| Bug #49515 | SIGHUP to mysqld with ndb causes SIGSEGV | ||
|---|---|---|---|
| Submitted: | 7 Dec 2009 16:18 | Modified: | 9 Dec 2009 12:20 |
| Reporter: | Yoshinori Matsunobu | Email Updates: | |
| Status: | Closed | Impact on me: | |
| Category: | MySQL Cluster: Cluster (NDB) storage engine | Severity: | S2 (Serious) |
| Version: | 7.0.9 | OS: | Solaris |
| Assigned to: | Jonas Oreland | CPU Architecture: | Any |
[8 Dec 2009 8:55]
Bugs System
A patch for this bug has been committed. After review, it may be pushed to the relevant source trees for release in the next version. You can access the patch from: http://lists.mysql.com/commits/93124 3183 Jonas Oreland 2009-12-08 ndb - bug#49515 - be careful in ndbcluster_binlog_wait, as THD might be null is called from signal handler
[8 Dec 2009 9:12]
Bugs System
A patch for this bug has been committed. After review, it may be pushed to the relevant source trees for release in the next version. You can access the patch from: http://lists.mysql.com/commits/93127 3184 Jonas Oreland 2009-12-08 ndb - bug#49515 - be careful in ndbcluster_binlog_wait, as THD might be null is called from signal handler
[8 Dec 2009 9:14]
Bugs System
Pushed into 5.1.39-ndb-7.1.0 (revid:jonas@mysql.com-20091208090936-qlflnt06fc37a1kw) (version source revid:jonas@mysql.com-20091208090936-qlflnt06fc37a1kw) (merge vers: 5.1.39-ndb-7.1.0) (pib:13)
[8 Dec 2009 9:19]
Jonas Oreland
pushed to 6.3.29 and 7.0.10
[8 Dec 2009 13:59]
Bugs System
A patch for this bug has been committed. After review, it may be pushed to the relevant source trees for release in the next version. You can access the patch from: http://lists.mysql.com/commits/93199 3181 Martin Skold 2009-12-08 [merge] Merge modified: configure.in mysql-test/std_data/ndb_config_mycnf1.cnf mysql-test/suite/ndb/r/ndb_config.result mysql-test/suite/ndb/t/ndb_config.test sql/ha_ndbcluster_binlog.cc storage/ndb/src/kernel/blocks/dbdih/DbdihMain.cpp storage/ndb/src/kernel/blocks/dblqh/Dblqh.hpp storage/ndb/src/kernel/blocks/dblqh/DblqhInit.cpp storage/ndb/src/kernel/blocks/dblqh/DblqhMain.cpp storage/ndb/src/ndbapi/Ndb.cpp storage/ndb/test/ndbapi/testBlobs.cpp
[9 Dec 2009 12:20]
Jon Stephens
Documented bugfix int he NDB-6.3.29 and 7.0.10 changelogs as follows:
Sending SIGHUP to a mysqld running with the --ndbcluster and
--log-bin options caused the process to crash instead of
refreshing its log files.
Closed.
Description: When sending SIGHUP to mysqld with ndbcluster and log-bin, mysqld crashes with SIGSEGV, instead of refreshing log files. Here is a stack trace I got. ----------------- lwp# 29 / thread# 29 -------------------- fffffd7ffed7cb8a _lwp_kill () + a 0000000000a8d399 my_write_core () + 21 00000000007b414e handle_segfault () + 1de fffffd7ffed77386 __sighndlr () + 6 fffffd7ffed6bc32 call_user_handler () + 252 fffffd7ffed6be4e sigacthandler (b, 0, fffffd7ffe90ea00) + de --- called from signal handler with signal 11 (SIGSEGV) --- 0000000000970e3b __1cWndbcluster_binlog_wait6FpnDTHD__v_ () + 6b 000000000097135b __1cVndbcluster_flush_logs6FpnKhandlerton__b_ () + 1b 00000000008bc0ee __1cQflush_handlerton6FpnDTHD_pnNst_plugin_int_pv_c_ () + 1e 0000000000941758 __1cYplugin_foreach_with_mask6FpnDTHD_pF1pnNst_plugin_int_pv_ciI4_b_ () + 4e8 00000000008bc150 __1cNha_flush_logs6FpnKhandlerton__b_ () + 50 00000000007cb81f __1cUreload_acl_and_cache6FpnDTHD_LpnKTABLE_LIST_pb_b_ () + 15f 00000000007b4784 signal_hand () + 214 fffffd7ffed7704b _thr_setup () + 5b fffffd7ffed77280 _lwp_start () reload_acl_and_cache is called here. case SIGHUP: if (!abort_loop) { bool not_used; mysql_print_status(); // Print some debug info reload_acl_and_cache((THD*) 0, (REFRESH_LOG | REFRESH_TABLES | REFRESH_FAST | REFRESH_GRANT | REFRESH_THREADS | REFRESH_HOSTS), (TABLE_LIST*) 0, ¬_used); // Flush logs In ndbcluster_binlog_wait, it will get SIGSEGV if thd is zero. static void ndbcluster_binlog_wait(THD *thd) { if (ndb_binlog_running) { DBUG_ENTER("ndbcluster_binlog_wait"); ulonglong wait_epoch= ndb_get_latest_trans_gci(); /* cluster not connected or no transactions done so nothing to wait for */ if (!wait_epoch) DBUG_VOID_RETURN; const char *save_info= thd ? thd->proc_info : 0; int count= 30; if (thd) thd->proc_info= "Waiting for ndbcluster binlog update to " "reach current position"; pthread_mutex_lock(&injector_mutex); while (!thd->killed && count && ndb_binlog_running && (ndb_latest_handled_binlog_epoch == 0 || ndb_latest_handled_binlog_epoch < wait_epoch)) How to repeat: 1. Run mysqld with ndbcluster and log-bin 2. kill -s HUP <pid of mysqld> Suggested fix: Strict null checking for THD is required in ndbcluster_binlog_wait.