SIGSEGV in Firebird 5.0.0.1306 embedded during update on cursor #8185

FabianoBiagini · 2024-07-18T15:26:55Z

The program was executing the following query:

update scheduler set 
  id_sched=gen_id(sched_gen,1),
  wndopen=current_timestamp,
  wndclose=current_timestamp+(cast(apertura as numeric(9,5))/1440.0),
  cache_vect=(select pv1.rif_vect from arcpervect pv1 where pv1.rif_per=scheduler.rif_per and pv1.idx_vect=1),
  lastresult=null
where current of SCHEDPEEK

Stack trace:

Program terminated with signal SIGSEGV, Segmentation fault.
#0  __memcpy_sse2_unaligned () at ../sysdeps/x86_64/multiarch/memcpy-sse2-unaligned.S:157
157     ../sysdeps/x86_64/multiarch/memcpy-sse2-unaligned.S: File o directory non esistente.
(gdb) bt
#0  __memcpy_sse2_unaligned () at ../sysdeps/x86_64/multiarch/memcpy-sse2-unaligned.S:157
#1  0x00007f6d4f705824 in MOVD_move(Jrd::thread_db*, dsc*, dsc*) () at /firebird/src/dsql/movd.cpp:40
#2  0x00007f6d4f6381e4 in Jrd::DsqlDmlRequest::mapInOut(Jrd::thread_db*, bool, Jrd::dsql_msg const*, Firebird::IMessageMetadata*, unsigned char*, unsigned char const*) () at /firebird/src/dsql/DsqlRequests.cpp:974
#3  0x00007f6d4f63a985 in Jrd::DsqlDmlRequest::execute(Jrd::thread_db*, Jrd::jrd_tra**, Firebird::IMessageMetadata*, unsigned char const*, Firebird::IMessageMetadata*, unsigned char*, bool) () at /firebird/src/dsql/DsqlRequests.cpp:723
#4  0x00007f6d4f6ecd9b in DSQL_execute_immediate(Jrd::thread_db*, Jrd::Attachment*, Jrd::jrd_tra**, unsigned int, char const*, unsigned short, Firebird::IMessageMetadata*, unsigned char const*, Firebird::IMessageMetadata*, unsigned char*, bool) () at /firebird/src/dsql/dsql.cpp:365
#5  0x00007f6d4f4fff02 in Jrd::JAttachment::execute(Firebird::CheckStatusWrapper*, Firebird::ITransaction*, unsigned int, char const*, unsigned int, Firebird::IMessageMetadata*, void*, Firebird::IMessageMetadata*, void*) () at /firebird/src/jrd/../jrd/jrd.h:810
#6  0x00007f6d4f51a4cf in Firebird::IAttachmentBaseImpl<Jrd::JAttachment, Firebird::CheckStatusWrapper, Firebird::IReferenceCountedImpl<Jrd::JAttachment, Firebird::CheckStatusWrapper, Firebird::Inherit<Firebird::IVersionedImpl<Jrd::JAttachment, Firebird::CheckStatusWrapper, Firebird::Inherit<Firebird::IAttachment> > > > >::cloopexecuteDispatcher (self=<optimized out>, status=0x7ffc5ad86388, transaction=0x7f6d52619248, stmtLength=0,
    sqlStmt=0x84ecb0 "update scheduler set id_sched=gen_id(sched_gen,1),wndopen=current_timestamp,wndclose=current_timestamp+(cast(apertura as numeric(9,5))/1440.0),  cache_vect=(select pv1.rif_vect from arcpervect pv1 whe"..., dialect=3, inMetadata=0x0, inBuffer=0x7ffc5ad863b8, outMetadata=0x0,
    outBuffer=0x0) at /firebird/src/include/firebird/IdlFbInterfaces.h:7545
#7  0x00007f6d52350fba in ?? () from /usr/lib/libfbembed.so.2.5
#8  0x00007f6d5236b16c in ?? () from /usr/lib/libfbembed.so.2.5
#9  0x00007f6d5236ff1d in isc_dsql_exec_immed2 () from /usr/lib/libfbembed.so.2.5
#10 0x00007f6d52370155 in isc_dsql_execute_immediate () from /usr/lib/libfbembed.so.2.5
#11 0x000000000053f2eb in TInterbaseConnection::ImmediateSQL (this=0x2370d70,
    sql=0x84ecb0 "update scheduler set id_sched=gen_id(sched_gen,1),wndopen=current_timestamp,wndclose=current_timestamp+(cast(apertura as numeric(9,5))/1440.0),  cache_vect=(select pv1.rif_vect from arcpervect pv1 whe"..., tr=0x7ffc5ad869b0) at common/gdb_ibase.cpp:269

Registers:

(gdb) info reg
rax            0xffff8092b52c6578       -140107383544456
rbx            0x7f6d4ad398c0   140107383544000
rcx            0x10     16
rdx            0x8      8
rsi            0x8      8
rdi            0x7f6d4ad39a88   140107383544456
rbp            0x7ffc5ad85c30   0x7ffc5ad85c30
rsp            0x7ffc5ad85be8   0x7ffc5ad85be8
r8             0x0      0
r9             0x202df  131807
r10            0x7ffc5ad85c90   140721832615056
r11            0x7ffc5ad85ca0   140721832615072
r12            0x7f6d4ad39530   140107383543088
r13            0x7f6d4ad39430   140107383542832
r14            0x7f6d4ad39a80   140107383544448
r15            0x7ffc5ad85fa8   140721832615848
rip            0x7f6d518558ed   0x7f6d518558ed <__memcpy_sse2_unaligned+493>
eflags         0x10202  [ IF RF ]
cs             0x33     51
ss             0x2b     43
ds             0x0      0
es             0x0      0
fs             0x0      0
gs             0x0      0

Shared library:

(gdb) info sharedlibrary
From                To                  Syms Read   Shared Object Library
                                        No          linux-vdso.so.1
0x00007f6d522efa70  0x00007f6d524be7a1  Yes (*)     /usr/lib/libfbembed.so.2.5
0x00007f6d52091220  0x00007f6d520a2469  Yes (*)     /lib/x86_64-linux-gnu/libz.so.1
0x00007f6d51ddf8b0  0x00007f6d51e47b1f  Yes (*)     /usr/lib/x86_64-linux-gnu/libstdc++.so.6
0x00007f6d51b70ab0  0x00007f6d51b809a5  Yes (*)     /lib/x86_64-linux-gnu/libgcc_s.so.1
0x00007f6d517e44a0  0x00007f6d5190e6a3  Yes         /lib/x86_64-linux-gnu/libc.so.6
0x00007f6d515bf350  0x00007f6d515c206c  Yes         /lib/x86_64-linux-gnu/librt.so.1
0x00007f6d513b9ed0  0x00007f6d513ba97e  Yes         /lib/x86_64-linux-gnu/libdl.so.2
0x00007f6d510bd580  0x00007f6d51128d96  Yes         /lib/x86_64-linux-gnu/libm.so.6
0x00007f6d50ea09f0  0x00007f6d50eac731  Yes         /lib/x86_64-linux-gnu/libpthread.so.0
0x00007f6d50c86d20  0x00007f6d50c9784f  Yes (*)     /usr/lib/../lib/libtommath.so.1
0x00007f6d52592ae0  0x00007f6d525ab170  Yes         /lib64/ld-linux-x86-64.so.2
0x00007f6d4f2cca20  0x00007f6d4f98b931  Yes         /opt/firebird/plugins/libEngine13.so
0x00007f6d526f5e90  0x00007f6d52731f57  Yes (*)     /opt/firebird/plugins/../lib/libtomcrypt.so.1
0x00007f6d526b7060  0x00007f6d526b7177  Yes (*)     /opt/firebird/lib/libib_util.so
0x00007f6d4f0c7d00  0x00007f6d4f174161  Yes (*)     /opt/firebird/intl/fbintl
0x00007f6d4ed7f6f0  0x00007f6d4ee33818  Yes (*)     /usr/lib/x86_64-linux-gnu/libicuuc.so.52.1
0x00007f6d4d4c0570  0x00007f6d4d4c0670  Yes (*)     /usr/lib/x86_64-linux-gnu/libicudata.so.52
0x00007f6d4d148770  0x00007f6d4d25ba15  Yes (*)     /usr/lib/x86_64-linux-gnu/libicui18n.so.52.1
0x00007f6d4cea73b0  0x00007f6d4ceabb4e  Yes         /lib/x86_64-linux-gnu/libnss_compat.so.2
0x00007f6d4cc92160  0x00007f6d4cc9d693  Yes         /lib/x86_64-linux-gnu/libnsl.so.1
0x00007f6d4ca851a0  0x00007f6d4ca8b1bc  Yes         /lib/x86_64-linux-gnu/libnss_nis.so.2
0x00007f6d4c8792a0  0x00007f6d4c87f803  Yes         /lib/x86_64-linux-gnu/libnss_files.so.2
0x00007f6d4c6aff50  0x00007f6d4c7ebf81  Yes (*)     /opt/firebird/plugins/libfbtrace.so
(*): Shared library is missing debugging information.

/usr/lib/libfbembed.so.2.5 is a link to /opt/firebird/lib/libfbclient.so for compatibility reasons.

/opt/firebird/firebird.conf:

ServerMode = Classic
RelaxedAliasChecking=1

I hope this can help.

The text was updated successfully, but these errors were encountered:

AlexPeshkoff · 2024-07-18T15:48:39Z

Can you also provide xz-compressed core dump?

FabianoBiagini · 2024-07-18T15:56:57Z

The core dump contains too many personal information of customers, but I can reproduce the problem with an anonymized version of the DB and send you the new core dump, maybe tomorrow

aafemt · 2024-07-18T16:18:16Z

Could you show us query that defines cursor SCHEDPEEK?

FabianoBiagini · 2024-07-19T06:45:00Z

Of course

select rif_per,rif_class,flags,giorni,apertura,loop 
from scheduler s 
where (s.alle=? and (s.giorni is null or s.giorni containing ? or s.giorni containing '8')) 
 and (s.wndclose is null or s.wndclose<current_timestamp) 
for update of id_sched,cache_vect,wndopen,wndclose,lastresult

aafemt · 2024-07-19T07:49:39Z

Is it opened as an unidirectional or scrollable cursor?

FabianoBiagini · 2024-07-19T07:53:10Z

Unidirectional, the code is written for Firebird 2.5

FabianoBiagini · 2024-07-19T08:42:37Z

Here we go!
I have reproduced the problem with an almost empty DB, so I can send you the cordeump.

(gdb) bt
#0  __memcpy_sse2_unaligned () at ../sysdeps/x86_64/multiarch/memcpy-sse2-unaligned.S:157
#1  0x00007fe3be747824 in MOVD_move(Jrd::thread_db*, dsc*, dsc*) () at /firebird/src/dsql/movd.cpp:40
#2  0x00007fe3be67a1e4 in Jrd::DsqlDmlRequest::mapInOut(Jrd::thread_db*, bool, Jrd::dsql_msg const*, Firebird::IMessageMetadata*, unsigned char*, unsigned char const*)
    () at /firebird/src/dsql/DsqlRequests.cpp:974
#3  0x00007fe3be67c985 in Jrd::DsqlDmlRequest::execute(Jrd::thread_db*, Jrd::jrd_tra**, Firebird::IMessageMetadata*, unsigned char const*, Firebird::IMessageMetadata*, unsigned char*, bool) () at /firebird/src/dsql/DsqlRequests.cpp:723
#4  0x00007fe3be72ed9b in DSQL_execute_immediate(Jrd::thread_db*, Jrd::Attachment*, Jrd::jrd_tra**, unsigned int, char const*, unsigned short, Firebird::IMessageMetadata*, unsigned char const*, Firebird::IMessageMetadata*, unsigned char*, bool) () at /firebird/src/dsql/dsql.cpp:365
#5  0x00007fe3be541f02 in Jrd::JAttachment::execute(Firebird::CheckStatusWrapper*, Firebird::ITransaction*, unsigned int, char const*, unsigned int, Firebird::IMessageMetadata*, void*, Firebird::IMessageMetadata*, void*) () at /firebird/src/jrd/../jrd/jrd.h:810
#6  0x00007fe3be55c4cf in Firebird::IAttachmentBaseImpl<Jrd::JAttachment, Firebird::CheckStatusWrapper, Firebird::IReferenceCountedImpl<Jrd::JAttachment, Firebird::CheckStatusWrapper, Firebird::Inherit<Firebird::IVersionedImpl<Jrd::JAttachment, Firebird::CheckStatusWrapper, Firebird::Inherit<Firebird::IAttachment> > > > >::cloopexecuteDispatcher (self=<optimized out>, status=0x7ffc3f8719c8, transaction=0x7fe3c165b228, stmtLength=0,
    sqlStmt=0x84ecb0 "update scheduler set id_sched=gen_id(sched_gen,1),wndopen=current_timestamp,wndclose=current_timestamp+(cast(apertura as numeric(9,5))/1440.0),  cache_vect=(select pv1.rif_vect from arcpervect pv1 whe"..., dialect=3, inMetadata=0x0, inBuffer=0x7ffc3f8719f8, outMetadata=0x0, outBuffer=0x0)
    at /firebird/src/include/firebird/IdlFbInterfaces.h:7545
#7  0x00007fe3c1392fba in ?? () from /usr/lib/libfbembed.so.2.5
#8  0x00007fe3c13ad16c in ?? () from /usr/lib/libfbembed.so.2.5
#9  0x00007fe3c13b1f1d in isc_dsql_exec_immed2 () from /usr/lib/libfbembed.so.2.5
#10 0x00007fe3c13b2155 in isc_dsql_execute_immediate () from /usr/lib/libfbembed.so.2.5
#11 0x000000000053f2eb in TInterbaseConnection::ImmediateSQL (this=0x12a2d70,
    sql=0x84ecb0 "update scheduler set id_sched=gen_id(sched_gen,1),wndopen=current_timestamp,wndclose=current_timestamp+(cast(apertura as numeric(9,5))/1440.0),  cache_vect=(select pv1.rif_vect from arcpervect pv1 whe"..., tr=0x7ffc3f871ff0) at common/gdb_ibase.cpp:269

(gdb) info sharedlibrary
From                To                  Syms Read   Shared Object Library
                                        No          linux-vdso.so.1
0x00007fe3c1331a70  0x00007fe3c15007a1  Yes (*)     /usr/lib/libfbembed.so.2.5
0x00007fe3c10d3220  0x00007fe3c10e4469  Yes (*)     /lib/x86_64-linux-gnu/libz.so.1
0x00007fe3c0e218b0  0x00007fe3c0e89b1f  Yes (*)     /usr/lib/x86_64-linux-gnu/libstdc++.so.6
0x00007fe3c0bb2ab0  0x00007fe3c0bc29a5  Yes (*)     /lib/x86_64-linux-gnu/libgcc_s.so.1
0x00007fe3c08264a0  0x00007fe3c09506a3  Yes         /lib/x86_64-linux-gnu/libc.so.6
0x00007fe3c0601350  0x00007fe3c060406c  Yes         /lib/x86_64-linux-gnu/librt.so.1
0x00007fe3c03fbed0  0x00007fe3c03fc97e  Yes         /lib/x86_64-linux-gnu/libdl.so.2
0x00007fe3c00ff580  0x00007fe3c016ad96  Yes         /lib/x86_64-linux-gnu/libm.so.6
0x00007fe3bfee29f0  0x00007fe3bfeee731  Yes         /lib/x86_64-linux-gnu/libpthread.so.0
0x00007fe3bfcc8d20  0x00007fe3bfcd984f  Yes (*)     /usr/lib/../lib/libtommath.so.1
0x00007fe3c15d4ae0  0x00007fe3c15ed170  Yes         /lib64/ld-linux-x86-64.so.2
0x00007fe3be30ea20  0x00007fe3be9cd931  Yes         /opt/firebird/plugins/libEngine13.so
0x00007fe3c1737e90  0x00007fe3c1773f57  Yes (*)     /opt/firebird/plugins/../lib/libtomcrypt.so.1
0x00007fe3c16f9060  0x00007fe3c16f9177  Yes (*)     /opt/firebird/lib/libib_util.so
0x00007fe3be109d00  0x00007fe3be1b6161  Yes (*)     /opt/firebird/intl/fbintl
0x00007fe3bddc16f0  0x00007fe3bde75818  Yes (*)     /usr/lib/x86_64-linux-gnu/libicuuc.so.52.1
0x00007fe3bc502570  0x00007fe3bc502670  Yes (*)     /usr/lib/x86_64-linux-gnu/libicudata.so.52
0x00007fe3bc18a770  0x00007fe3bc29da15  Yes (*)     /usr/lib/x86_64-linux-gnu/libicui18n.so.52.1
0x00007fe3bbee93b0  0x00007fe3bbeedb4e  Yes         /lib/x86_64-linux-gnu/libnss_compat.so.2
0x00007fe3bbcd4160  0x00007fe3bbcdf693  Yes         /lib/x86_64-linux-gnu/libnsl.so.1
0x00007fe3bbac71a0  0x00007fe3bbacd1bc  Yes         /lib/x86_64-linux-gnu/libnss_nis.so.2
0x00007fe3bb8bb2a0  0x00007fe3bb8c1803  Yes         /lib/x86_64-linux-gnu/libnss_files.so.2
0x00007fe3bb6f1f50  0x00007fe3bb82df81  Yes (*)     /opt/firebird/plugins/libfbtrace.so

(gdb) info reg
rax            0xffff801c46a727a8       -140616043911256
rbx            0x7fe3b958d6c0   140616043910848
rcx            0x10     16
rdx            0x8      8
rsi            0x8      8
rdi            0x7fe3b958d858   140616043911256
rbp            0x7ffc3f871270   0x7ffc3f871270
rsp            0x7ffc3f871228   0x7ffc3f871228
r8             0x0      0
r9             0x202df  131807
r10            0x7ffc3f8712d0   140721374302928
r11            0x7ffc3f8712e0   140721374302944
r12            0x7fe3b958d330   140616043909936
r13            0x7fe3b958d230   140616043909680
r14            0x7fe3b958d850   140616043911248
r15            0x7ffc3f8715e8   140721374303720
rip            0x7fe3c08978ed   0x7fe3c08978ed <__memcpy_sse2_unaligned+493>
eflags         0x10202  [ IF RF ]
cs             0x33     51
ss             0x2b     43
ds             0x0      0
es             0x0      0
fs             0x0      0
gs             0x0      0

core.6617.gz

FabianoBiagini · 2024-07-19T08:48:46Z

I forgot some possibly useful information:

the "select ... for update" run every minute,
the first minute it finds a record, process with "update ... where current of ..." and everything ok! no SIGSEGV
the second minute: "select ... for update" restart, find another record and process it also with "update ... where current of ...",
now it trigger the SIGSEGV

The code runs well with Firebird 2.5 and 3.0.

dyemanov · 2024-07-19T09:32:35Z

It seems I'm able to reproduce it. Working on it.

aafemt · 2024-07-20T12:32:20Z

The code runs well with Firebird 2.5 and 3.0.

Try to set MaxStatementCacheSize to zero.

FabianoBiagini · 2024-07-22T07:12:32Z

Try to set MaxStatementCacheSize to zero.

You are right, the bug is evidently related to the statement cache. Putting the cache size to zero make the SIGSEGV disappear, but also disappear performance gaining, the main reason for me to adopt FB 5.0

aafemt · 2024-07-22T08:19:31Z

Do you mean performance of this particular query or performance of your application as whole? I can't believe that preparation of this "update" can be slow.

FabianoBiagini · 2024-07-22T08:30:35Z

Do you mean performance of this particular query or performance of your application as whole? I can't believe that preparation of this "update" can be slow.

I mean performance as measured by my personal benchmark with prepared and unprepared query loop. I can't say that preparation of this specific "update" is slow.

aafemt · 2024-07-22T09:06:11Z

In this case I think that the simplest fix would be to disable caching of positional DML statements.

dyemanov · 2024-07-22T09:19:01Z

I've already implemented this, but honestly expected to do something more clever.

aafemt · 2024-07-22T09:39:38Z

Alternatives could be "partial release/prepare" mechanic which is too big work for v5 or parent statement lookup on each execution which may be not good for performance.

asfernandes · 2024-07-22T11:05:12Z

I forgot some possibly useful information:

the "select ... for update" run every minute,

the first minute it finds a record, process with "update ... where current of ..." and everything ok! no SIGSEGV

the second minute: "select ... for update" restart, find another record and process it also with "update ... where current of ...",

now it trigger the SIGSEGV

The code runs well with Firebird 2.5 and 3.0.

Is this dependent on table and query structure?
Is this "every minute" time related really or execution count related? Or just random?

I've tried simple table with many executions and not reproduced it.

asfernandes · 2024-07-22T11:37:01Z

In my tests, I'm having error invalid request handle. But I imagine, if many more statements run, that orphan statement may go out of cache and crash happens.

aafemt · 2024-07-22T11:39:31Z

The worse is the case when there are two parent statements in the cache and by chance DML statement is used referring to the wrong SELECT statement.

asfernandes · 2024-07-22T11:43:06Z

The worse is the case when there are two parent statements in the cache and by chance DML statement is used referring to the wrong SELECT statement.

Cursor names are attached to dsql requests. What is cached is dsql statements.

aafemt · 2024-07-22T11:47:58Z

Right. But still actual problem is that parentRequest, parentDbKey and parentRecVersion are members of DsqlStatement.

asfernandes · 2024-07-22T11:52:10Z

@dyemanov let me know if you are actively working on a fix or want that I look at it.

FabianoBiagini · 2024-07-22T13:06:50Z

I've just made two tests.

Is this "every minute" time related really or execution count related? Or just random?

I removed the time gap and forced the scheduler to run for "10:00" and then for "10:01" without any delay between the calls, the SIGSEGV appeared in the same way. So the one minute delay is not involved.

In my tests, I'm having error invalid request handle.

This happen to me too if I try to rewrite the code in attempt to reproduce the bug in a simpler situation.
So, using the same "select .. for update" and the same "update ... where current of" in a simplified piece of code I obtain only the invalid request handle and no more the SIGSEGV. Please note that in the production scenario there are other concurrent processes that could be accessing the same DB.

aafemt · 2024-07-23T08:07:34Z

honestly expected to do something more clever.

I'll have to solve it anyway in #8145 so you don't need to spend much of your time to find something more clever.

…che turned on.

…g update on cursor - disallow caching for positioned updates/deletes

pavel-zotov · 2024-07-29T16:55:44Z

QA note.
Could not reproduce using script from attached .sql
gh-8185.sql.zip

aafemt · 2024-07-29T19:03:20Z

@pavel-zotov PSQL cursors and DSQL cursors are completely different beasts. This problem cannot happen with PSQL cursors which you are trying to use.

aafemt · 2024-07-30T15:14:34Z

@pavel-zotov Steps to reproduce should be like this:

Prepare statement 1 "select ... for update"
Set cursor name for statement 1
Prepare statement 2 "update ... where current of <cursor name from step 2>"
Prepare statement 3 similar to statement 1
Release statement 1
Release statement 2
Set cursor name to statement 3 as in step 2
Prepare statement 2 again (it will be got from cache keeping reference to statement 1)
Run statement 3 and fetch one record
Execute statement 2

At step 10 you can get "Invalid handle" error or a crash if you swap steps 5 and 6.
ISQL cannot perform steps 2 and 9 but I believe Python program can.

pavel-zotov · 2024-11-05T13:54:52Z

::: QA NOTE :::
Test has been implemented and uses steps described above by aafemt.
Checked on snapshots 5.0.1.1452-08dc25f (27.07.2024 11:50) and 6.0.0.401-a7d10a4 (29.07.2024 01:33) -- all OK.
Current FB 6.x also works fine.
But it is not so for fresh FB 5.x: it fails since 2024-10-23, commit: c84d945

…che turned on.

dyemanov self-assigned this Jul 19, 2024

dyemanov added component: engine type: bug affect-version: 6.0 Initial affect-version: 5.0.0 labels Jul 19, 2024

asfernandes added a commit that referenced this issue Jul 23, 2024

Fix #8185 - SIGSEGV with WHERE CURRENT OF statement with statement ca…

83c97a2

…che turned on.

dyemanov added a commit that referenced this issue Jul 27, 2024

Simplest fix for #8185: SIGSEGV in Firebird 5.0.0.1306 embedded durin…

08dc25f

…g update on cursor - disallow caching for positioned updates/deletes

dyemanov added the fix-version: 5.0.1 label Jul 27, 2024

asfernandes closed this as completed in a7d10a4 Jul 29, 2024

asfernandes added the fix-version: 6.0 Alpha 1 label Jul 29, 2024

asfernandes self-assigned this Jul 29, 2024

aafemt mentioned this issue Jul 29, 2024

Fix for #8082 by making engine to use user buffers directly #8145

Open

pavel-zotov added the qa: not enough information label Jul 29, 2024

pavel-zotov added qa: done successfully and removed qa: not enough information labels Nov 1, 2024

dyemanov pushed a commit that referenced this issue Nov 16, 2024

Fix #8185 - SIGSEGV with WHERE CURRENT OF statement with statement ca…

5ab64aa

…che turned on.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

SIGSEGV in Firebird 5.0.0.1306 embedded during update on cursor #8185

SIGSEGV in Firebird 5.0.0.1306 embedded during update on cursor #8185

FabianoBiagini commented Jul 18, 2024

AlexPeshkoff commented Jul 18, 2024

FabianoBiagini commented Jul 18, 2024

aafemt commented Jul 18, 2024

FabianoBiagini commented Jul 19, 2024

aafemt commented Jul 19, 2024

FabianoBiagini commented Jul 19, 2024

FabianoBiagini commented Jul 19, 2024

FabianoBiagini commented Jul 19, 2024 •

edited

Loading

dyemanov commented Jul 19, 2024

aafemt commented Jul 20, 2024

FabianoBiagini commented Jul 22, 2024

aafemt commented Jul 22, 2024

FabianoBiagini commented Jul 22, 2024

aafemt commented Jul 22, 2024

dyemanov commented Jul 22, 2024

aafemt commented Jul 22, 2024

asfernandes commented Jul 22, 2024

asfernandes commented Jul 22, 2024

aafemt commented Jul 22, 2024

asfernandes commented Jul 22, 2024

aafemt commented Jul 22, 2024

asfernandes commented Jul 22, 2024

FabianoBiagini commented Jul 22, 2024

aafemt commented Jul 23, 2024

pavel-zotov commented Jul 29, 2024

aafemt commented Jul 29, 2024

aafemt commented Jul 30, 2024

pavel-zotov commented Nov 5, 2024

SIGSEGV in Firebird 5.0.0.1306 embedded during update on cursor #8185

SIGSEGV in Firebird 5.0.0.1306 embedded during update on cursor #8185

Comments

FabianoBiagini commented Jul 18, 2024

AlexPeshkoff commented Jul 18, 2024

FabianoBiagini commented Jul 18, 2024

aafemt commented Jul 18, 2024

FabianoBiagini commented Jul 19, 2024

aafemt commented Jul 19, 2024

FabianoBiagini commented Jul 19, 2024

FabianoBiagini commented Jul 19, 2024

FabianoBiagini commented Jul 19, 2024 • edited Loading

dyemanov commented Jul 19, 2024

aafemt commented Jul 20, 2024

FabianoBiagini commented Jul 22, 2024

aafemt commented Jul 22, 2024

FabianoBiagini commented Jul 22, 2024

aafemt commented Jul 22, 2024

dyemanov commented Jul 22, 2024

aafemt commented Jul 22, 2024

asfernandes commented Jul 22, 2024

asfernandes commented Jul 22, 2024

aafemt commented Jul 22, 2024

asfernandes commented Jul 22, 2024

aafemt commented Jul 22, 2024

asfernandes commented Jul 22, 2024

FabianoBiagini commented Jul 22, 2024

aafemt commented Jul 23, 2024

pavel-zotov commented Jul 29, 2024

aafemt commented Jul 29, 2024

aafemt commented Jul 30, 2024

pavel-zotov commented Nov 5, 2024

FabianoBiagini commented Jul 19, 2024 •

edited

Loading