| Bug #47329 | Endless 4006 in startTransaction()/getScanOperation() + node-failure scenario | ||
|---|---|---|---|
| Submitted: | 15 Sep 2009 13:56 | Modified: | 16 Sep 2009 13:38 |
| Reporter: | Jonas Oreland | Email Updates: | |
| Status: | Closed | Impact on me: | |
| Category: | MySQL Cluster: Cluster (NDB) storage engine | Severity: | S3 (Non-critical) |
| Version: | mysql-5.1-telco-6.2 | OS: | Any |
| Assigned to: | Jonas Oreland | CPU Architecture: | Any |
[15 Sep 2009 16:52]
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/83313 2999 Jonas Oreland 2009-09-15 ndb - bug#47329 - release unusable NdbTransaction-object on closeTransaction
[15 Sep 2009 17:02]
Bugs System
Pushed into 5.1.37-ndb-6.3.27 (revid:jonas@mysql.com-20090915165710-mro7wyb4547aqje1) (version source revid:jonas@mysql.com-20090915165710-mro7wyb4547aqje1) (merge vers: 5.1.37-ndb-6.3.27) (pib:11)
[15 Sep 2009 17:02]
Bugs System
Pushed into 5.1.37-ndb-7.0.8 (revid:jonas@mysql.com-20090915165909-j9hf2cmeavek8t5l) (version source revid:jonas@mysql.com-20090915165909-j9hf2cmeavek8t5l) (merge vers: 5.1.37-ndb-7.0.8) (pib:11)
[15 Sep 2009 17:03]
Bugs System
Pushed into 5.1.35-ndb-7.1.0 (revid:jonas@mysql.com-20090915170148-jh1wiy6upn3yipjl) (version source revid:jonas@mysql.com-20090915170148-jh1wiy6upn3yipjl) (merge vers: 5.1.35-ndb-7.1.0) (pib:11)
[15 Sep 2009 17:06]
Jonas Oreland
also pushed to 6.2.19
[15 Sep 2009 20:40]
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/83380 3008 Martin Skold 2009-09-15 [merge] Merge added: storage/ndb/test/run-test/conf-ndb07.cnf modified: storage/ndb/src/common/portlib/NdbDir.cpp storage/ndb/src/kernel/blocks/ERROR_codes.txt storage/ndb/src/kernel/blocks/dbdih/DbdihMain.cpp storage/ndb/src/kernel/blocks/ndbfs/AsyncFile.cpp storage/ndb/src/ndbapi/Ndb.cpp storage/ndb/test/ndbapi/testSystemRestart.cpp storage/ndb/test/run-test/Makefile.am storage/ndb/test/src/UtilTransactions.cpp
[15 Sep 2009 21:04]
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/83385 3010 Martin Skold 2009-09-15 [merge] Merge added: storage/ndb/test/run-test/conf-ndb07.cnf modified: storage/ndb/src/common/portlib/NdbDir.cpp storage/ndb/src/kernel/blocks/ERROR_codes.txt storage/ndb/src/kernel/blocks/dbdih/DbdihMain.cpp storage/ndb/src/ndbapi/Ndb.cpp storage/ndb/test/ndbapi/testSystemRestart.cpp storage/ndb/test/run-test/Makefile.am storage/ndb/test/src/UtilTransactions.cpp
[15 Sep 2009 21:05]
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/83388 3051 Martin Skold 2009-09-15 [merge] Merge added: storage/ndb/test/run-test/conf-ndb07.cnf modified: storage/ndb/src/kernel/blocks/ERROR_codes.txt storage/ndb/src/kernel/blocks/dbdih/DbdihMain.cpp storage/ndb/src/ndbapi/Ndb.cpp storage/ndb/test/ndbapi/testSystemRestart.cpp storage/ndb/test/run-test/Makefile.am storage/ndb/test/src/UtilTransactions.cpp
[15 Sep 2009 21:10]
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/83390 2996 Martin Skold 2009-09-15 [merge] Merge modified: storage/ndb/src/ndbapi/Ndb.cpp storage/ndb/test/src/UtilTransactions.cpp
[16 Sep 2009 13:38]
Jon Stephens
Documetned bugfix in the NDB-6.2.19, 6.3.27, and 7.0.8 changelogs, as follows:
In some circumstances, if an API node encountered a data node
failure between the creation of a transaction and the start of a
scan using that transaction, then any subsequent calls to
startTransaction() and closeTransaction() could cause the same
transaction to be started and closed repeatedly.
Closed.
Description: Using a pattern like this 0 loop: 1 p = startTransaction(); 2 o = p->getNdbScanOperation(); 3 if (o == 0 && p->getNdbError().status == NdbError::TemporaryError) 4 { 5 p->close(); 6 goto loop; 7 } Could in some (rare) scenarios end up in infinite loop. (if api-node discovered node-failure in between 1 and 2) This as getNdbScanOperation() will detect if node (that p is connected to) is dead, but startTransaction()/closeTransaction() will not, causing the same transaction to be started/closed endlessly. How to repeat: "testIndex -n NFNR3 T6 T13" reproduces semi-consistent in autotest. Suggested fix: in closeTransaction() (step 5 above) check if sequence no is correct, and if not remove the transaction from pool transactions allocated from ndbd.