v$SESSION是我们常用的动态性能视图之一,其SQL_ID字段常用来检测会话当前执行的SQL语句,但在少数版本中V$SESSION的SQL_ID列可能不正常地显示SQL_ID,一般是在Session Active或执行Pl/SQL的时候SQL_ID为NULL,以下是部分相关的部分BUG列表:
BUG# | Affected Version | Symptom | Fix Version |
Bug 13068790 – the value of v$session.sql_id of active session is null [ID 13068790.8] | 11.2.0.2.4 Patch Set Update 11.2.0.2.3 Patch Set Update 11.2.0.2.1 Patch Set Update 11.2.0.2 11.2.0.1 |
active session’s sql_id, sql_address may be null or 0 out while a recursive call is executing. | 11.2.0.3 (Server Patch Set) |
Bug 14393463 : IN V$SESSION, SQL_ID IS NOT NULL BUT STATUS IS INACTIVE | 11.2.0.3 | Abstract: IN V$SESSION, SQL_ID IS NOT NULL BUT STATUS IS INACTIVE | No Workaround Now |
Bug 11670241 : NULL V$SESSION.SQL_ID WHEN RUNNING SLOW SQL IN PL/SQL | 11.2.0.1 | In 11.2 (maybe also in 10g), when running a slow sql in pl/sql, we found that v$session.sql_id will be null. This is very inconvenient for troubleshooting, because we cannot tell what sql is running by a session. The only workaround seems to turn on sql_trace. |
No Workaround Now |
Bug 13089859 : THE VALUE OF V$SESSION.SQL_ID OF ACTIVE SESSION IS NULL | 11.2.0.2 | In some cases customer needs to find out SQL statements which are executed in active sessions and tune them. So they needs the values of SQL_ID in v$session.But that column occasionally have NULL values and then it make a difficulty in tuning problematic SQL statements. The following columns of v$session have NULL or zero (0) values even though the session status is ACTIVE: SQL_ID, SQL_ADDRESS, SQL_HASH_VALUE, SQL_EXEC_START, SQL_EXEC_ID. |
12.1 |
另附V$SESSION视图相关的BUG列表:
NB | Bug | Fixed | Description |
13545355 | 12.1.0.0 | V$SESSION.FIXED_TABLE_SEQUENCE may show negative values if DB up for a long time | |
10299006 | 11.2.0.2.GIPSU03, 11.2.0.3, 12.1.0.0 | Excess oraagent.bin sessions in the database (in V$SESSION) / ORA-20 | |
9339310 | 11.2.0.2, 12.1.0.0 | V$SESSION may not show correct SQL_ID | |
P | 7214679 | 11.2.0.2, 12.1.0.0 | OSUSER in V$SESSION is missing extended characters |
13068790 | 11.2.0.3 | the value of v$session.sql_id of active session is null | |
8655331 | 11.2.0.2 | V$SESSION.COMMAND has wrong code for commands >= 128 | |
6994490 | 10.2.0.5, 11.2.0.1 | Multibyte characters garbled in V$SESSION client information | |
6993310 | 11.2.0.1 | V$SESSION.CLIENT_INFO cannot be set to NULL with fix for bug 5915741 | |
6661393 | 10.2.0.5, 11.1.0.7, 11.2.0.1 | Setting CLIENT_INFO does not affect CLIENT_IDENTIFIER | |
8674660 | 10.2.0.5, 11.1.0.7 | DIAG process gets PGA memory corruption or ORA-600[kjzhablar:idx] or dump in RAC env | |
5928612 | 10.2.0.4, 11.1.0.7 | V$SESSION SQL_ADDRESS / SQL_HASH_VALUE not set for DBMS_JOBs | |
5915741 | 10.2.0.5, 11.1.0.6 | ORA-29275 selecting from V$SESSION with multibyte DB | |
5246867 | 10.2.0.3, 11.1.0.6 | V$SESSION.PROGRAM is not populated properly for JDBC Thin | |
5140631 | 10.2.0.4, 11.1.0.6 | V$SESSION.sql_address not set by DBMS_SCHEDULER | |
5078627 | 10.2.0.4, 11.1.0.6 | Audit sessionid is zero for jobs invoked by job scheduler | |
5010879 | 10.2.0.4, 11.1.0.6 | V$SESSION slow and does not show any BLOCKING_SESSION column data | |
4507211 | 10.2.0.3, 11.1.0.6 | Thin JDBC connection properties terminal not reflected in V$SESSION | |
4496189 | 9.2.0.8, 10.2.0.2, 11.1.0.6 | V$SESSION.PROGRAM can contain partial MULTIBYTE characters | |
4493741 | 10.2.0.4, 11.1.0.6 | Cannot see SQL_TEXT for procedure calls from EXECUTE IMMEDIATE | |
4383610 | 10.1.0.5, 10.2.0.2, 11.1.0.6 | application info attributes are not translated correctly in utf16 environment | |
3735857 | 10.2.0.2, 11.1.0.6 | V$SESSION.OSUSER not populated for JDBC clients | |
9322219 | 10.2.0.5.5 | Session dump with stack memcpy <- kjzhgigblk in RAC env | |
5884519 | 10.2.0.4 | V$SESSION is slow with fix for bug 5010879 | |
5481650 | 10.2.0.4 | GV$SESSION.blocking_session has incorrect value | |
4393134 | 10.2.0.1 | OracleConnectionCachImpl does not set connection properties (eg V$SESSION.PROGRAM) | |
3258390 | 9.2.0.6, 10.1.0.4, 10.2.0.1 | V$SESSION.SCHEMANAME may return wrong value in PLSQL of another user | |
2740805 | 9.2.0.5, 10.1.0.2 | V$SESSION.OS_USER set incorrectly if client uses RADIUS | |
P | 2661173 | 9.2.0.4, 10.1.0.2 | Linux: V$SESSION.PROGRAM shows full path instead of executable name |
P | 2628258 | 9.2.0.3, 10.1.0.2 | Win: Trailing “\0” added to MACHINE column of V$SESSION |
2123156 | 9.0.1.4, 9.2.0.2, 10.1.0.2 | FAILOVER: V$SESSION.FAILOVER_METHOD / FAILOVER _TYPE may be wrong for second connection | |
2106360 | 9.2.0.2, 10.1.0.2 | MODULE (V$SESSION/V$SQL) should default to the PROGRAM name | |
P | 2026123 | 8.1.7.3, 9.0.1.4, 9.2.0.1 | V$SESSION.PROGRAM displays ‘?’ for OCI clients |
1326191 | 9.2.0.1 | V$SESSION.OSUSER is always ‘ORACLE’ using JDBC Thin | |
1249631 | 8.1.7.3, 9.0.1.3, 9.2.0.1 | V$SESSION.SQL_ADDRESS is not cleared when a cursor is unmapped | |
1540012 | 9.0.1.0 | V$SESSION and V$SESSTAT “IDLE TIME” may not increment | |
1500535 | 8.1.7.1.B, 9.0.1.0 | V$SESSION.PROCESS not set by the THIN driver | |
1290469 | 8.1.6.3.J, 8.1.7.0 | V$SESSION does not show useful information for JDBC thin clients | |
1237128 | 8.1.7.0 | V$SESSION.PROGRAM may not contain a value | |
889678 | 8.1.7.0 | V$SESSION had short sizes for TERMINAL and MACHINE fields | |
P | 759086 | 7.3.4.4 | SP2: Program column in V$SESSION shows up as ? (broken fix) |
Leave a Reply