[ https://issues.apache.org/jira/browse/ARROW-1380?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16587666#comment-16587666 ]
Wes McKinney commented on ARROW-1380: ------------------------------------- I'm still seeing "still reachable" warnings. Note that the thing that is being valgrinded is the Plasma store not the Python process. Sorry that the JIRA was not more clear {code} $ export PLASMA_VALGRIND=1 $ py.test pyarrow/tests/test_plasma.py -s -v =================================================== test session starts ==================================================== platform linux -- Python 3.6.5, pytest-3.6.0, py-1.5.3, pluggy-0.6.0 -- /home/wesm/miniconda/envs/arrow-dev/bin/python cachedir: .pytest_cache rootdir: /home/wesm/code/arrow/python, inifile: setup.cfg collected 24 items pyarrow/tests/test_plasma.py::TestPlasmaClient::test_connection_failure_raises_exception ==32111== Memcheck, a memory error detector ==32111== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al. ==32111== Using Valgrind-3.13.0 and LibVEX; rerun with -h for copyright info ==32111== Command: /home/wesm/code/arrow/python/pyarrow/plasma_store -s /tmp/test_plasma-im2a8xhc/plasma.sock -m 100000000 ==32111== Allowing the Plasma store to use up to 0.1GB of memory. Starting object store with directory /dev/shm and huge page support disabled Connection to IPC socket failed for pathname unknown-store-name, retrying 1 more times PASSED==32111== ==32111== HEAP SUMMARY: ==32111== in use at exit: 33 bytes in 1 blocks ==32111== total heap usage: 33 allocs, 32 frees, 129,757 bytes allocated ==32111== ==32111== 33 bytes in 1 blocks are still reachable in loss record 1 of 1 ==32111== at 0x4C3017F: operator new(unsigned long) (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==32111== by 0x5739BBE: allocate (new_allocator.h:111) ==32111== by 0x5739BBE: std::string::_Rep::_S_create(unsigned long, unsigned long, std::allocator<char> const&) (basic_string.tcc:1057) ==32111== by 0x5739D47: std::string::_M_mutate(unsigned long, unsigned long, unsigned long) (basic_string.tcc:929) ==32111== by 0x573A16B: std::string::_M_replace_safe(unsigned long, unsigned long, char const*, unsigned long) (basic_string.tcc:1134) ==32111== by 0x47E54D: main (store.cc:937) ==32111== ==32111== LEAK SUMMARY: ==32111== definitely lost: 0 bytes in 0 blocks ==32111== indirectly lost: 0 bytes in 0 blocks ==32111== possibly lost: 0 bytes in 0 blocks ==32111== still reachable: 33 bytes in 1 blocks ==32111== of which reachable via heuristic: ==32111== stdstring : 33 bytes in 1 blocks ==32111== suppressed: 0 bytes in 0 blocks ==32111== ==32111== For counts of detected and suppressed errors, rerun with: -v ==32111== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0) {code} > [C++] Fix "still reachable" valgrind warnings in Plasma Python unit tests > ------------------------------------------------------------------------- > > Key: ARROW-1380 > URL: https://issues.apache.org/jira/browse/ARROW-1380 > Project: Apache Arrow > Issue Type: Bug > Components: Plasma (C++) > Reporter: Wes McKinney > Priority: Major > Fix For: 0.11.0 > > Attachments: LastTest.log, valgrind.supp_ > > > I thought I fixed this, but they seem to have recurred: > https://travis-ci.org/apache/arrow/jobs/266421430#L5220 -- This message was sent by Atlassian JIRA (v7.6.3#76005)