Project

General

Profile

Actions
Copy link

Bug #3118

closed

asan leaks with 5.0.0-dev (9e126b210 2019-08-07)

Added by Peter Manev about 5 years ago. Updated almost 5 years ago.

Status:
Closed
Priority:
Normal
Assignee:
-
Target version:
-
Affected Versions:
Effort:
Difficulty:
Label:

Description

Compiled like:

git clone https://github.com/OISF/suricata.git && cd suricata && \
git clone https://github.com/OISF/libhtp.git -b 0.5.x && ./autogen.sh &&  \
./configure --prefix=/opt/suritest-asan/ --sysconfdir=/opt/suritest-asan/etc \
--localstatedir=/opt/suritest-asan/var  \
CFLAGS="-ggdb3 -Werror -Wchar-subscripts -fno-strict-aliasing -fstack-protector-all -fsanitize=address -fno-omit-frame-pointer -Wno-unused-parameter -Wno-unused-function"  ac_cv_func_malloc_0_nonnull=yes ac_cv_func_realloc_0_nonnull=yes  \
--enable-geoip --enable-rust-strict  --enable-luajit  &&  \
make clean &&   make -j && make install && \
make install && make install-full &&  ldconfig

run like:

time LSAN_OPTIONS=suppressions=suricata/qa/lsan.suppress ASAN_SYMBOLIZER_PATH=/usr/lib/llvm-8/bin/llvm-symbolizer \
> /opt/suritest-asan/bin/suricata -k none -r qa.pcap --set "flow.memcap=8gb" \
> --set "stream.memcap=6gb" --set stream.reassembly.memcap=10gb --set max-pending-packets=20000 \
> -l log/ --runmode=autofp -S "rules/*.rules" 
[2214] 18/8/2019 -- 10:33:12 - (suricata.c:1071) <Notice> (LogVersion) -- This is Suricata version 5.0.0-dev (3a912446a 2019-07-22) running in USER mode
[2214] 18/8/2019 -- 10:35:43 - (tm-threads.c:2145) <Notice> (TmThreadWaitOnThreadInit) -- all 9 packet processing threads, 4 management threads initialized, engine started.
[2214] 18/8/2019 -- 11:15:56 - (suricata.c:2851) <Notice> (SuricataMainLoop) -- Signal Received.  Stopping engine.
[2270] 18/8/2019 -- 11:49:54 - (source-pcap-file.c:378) <Notice> (ReceivePcapFileThreadExitStats) -- Pcap-file module read 1 files, 275652306 packets, 152382822719 bytes

=================================================================
==2214==ERROR: LeakSanitizer: detected memory leaks

Direct leak of 64 byte(s) in 2 object(s) allocated from:
    #0 0x7f90ce0fdd38 in __interceptor_calloc (/usr/lib/x86_64-linux-gnu/libasan.so.4+0xded38)
    #1 0x5613529b108c in FTPCalloc /home/pevma/Work/tmp/suricata/src/app-layer-ftp.c:213
    #2 0x5613529b43cb in FTPParseRequest /home/pevma/Work/tmp/suricata/src/app-layer-ftp.c:644
    #3 0x5613529e2083 in AppLayerParserParse /home/pevma/Work/tmp/suricata/src/app-layer-parser.c:1210
    #4 0x56135290a216 in AppLayerHandleTCPData /home/pevma/Work/tmp/suricata/src/app-layer.c:660
    #5 0x561352e727e3 in ReassembleUpdateAppLayer /home/pevma/Work/tmp/suricata/src/stream-tcp-reassemble.c:1066
    #6 0x561352e72c4b in StreamTcpReassembleAppLayer /home/pevma/Work/tmp/suricata/src/stream-tcp-reassemble.c:1123
    #7 0x561352e7634d in StreamTcpReassembleHandleSegmentUpdateACK /home/pevma/Work/tmp/suricata/src/stream-tcp-reassemble.c:1689
    #8 0x561352e76608 in StreamTcpReassembleHandleSegment /home/pevma/Work/tmp/suricata/src/stream-tcp-reassemble.c:1732
    #9 0x561352e320f6 in HandleEstablishedPacketToClient /home/pevma/Work/tmp/suricata/src/stream-tcp.c:2375
    #10 0x561352e34b1e in StreamTcpPacketStateEstablished /home/pevma/Work/tmp/suricata/src/stream-tcp.c:2612
    #11 0x561352e50beb in StreamTcpStateDispatch /home/pevma/Work/tmp/suricata/src/stream-tcp.c:4617
    #12 0x561352e5224d in StreamTcpPacket /home/pevma/Work/tmp/suricata/src/stream-tcp.c:4798
    #13 0x561352e53e4f in StreamTcp /home/pevma/Work/tmp/suricata/src/stream-tcp.c:5134
    #14 0x561352ca5443 in FlowWorker /home/pevma/Work/tmp/suricata/src/flow-worker.c:231
    #15 0x561352ea0b8e in TmThreadsSlotVarRun /home/pevma/Work/tmp/suricata/src/tm-threads.c:128
    #16 0x561352ea2ee5 in TmThreadsSlotVar /home/pevma/Work/tmp/suricata/src/tm-threads.c:585
    #17 0x7f90cc4f26da in start_thread (/lib/x86_64-linux-gnu/libpthread.so.0+0x76da)

Indirect leak of 18 byte(s) in 2 object(s) allocated from:
    #0 0x7f90ce0fdd38 in __interceptor_calloc (/usr/lib/x86_64-linux-gnu/libasan.so.4+0xded38)
    #1 0x5613529b108c in FTPCalloc /home/pevma/Work/tmp/suricata/src/app-layer-ftp.c:213
    #2 0x5613529b4473 in FTPParseRequest /home/pevma/Work/tmp/suricata/src/app-layer-ftp.c:649
    #3 0x5613529e2083 in AppLayerParserParse /home/pevma/Work/tmp/suricata/src/app-layer-parser.c:1210
    #4 0x56135290a216 in AppLayerHandleTCPData /home/pevma/Work/tmp/suricata/src/app-layer.c:660
    #5 0x561352e727e3 in ReassembleUpdateAppLayer /home/pevma/Work/tmp/suricata/src/stream-tcp-reassemble.c:1066
    #6 0x561352e72c4b in StreamTcpReassembleAppLayer /home/pevma/Work/tmp/suricata/src/stream-tcp-reassemble.c:1123
    #7 0x561352e7634d in StreamTcpReassembleHandleSegmentUpdateACK /home/pevma/Work/tmp/suricata/src/stream-tcp-reassemble.c:1689
    #8 0x561352e76608 in StreamTcpReassembleHandleSegment /home/pevma/Work/tmp/suricata/src/stream-tcp-reassemble.c:1732
    #9 0x561352e320f6 in HandleEstablishedPacketToClient /home/pevma/Work/tmp/suricata/src/stream-tcp.c:2375
    #10 0x561352e34b1e in StreamTcpPacketStateEstablished /home/pevma/Work/tmp/suricata/src/stream-tcp.c:2612
    #11 0x561352e50beb in StreamTcpStateDispatch /home/pevma/Work/tmp/suricata/src/stream-tcp.c:4617
    #12 0x561352e5224d in StreamTcpPacket /home/pevma/Work/tmp/suricata/src/stream-tcp.c:4798
    #13 0x561352e53e4f in StreamTcp /home/pevma/Work/tmp/suricata/src/stream-tcp.c:5134
    #14 0x561352ca5443 in FlowWorker /home/pevma/Work/tmp/suricata/src/flow-worker.c:231
    #15 0x561352ea0b8e in TmThreadsSlotVarRun /home/pevma/Work/tmp/suricata/src/tm-threads.c:128
    #16 0x561352ea2ee5 in TmThreadsSlotVar /home/pevma/Work/tmp/suricata/src/tm-threads.c:585
    #17 0x7f90cc4f26da in start_thread (/lib/x86_64-linux-gnu/libpthread.so.0+0x76da)

SUMMARY: AddressSanitizer: 82 byte(s) leaked in 4 allocation(s).

real    77m4.883s
user    340m49.133s
sys     9m48.217s

Related issues 1 (0 open — 1 closed)

Related to Suricata - Bug #2458: memleak: gitmaster - 4.1.0-dev (rev c60decd)RejectedActions
Actions
Copy link
 #1

Updated by Victor Julien about 5 years ago

  • Status changed from New to Assigned
  • Assignee set to Jeff Lucovsky
  • Target version set to 70

Peter can you make sure Jeff has access to the pcap?

Actions
Copy link
 #2

Updated by Peter Manev about 5 years ago

I've been trying to carve a pcap out of the original pcap (140GB) that I was able to generate this issue with. So far have been unsuccessful , for starters I tried to carve out just "ftp" traffic, the resulting pcap was much smaller but could not reproduce the issue. Will keep trying and update the issue here.

Actions
Copy link
 #3

Updated by Jeff Lucovsky about 5 years ago

If this issue is reproducible, then I suggest we try patching the following function -- adding an assert

int StorageSetById(Storage *storage, const StorageEnum type, const int id, void *ptr)
{
#ifdef DEBUG
    BUG_ON(!storage_registraton_closed);
#endif
    SCLogDebug("storage %p id %d", storage, id);
    if (storage == NULL)
        return -1;
    assert(storage[id] == NULL); // If not null, memory leak
    storage[id] = ptr;
    return 0;
}
Actions
Copy link
 #4

Updated by Jeff Lucovsky about 5 years ago

  • Related to Bug #2458: memleak: gitmaster - 4.1.0-dev (rev c60decd) added
Actions
Copy link
 #5

Updated by Peter Manev about 5 years ago

I can rerun it and will report back.

Actions
Copy link
 #6

Updated by Jeff Lucovsky about 5 years ago

Great ... this is to test a hunch. This was a place where it looked like there could be a memory leak related to the expectation handling.

Actions
Copy link
 #7

Updated by Peter Manev almost 5 years ago

I dont see this any more on any runs that i tried.

Actions
Copy link
 #8

Updated by Victor Julien almost 5 years ago

  • Status changed from Assigned to Closed
  • Assignee deleted (Jeff Lucovsky)
  • Target version deleted (70)
Actions
Copy link

Also available in: Atom PDF