[MDEV-28440] MariaDB crashes during select Created: 2022-04-28 Updated: 2022-08-04 Resolved: 2022-08-04 |
|
| Status: | Closed |
| Project: | MariaDB Server |
| Component/s: | Data Manipulation - Update |
| Affects Version/s: | 10.5.15 |
| Fix Version/s: | N/A |
| Type: | Bug | Priority: | Major |
| Reporter: | Willian Vieira Rosa | Assignee: | Unassigned |
| Resolution: | Incomplete | Votes: | 0 |
| Labels: | crash, innodb | ||
| Environment: |
CentOS Linux release 7.9.2009 (Core) |
||
| Attachments: |
|
| Description |
|
MariaDB crashes during queries trying to read, delete or update a row. 2022-04-28 16:43:41 16 [ERROR] [FATAL] InnoDB: FIL_PAGE_TYPE=0 on BLOB read file ./dbpro_10080_sei/versao_secao_documento.ibd page 13654973 To report this bug, see https://mariadb.com/kb/en/reporting-bugs We will try our best to scrape up some info that will hopefully help Server version: 10.5.15-MariaDB Thread pointer: 0x7f1bac0009b8 Trying to get some variables. Connection ID (thread ID): 16 Optimizer switch: index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on,index_merge_sort_intersection=off,engine_condition_pushdown=off,index_condition_pushdown=on,derived_merge=on,derived_with_keys=on,firstmatch=on,loosescan=on,materialization=on,in_to_exists=on,semijoin=on,partial_match_rowid_merge=on,partial_match_table_scan=on,subquery_cache=on,mrr=off,mrr_cost_based=off,mrr_sort_keys=off,outer_join_with_cache=on,semijoin_with_cache=on,join_cache_incremental=on,join_cache_hashed=on,join_cache_bka=on,optimize_join_buffer_size=on,table_elimination=on,extended_keys=on,exists_to_in=on,orderby_uses_equalities=on,condition_pushdown_for_derived=on,split_materialized=on,condition_pushdown_for_subquery=on,rowid_filter=on,condition_pushdown_from_having=on,not_null_range_scan=off The manual page at https://mariadb.com/kb/en/how-to-produce-a-full-stack-trace-for-mysqld/ contains |
| Comments |
| Comment by Alice Sherepa [ 2022-04-29 ] | ||||||||
|
Could you please add the output of SHOW CREATE TABLE versao_secao_documento ? | ||||||||
| Comment by Willian Vieira Rosa [ 2022-04-29 ] | ||||||||
|
This is the result of SHOW CREATE TABLE: MariaDB [dbpro_10080_sei]> SHOW CREATE TABLE versao_secao_documento;
-----------------------
----------------------- | ||||||||
| Comment by Willian Vieira Rosa [ 2022-04-29 ] | ||||||||
|
CHECK TABLE don't show errors: MariaDB [dbpro_10080_sei]> CHECK TABLE versao_secao_documento;
---------------------------------------
--------------------------------------- | ||||||||
| Comment by Willian Vieira Rosa [ 2022-05-26 ] | ||||||||
|
To show possible solutions to people who might have the same problem, I will show how er recovered the database. "innochecksum -a 3 /opt/dbfiles/mariadb/10/pro_10080_sei/data/dbpro_10080_sei/versao_secao_documento.ibd The table has 6892719 registries, so I create a Select using the PK trying to identify the group of corrupted registries. "select id_versao_secao_documento from versao_secao_documento where id_versao_secao_documento > 1 and id_versao_secao_documento < 1000000 ; Then I created a new table using a select to exclude the corrupted registries. It was really impossible to recovery this 3 registries, but now we have a working dababase. | ||||||||
| Comment by Alice Sherepa [ 2022-08-04 ] | ||||||||
|
There were some bug reports about CHECK TABLE in some cases does not detect table corruption, so it might be also the same problem. |