Error During Save Point Backout
IBSurgeon FirstAID 4.0 IBUndelete 2.6 Firebird Encryption Plugin Framework Upgrades Previous versions IBSurgeon ISV Subscription IBSurgeon Recovery Pack FBScanner IBAnalyst FBDataGuard FBMonLogger IBTM Transaction Monitor IBBackupSurgeon IBSurgeon Free Tools Articles Tests Firebird 2.5 Performance Logs Firebird 3.0 Performance Logs Firebird 4.0 Performance Logs Documentation Firebird Training Articles Home Articles Common InterBase/Firebird errors caused by corruptions and their recovery chances Common InterBase/Firebird errors caused by corruptions and their recovery chances /en/articles/how-to-check-ram-and-avoid-database-corruptions/Alexey Kovyazin, 31-03-2014 Below is the description of common errors and problems in InterBase/Firebird databases and their recovery chances. To get exact recovery price and time please contact us via email. For approximate pricing please see "Firebird and InterBase Recovery" service description. There is no 100% warranty that described errors exactly correspond to the described reasons. Internal gds software consistency check (cannot find tip page (165)) Error description Database cannot be opened using Firebird or InterBase engine, and the following message appears: Internal gds software consistency check (cannot find tip page (165)) Abnormal shutdown or physical database file corruption. Transaction inventory page has been lost (TIP). Corruption area can vary from several pages to the whole database, so additional investigation needed. Reason (why it happened?) The most probable reasons are abnormal server shutdown (using Reset button), wrong backup approach or backup tools. On Windows XP such corruption can be caused by "System Restore" feature for "gdb" files. Recovery process First, the database should be scanned with FirstAID Diagnostician. If FirstAID does not warn about serious corruption, corruption can be fixed with the full version of FirstAID. In the case of serious corruption, the custom recovery needed. Chances to fix corruption 99% Database file appears corrupt. Wrong page type. Page NNN is of wrong type (expected X, found Y) Error description Error message appeared in standard output or in firebird.log or interbase.log: Database file appears corrupt. Wrong page type. Page NNN is of wrong type (expected X, found Y) Due to the physical corruption or another reason, the sequence of database file pages has been changed, or wrong values appeared on pointer pages or index root pages, etc. Reason The most probable reasons are abnormal server shutdown (using Reset button), wrong backup procedure or wrong backup tools/approach. Recovery pr
tables Remote You are not logged in. Help Print Public Report Report From: InterBase/Server/Internal/JRD [ Add a report in this area ] Report #: 27525 Status: Closed (Pulled) Heavy index inserts causes internal gds consistency check- InterBase V7.1 SP2 (without and with Security-Patch)- Windows 2 Project:InterBase Build #: Version: 7.5 Submitted By: John Kaster Report Type: Basic functionality failure Date Reported: 11/19/2004 12:00:00 AM Severity: Infrequently encountered problem Last Updated: 3/20/2012 2:24:39 AM Platform: All versions Internal Tracking #: 217539 Resolution: Fixed(Resolution Comments) Resolved http://ib-aid.com/en/articles/common-interbase-firebird-errors-caused-by-corruptions-and-their-recovery-chances/ in Build: : 7.5.1.162 Duplicate of:None Voting and Rating Overall Rating: No Ratings Yet 0.00 out of 5 Total Votes: None Description Heavy index inserts causes internal gds consistency check - InterBase V7.1 SP2 (without and with Security-Patch) - Windows 2000 SP4 Server - Windows 2000 SP4 Clients - DB now is about 26GB - NETBEUI connect (because the lack of http://qc.embarcadero.com/wc/qcmain.aspx?d=27525 Events with Windwos 2000 SP4) All 1-2 weeks we have the following message in InterBase.log: SERVER (Server) Fri Jul 02 01:06:49 2004 Database: D:\DATEN\RZKOM\DB\RZKOM5.IB deadlock internal gds software consistency check (error during savepoint backout (290), file: exe.c line: 1792) After that the DB is corrupt and we have to do a backup/restore cycle (Backup take around 2.5 hours, Restore take about 7 hours (?)). At the time it happens (usually in the night), there is a lot (?) of processing: There are 10 automated clients which are trying to insert about 100 records about every 20 seconds. In the newsgroups I get several hints refer to transaction management. As to that I can say: - We do NOT use commitretaining (never). - We use read only transactions where it is possible. - We have NO long running transactions (neither in concept of the application nor in Performance Monitor). - In one transaction there may be insert's of about 1000 records (but the transaction is as short as possible to do the insert) with a lot of triggers and stored procedures take part. This transactions can not be divided. -
Programmer's Town »Databases »Forerunner : "error during savepoint backout (290), file: exe.cpp line: 1618"... Pages 1 You must login or register to post a reply Topic RSS feed Posts [ 2 ] http://www.progtown.com/topic1948613-forerunner-error-during-savepoint-backout-290-file-execpp-line-1618.html 1 Topic by Tabloid 2015-01-12 14:49:23 Tabloid Member Offline Registered: 2012-04-21 Posts: 7,999 Topic: http://tracker.firebirdsql.org/browse/CORE-4914?page=com.atlassian.jira.plugin.system.issuetabpanels%3Aall-tabpanel Forerunner : "error during savepoint backout (290), file: exe.cpp line: 1618"... ... Appearing normally after several tens "page type 5 lock denied (216)", like - only on SC / CS: [spoiler] oel64 Thu Jan 8 1:23:49 PM 2015 Database: oltp30 page 75993, page type 5 lock denied (216) oel64 Thu Jan 8 1:24:40 PM 2015 error during Database: oltp30 page 81863, page type 5 lock denied (216) oel64 Thu Jan 8 1:51:14 PM 2015 Sweep is started by SWEEPER Database "oltp30" OIT 148724, OAT 174066, OST 171764, Next 180398 oel64 Thu Jan 8 1:57:25 PM 2015 Database: oltp30 page 149346, page type 5 lock denied (216) oel64 Thu Jan 8 2:00:14 PM 2015 Sweep is finished Database "oltp30" OIT 171763, OAT 177615, OST 172594, Next error during save 192652 oel64 Thu Jan 8 2:00:40 PM 2015 Database: oltp30 page 3688, page type 5 lock denied (216) oel64 Thu Jan 8 2:06:13 PM 2015 Database: oltp30 page 93826, page type 5 lock denied (216) oel64 Thu Jan 8 2:21:10 PM 2015 Database: oltp30 page 50968, page type 5 lock denied (216) oel64 Thu Jan 8 2:23:43 PM 2015 Database: oltp30 page 42292, page type 5 lock denied (216) oel64 Thu Jan 8 2:24:58 PM 2015 Database: oltp30 page 63484, page type 5 lock denied (216) oel64 Thu Jan 8 2:26:36 PM 2015 Database: oltp30 page 70116, page type 5 lock denied (216) oel64 Thu Jan 8 2:34:35 PM 2015 Database: oltp30 page 136692, page type 5 lock denied (216) oel64 Thu Jan 8 2:41:33 PM 2015 Database: oltp30 page 124896, page type 5 lock denied (216) oel64 Thu Jan 8 2:41:33 PM 2015 Database:/var/db/fb30/oltp30.fdb page 124896, page type 5 lock denied internal Firebird consistency check (error during savepoint backout (290), file: exe.cpp line: 1618) oel64 Thu Jan 8 2:41:33 PM 2015 Database:/var/db/fb30/oltp30.fdb page 124896, page type 5 lock denied internal Firebird consistency check (error during savepoint backout (290), file: exe.cpp line: 1618) oel64 Thu Jan 8 2:41:33 PM 2015 Database:/var/db/fb30/oltp30.fdb internal Firebird consist
Zotov Votes: 0 Watchers: 2 Operations If you were logged in you would be able to see more operations. Firebird Core Moving database to 'full shutdown' state can corrupt index (at least) on long-key text field. Shutdown process can hang w/o returning control to outer environment. Created: 24/Aug/15 05:44 PM Updated: 09/Nov/15 09:49 PM Component/s: Engine, SVCMGR Affects Version/s: None Fix Version/s: 3.0 RC1 File Attachments: 1. dbshut-fb30ss-15attaches-missing-some-errors-in-firebird-log.zip (46 kB) 2. fb3ss_result.zip (3.55 MB) Issue Links: Relate This issue is related to: CORE-4904 Index corruption when add data in long-key-indexed field QA Status: No test Description « Hide I've encountered with lot of errors which can be produced when database is moved to shutdown state during intensive DML workload. It seems that almost all of these errors are caused by presence of long-key indexed field, especially when this field is filled with the same text value (i.e. 100% duplicates). Errors can happen even on 5-7 attachments when test running on machine with slow CPU and/or IO subsystem. All tests was run on Windows. To reproduce: 1) Download following two files using your svn client: 1.1) svn.code.sf.net/p/firebird/code/qa/fbt-repository/trunk/tests/functional/tabloid/batches/dbshut.bat.txt 1.2) svn.code.sf.net/p/firebird/code/qa/fbt-repository/trunk/tests/functional/tabloid/batches/dbshut.conf 2) Rename 'dbshut.bat.tx' to 'dbshut.bat' (i.e. remove last part of its 'long' extension). 3) Open 'dbshut.conf' and correct settings related to your FB instance(s) (hereafter: 'Cs' = Classic; 'sC' = SuperClassic; 'sS' = SuperServer): === fb25Cshome=