failed: Dynamic Server must abort > 15:27:49 IBM Informix Dynamic Server Version 10.00.UC4 > 15:27:49 Who: Session(11, informix@susesrvr, 0, 0xa54e8a2c) > > Thread(59, xchg_2.0, a54c3d30, 1) > > File: rslog.c Line: 3589 > 15:27:49 Results: Dynamic Server must abort > 15:27:49 Action: Reinitialize shared memory > 15:27:49 stack trace for pid 6791 written to /home/informix/tmp/af.4231a44 > 15:27:49 See Also: /home/informix/tmp/af.4231a44, shmem.4231a44.0 > 15:27:58 rslog.c, line 3589, thread 59, proc id 6791, Dynamic Server must > abort. > 15:27:59 PANIC: Attempting to bring system down > > I have seen the ONLY way to resolve this is dail in for support with IBM, am > just wondering is there anyone on this user group who knows how to do this > trancation or has ever done this excersis and has some experience with > process? The only people likely to have the relevant experience are people who do or who used to work for IBM Informix Tech Support. Time to brush off that support contract. It appears that your server went down suddenly and for some unclear reason, the server was not able to rollback an uncommitted transaction (and error 126 might shed some light on the problem). > And what are the couses for this problem? It is most likely a combination of an obscure bug and outrageously bad timing. It might just be bad timing, but that is less likely. -- Jonathan Leffler #include <disclaimer.h> Email: jleffler@earthlink.net, jleffler@us.ibm.com Guardian of DBD::Informix v2007.0226 -- http://dbi.perl.org/ NB: Please do not use this email for correspondence. I don't necessarily read it every week, even.