MySQL Bugs: #32951: stack badness in ndb

Bug #32951	stack badness in ndb
Submitted:	4 Dec 2007 8:04	Modified:	13 May 2008 8:45
Reporter:	Tatiana Azundris Nuernberg	Email Updates:
Status:	Can't repeat	Impact on me:	None
Category:	MySQL Cluster: Cluster (NDB) storage engine	Severity:	S3 (Non-critical)
Version:	6.0	OS:	Any
Assigned to:		CPU Architecture:	Any

Description:
tests fail with broken stack, corruption seems to appear in or below
NdbOperation::setVarValue in NdbOperationDefine.cpp

#6  0x089ddfc4 in __stack_chk_fail_local ()
#7  0x08840d4b in NdbOperation::setValue (this=0x8e7ef20, tAttrInfo=0x8e85180, 
    aValuePassed=0xb6b95bc0 "@\037lcrwtfsqabknvtfcbpoonxsjiqvmhnfikxxhcgoexlkoez
vahdbnfqyzgtqxalcrwtfsqabknvtfcbpoonxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxalcrw
tfsqabknvtfcbpoonxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxalcrwtfsqabknvtf"...)
    at NdbOperationDefine.cpp:621
#8  0x08840e9a in NdbOperation::setVarValue (this=0x8e7ef20, 
    tAttrInfo=0x8e85180, 
    aBareValue=0x8eb8be0 "lcrwtfsqabknvtfcbpoonxsjiqvmhnfikxxhcgoexlkoezvahdbnfq
yzgtqxalcrwtfsqabknvtfcbpoonxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxalcrwtfsqabkn
vtfcbpoonxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxalcrwtfsqabknvtfcb"..., 
    aLen=@0x8e7b230) at NdbOperationDefine.cpp:859
#9  0x08897ef6 in NdbBlob::setPartDataValue (this=0x8e7b150, anOp=0x8e7ef20, 
    buf=0x8eb8be0 "lcrwtfsqabknvtfcbpoonxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxa
lcrwtfsqabknvtfcbpoonxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxalcrwtfsqabknvtfcbpo
onxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxalcrwtfsqabknvtfcb"..., 
    aLen=@0x8e7b230) at NdbBlob.cpp:960
---Type <return> to continue, or q <return> to quit---
#10 0x0889aae1 in NdbBlob::insertPart (this=0x8e7b150, 
    buf=0x8eb8be0 "lcrwtfsqabknvtfcbpoonxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxa
lcrwtfsqabknvtfcbpoonxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxalcrwtfsqabknvtfcbpo
onxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxalcrwtfsqabknvtfcb"..., part=0, 
    len=@0x8e7b230) at NdbBlob.cpp:1686
#11 0x0889acff in NdbBlob::insertParts (this=0x8e7b150, 
    buf=0x8eb8be0 "lcrwtfsqabknvtfcbpoonxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxa
lcrwtfsqabknvtfcbpoonxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxalcrwtfsqabknvtfcbpo
onxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxalcrwtfsqabknvtfcb"..., part=0, 
    count=1) at NdbBlob.cpp:1673
#12 0x0889bc5e in NdbBlob::writeDataPrivate (this=0x8e7b150, 
    buf=0x8eb8be0 "lcrwtfsqabknvtfcbpoonxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxa
lcrwtfsqabknvtfcbpoonxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxalcrwtfsqabknvtfcbpo
onxsjiqvmhnfikxxhcgoexlkoezvahdbnfqyzgtqxalcrwtfsqabknvtfcb"..., bytes=20789)
    at NdbBlob.cpp:1499
#13 0x0889c504 in NdbBlob::preExecute (this=0x8e7b150, anExecType=NoCommit, 
    batch=@0xb6b99f2f) at NdbBlob.cpp:2310
#14 0x088344c0 in NdbTransaction::execute (this=0x8dd6c70, 
    aTypeOfExec=NdbTransaction::NoCommit, 
    abortOption=NdbOperation::AbortOnError, forceSend=1)
    at NdbTransaction.cpp:307
#15 0x085786d4 in execute_no_commit (h=0x8e69268, trans=0x8dd6c70, 
    force_release=false) at ha_ndbcluster.cc:297
---Type <return> to continue, or q <return> to quit---
#16 0x0856359c in ha_ndbcluster::flush_bulk_insert (this=0x8e69268)
    at ha_ndbcluster.cc:4587
#17 0x08567989 in ha_ndbcluster::ndb_write_row (this=0x8e69268, 
    record=0x8e73a98 "ü", primary_key_update=false, batched_update=false)
    at ha_ndbcluster.cc:3410
#18 0x085688e6 in ha_ndbcluster::write_row (this=0x8e69268, 
    record=0x8e73a98 "ü") at ha_ndbcluster.cc:3186
#19 0x08460206 in handler::ha_write_row (this=0x8e69268, buf=0x8e73a98 "ü")
    at handler.cc:4547
#20 0x083c8156 in write_record (thd=0x8e005b8, table=0x8e796c0, 
    info=0xb6b9a2e8) at sql_insert.cc:1549
#21 0x083cd35d in mysql_insert (thd=0x8e005b8, table_list=0x8df9b18, 
    fields=@0x8e0193c, values_list=@0x8e01960, update_fields=@0x8e01954, 
    update_values=@0x8e01948, duplic=DUP_ERROR, ignore=false)
    at sql_insert.cc:803
#22 0x08326d7a in mysql_execute_command (thd=0x8e005b8) at sql_parse.cc:2608
#23 0x0832d673 in mysql_parse (thd=0x8e005b8, 
    inBuf=0x8df9a18 "insert into t1 values (\n0, repeat(@s2,454), 100, repeat(@s
2,345), NULL)", length=71, found_semicolon=0xb6b9ae98) at sql_parse.cc:5369
#24 0x0832e285 in dispatch_command (command=COM_QUERY, thd=0x8e005b8, 
    packet=0x8e3baa9 "insert into t1 values (\n0, repeat(@s2,454), 100, repeat(@
s2,345), NULL)", packet_length=71) at sql_parse.cc:890
#25 0x0832f734 in do_command (thd=0x8e005b8) at sql_parse.cc:657
---Type <return> to continue, or q <return> to quit---
#26 0x0831bf72 in handle_one_connection (arg=0x8e005b8) at sql_connect.cc:1119
#27 0xb7ec6112 in start_thread () from /lib/libpthread.so.0
#28 0xb7cd248e in clone () from /lib/libc.so.6

How to repeat:
compile with -fstack-protector-all

run any of the flwg tests:
rpl_ndb.rpl_ndb_blob rpl_ndb.rpl_ndb_blob2 ndb.ndb_blob ndb.ndb_dd_backuprestore

Suggested fix:
find out what happens, and whether it's toxic

Thank you for the report.

I can not repeat described behavior with current mysql-6.0 BK sources on Linux. Please try with current sources in your environment and if bug is still repeatable indicate your operating system and configure options you used.

No feedback was provided for this bug for over a month, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".

cannot repeat with 5.1-bk and 6.0-bk, presumed fixed. :-S
thank you for your time and effort!