in_stdout: overflow when reading in_mem generated msgpack maps #6
Comments
edsiper
added a commit
that referenced
this issue
Jun 3, 2015
in_mem: fix overflow when reading in_mem generated msgpack maps #6
|
fixed by #7 |
fujimotos
pushed a commit
to fujimotos/fluent-bit
that referenced
this issue
Jul 22, 2019
development: modify return value of API
fujimotos
pushed a commit
to fujimotos/fluent-bit
that referenced
this issue
Jan 15, 2020
When Fluent Bit encounters with a partial parser definition, it
crashes badly with a segmentation fault.
$ ./bin/fluent-bit -R parser.conf -c tail.conf
...
[2020/01/15 16:11:21] [error] [parser] no parser 'format' found for 'simple' in file 'conf/timestamp.parser'
[engine] caught signal (SIGSEGV)
#0 0x558bc4a0a226 in flb_parser_decoder_list_destroy() at src/flb_parser_decoder.c:700
fluent#1 0x558bc4a05d75 in flb_parser_conf_file() at src/flb_parser.c:566
fluent#2 0x558bc49f4bdd in flb_config_set_property() at src/flb_config.c:406
fluent#3 0x558bc49e24ae in flb_service_conf() at src/fluent-bit.c:446
fluent#4 0x558bc49e2f90 in main() at src/fluent-bit.c:807
fluent#5 0x7fa1cb7f109a in ???() at ???:0
fluent#6 0x558bc49e13a9 in ???() at ???:0
fluent#7 0xffffffffffffffff in ???() at ???:0
Aborted
This is just because `decoders` is not being initialized properly,
and that confuses Fluent Bit to deallocate a random memmory block
on the cleanup path. Fix it.
Signed-off-by: Fujimoto Seiji <fujimoto@clear-code.com>
edsiper
pushed a commit
that referenced
this issue
Jan 16, 2020
When Fluent Bit encounters with a partial parser definition, it
crashes badly with a segmentation fault.
$ ./bin/fluent-bit -R parser.conf -c tail.conf
...
[2020/01/15 16:11:21] [error] [parser] no parser 'format' found for 'simple' in file 'conf/timestamp.parser'
[engine] caught signal (SIGSEGV)
#0 0x558bc4a0a226 in flb_parser_decoder_list_destroy() at src/flb_parser_decoder.c:700
#1 0x558bc4a05d75 in flb_parser_conf_file() at src/flb_parser.c:566
#2 0x558bc49f4bdd in flb_config_set_property() at src/flb_config.c:406
#3 0x558bc49e24ae in flb_service_conf() at src/fluent-bit.c:446
#4 0x558bc49e2f90 in main() at src/fluent-bit.c:807
#5 0x7fa1cb7f109a in ???() at ???:0
#6 0x558bc49e13a9 in ???() at ???:0
#7 0xffffffffffffffff in ???() at ???:0
Aborted
This is just because `decoders` is not being initialized properly,
and that confuses Fluent Bit to deallocate a random memmory block
on the cleanup path. Fix it.
Signed-off-by: Fujimoto Seiji <fujimoto@clear-code.com>
edsiper
pushed a commit
that referenced
this issue
Jan 17, 2020
When Fluent Bit encounters with a partial parser definition, it
crashes badly with a segmentation fault.
$ ./bin/fluent-bit -R parser.conf -c tail.conf
...
[2020/01/15 16:11:21] [error] [parser] no parser 'format' found for 'simple' in file 'conf/timestamp.parser'
[engine] caught signal (SIGSEGV)
#0 0x558bc4a0a226 in flb_parser_decoder_list_destroy() at src/flb_parser_decoder.c:700
#1 0x558bc4a05d75 in flb_parser_conf_file() at src/flb_parser.c:566
#2 0x558bc49f4bdd in flb_config_set_property() at src/flb_config.c:406
#3 0x558bc49e24ae in flb_service_conf() at src/fluent-bit.c:446
#4 0x558bc49e2f90 in main() at src/fluent-bit.c:807
#5 0x7fa1cb7f109a in ???() at ???:0
#6 0x558bc49e13a9 in ???() at ???:0
#7 0xffffffffffffffff in ???() at ???:0
Aborted
This is just because `decoders` is not being initialized properly,
and that confuses Fluent Bit to deallocate a random memmory block
on the cleanup path. Fix it.
Signed-off-by: Fujimoto Seiji <fujimoto@clear-code.com>
edsiper
pushed a commit
that referenced
this issue
Jan 23, 2020
When Fluent Bit encounters with a partial parser definition, it
crashes badly with a segmentation fault.
$ ./bin/fluent-bit -R parser.conf -c tail.conf
...
[2020/01/15 16:11:21] [error] [parser] no parser 'format' found for 'simple' in file 'conf/timestamp.parser'
[engine] caught signal (SIGSEGV)
#0 0x558bc4a0a226 in flb_parser_decoder_list_destroy() at src/flb_parser_decoder.c:700
#1 0x558bc4a05d75 in flb_parser_conf_file() at src/flb_parser.c:566
#2 0x558bc49f4bdd in flb_config_set_property() at src/flb_config.c:406
#3 0x558bc49e24ae in flb_service_conf() at src/fluent-bit.c:446
#4 0x558bc49e2f90 in main() at src/fluent-bit.c:807
#5 0x7fa1cb7f109a in ???() at ???:0
#6 0x558bc49e13a9 in ???() at ???:0
#7 0xffffffffffffffff in ???() at ???:0
Aborted
This is just because `decoders` is not being initialized properly,
and that confuses Fluent Bit to deallocate a random memmory block
on the cleanup path. Fix it.
Signed-off-by: Fujimoto Seiji <fujimoto@clear-code.com>
4 tasks
This was referenced Jul 25, 2021
This was referenced Jun 21, 2022
cosmo0920
added a commit
that referenced
this issue
Oct 5, 2022
…es strictly
Without this check, the following weird error is occurred
intermittently:
```log
[0] dummy.0: [1664938706.407551000, {"message"=>"dummy"}]
[2022/10/05 11:58:27] [ info] [test] flush record
flb-rt-core_chunk_trace(32205,0x16fe87000) malloc: *** error for object 0x600002600074: pointer being realloc'd was not allocated
flb-rt-core_chunk_trace(32205,0x16fe87000) malloc: *** set a breakpoint in malloc_error_break to debug
```
The main reason is, num_records index is broken in some cases:
```
flb-rt-core_chunk_trace(32205,0x16fe87000) malloc: *** error for object 0x600002600074: pointer being realloc'd was not allocated
flb-rt-core_chunk_trace(32205,0x16fe87000) malloc: *** set a breakpoint in malloc_error_break to debug
[2022/10/05 11:58:27] [ info] [input] pausing dummy.0
Process 32205 stopped
* thread #2, name = 'flb-pipeline', stop reason = breakpoint 1.1
frame #0: 0x00000001b34a3120 libsystem_malloc.dylib`malloc_error_break
libsystem_malloc.dylib`malloc_error_break:
-> 0x1b34a3120 <+0>: pacibsp
0x1b34a3124 <+4>: stp x29, x30, [sp, #-0x10]!
0x1b34a3128 <+8>: mov x29, sp
0x1b34a312c <+12>: nop
Target 0: (flb-rt-core_chunk_trace) stopped.
(lldb) bt
* thread #2, name = 'flb-pipeline', stop reason = breakpoint 1.1
* frame #0: 0x00000001b34a3120 libsystem_malloc.dylib`malloc_error_break
frame #1: 0x00000001b3494844 libsystem_malloc.dylib`malloc_vreport + 428
frame #2: 0x00000001b3497f34 libsystem_malloc.dylib`malloc_report + 64
frame #3: 0x00000001b3488210 libsystem_malloc.dylib`realloc + 328
frame #4: 0x0000000100006154 flb-rt-core_chunk_trace`flb_realloc(ptr=0x0000600002600074, size=18446744064764412176) at flb_mem.h:94:12
frame #5: 0x0000000100005fc8 flb-rt-core_chunk_trace`callback_add_record(data=0x0000600003014000, size=135, cb_data=0x0000600000004010) at core_chunk_trace.c:51:28
frame #6: 0x00000001001268b0 flb-rt-core_chunk_trace`out_lib_flush(event_chunk=0x0000600000c14000, out_flush=0x0000600001714000, i_ins=0x0000000100b09ab0, out_context=0x0000600000204a80, config=0x000000010181d200) at out_lib.c:197:9
frame #7: 0x0000000100029d70 flb-rt-core_chunk_trace`output_pre_cb_flush at flb_output.h:517:5
frame #8: 0x000000010044fa64 flb-rt-core_chunk_trace`co_switch(handle=0x000000010044fa64) at aarch64.c:133:4
(lldb) frane select 5
error: 'frane' is not a valid command.
(lldb) frame select 5
frame #5: 0x0000000100005fc8 flb-rt-core_chunk_trace`callback_add_record(data=0x0000600003014000, size=135, cb_data=0x0000600000004010) at core_chunk_trace.c:51:28
48 flb_calloc(1, sizeof(struct callback_record));
49 } else {
50 ctx->records = (struct callback_record *)
-> 51 flb_realloc(ctx->records,
52 (ctx->num_records+1)*sizeof(struct callback_record));
53 }
54 if (ctx->records == NULL) {
(lldb) po ctx->records
0x0000600002600074
(lldb) po ctx->records
0x0000600002600074
(lldb) po ctx->num_records
-559071216
```
Signed-off-by: Hiroshi Hatake <hatake@calyptia.com>
zecke
added a commit
to zecke/fluent-bit
that referenced
this issue
May 25, 2024
The tls variable for out_flush_params is not initialized as the
flb_start function is not called during the dry run. Call flb_init
directly and then shutdown the engine.
configuration test is successful
=================================================================
==63633==ERROR: AddressSanitizer: attempting free on address which was not malloc()-ed: 0x0001f71b3ac0 in thread T0
#0 0x103c9f260 in wrap_free+0x98 (libclang_rt.asan_osx_dynamic.dylib:arm64e+0x53260)
fluent#1 0x100179d9c in flb_free flb_mem.h:127
fluent#2 0x10017f4a0 in flb_output_exit flb_output.c:481
fluent#3 0x1001cb038 in flb_engine_shutdown flb_engine.c:1119
fluent#4 0x10010d45c in flb_destroy flb_lib.c:240
fluent#5 0x100008c40 in flb_main fluent-bit.c:1348
fluent#6 0x10000c644 in main fluent-bit.c:1456
fluent#7 0x18f11e0dc (<unknown module>)
frame fluent#6: 0x000000010017f4a4 fluent-bit`flb_output_exit(config=0x0000000102b00200) at flb_output.c:481:9
478
479 params = FLB_TLS_GET(out_flush_params);
480 if (params) {
-> 481 flb_free(params);
482 }
483 }
Signed-off-by: Holger Hans Peter Freyther <holger@freyther.de>
edsiper
pushed a commit
that referenced
this issue
May 26, 2024
The tls variable for out_flush_params is not initialized as the
flb_start function is not called during the dry run. Call flb_init
directly and then shutdown the engine.
configuration test is successful
=================================================================
==63633==ERROR: AddressSanitizer: attempting free on address which was not malloc()-ed: 0x0001f71b3ac0 in thread T0
#0 0x103c9f260 in wrap_free+0x98 (libclang_rt.asan_osx_dynamic.dylib:arm64e+0x53260)
#1 0x100179d9c in flb_free flb_mem.h:127
#2 0x10017f4a0 in flb_output_exit flb_output.c:481
#3 0x1001cb038 in flb_engine_shutdown flb_engine.c:1119
#4 0x10010d45c in flb_destroy flb_lib.c:240
#5 0x100008c40 in flb_main fluent-bit.c:1348
#6 0x10000c644 in main fluent-bit.c:1456
#7 0x18f11e0dc (<unknown module>)
frame #6: 0x000000010017f4a4 fluent-bit`flb_output_exit(config=0x0000000102b00200) at flb_output.c:481:9
478
479 params = FLB_TLS_GET(out_flush_params);
480 if (params) {
-> 481 flb_free(params);
482 }
483 }
Signed-off-by: Holger Hans Peter Freyther <holger@freyther.de>
markuman
pushed a commit
to markuman/fluent-bit
that referenced
this issue
May 29, 2024
The tls variable for out_flush_params is not initialized as the
flb_start function is not called during the dry run. Call flb_init
directly and then shutdown the engine.
configuration test is successful
=================================================================
==63633==ERROR: AddressSanitizer: attempting free on address which was not malloc()-ed: 0x0001f71b3ac0 in thread T0
#0 0x103c9f260 in wrap_free+0x98 (libclang_rt.asan_osx_dynamic.dylib:arm64e+0x53260)
fluent#1 0x100179d9c in flb_free flb_mem.h:127
fluent#2 0x10017f4a0 in flb_output_exit flb_output.c:481
fluent#3 0x1001cb038 in flb_engine_shutdown flb_engine.c:1119
fluent#4 0x10010d45c in flb_destroy flb_lib.c:240
fluent#5 0x100008c40 in flb_main fluent-bit.c:1348
fluent#6 0x10000c644 in main fluent-bit.c:1456
fluent#7 0x18f11e0dc (<unknown module>)
frame fluent#6: 0x000000010017f4a4 fluent-bit`flb_output_exit(config=0x0000000102b00200) at flb_output.c:481:9
478
479 params = FLB_TLS_GET(out_flush_params);
480 if (params) {
-> 481 flb_free(params);
482 }
483 }
Signed-off-by: Holger Hans Peter Freyther <holger@freyther.de>
Signed-off-by: Markus Bergholz <git@osuv.de>
nathalapooja
referenced
this issue
in amazon-contributing/upstream-to-fluent-bit
Oct 16, 2024
…pace and node (#6) * Add PutLogEvents entity scraping logic for namespace and node * Fix entity formatting * Remove unnecessary non-NULL scenario checks for entity attributes
# for free
to join this conversation on GitHub.
Already have an account?
# to comment
$ bin/fluent-bit -i mem -o stdout
[2015/06/02 18:41:02] [ info] starting engine
[0] {"time"=>1433292063, "total"=>8081592, "free"=>1595860}
[1] {"time"=>1433292064, "total"=>8081592, "free"=>1596028}
[2] {"time"=>1433292065, "total"=>8081592, "free"=>1595900}
[3] {"time"=>1433292066, "total"=>8081592, "free"=>1600272}
[4] 0
[5] 0
[6] 0
[7] 0
[8] 0
[9] 0
[10] 0
[11] 0
[12] 33
[13] 1
[14] 0
[15] 0
[16] 0
[17] 0
[18] 0
[19] 0
[20] 104
[21] -19
[22] 86
[23] "d^?^@^@hVd^?^@^@^@"
[24] 0
[25] 0
[26] 0
[27] 0
[28] 0
The text was updated successfully, but these errors were encountered: