Andrew Wong has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/12170 )

Change subject: client: avoid KLOG_EVERY_N_SECS with ScopedCleanup
......................................................................


Patch Set 1:

(2 comments)

http://gerrit.cloudera.org:8080/#/c/12170/1//COMMIT_MSG
Commit Message:

http://gerrit.cloudera.org:8080/#/c/12170/1//COMMIT_MSG@11
PS1, Line 11: User stack:
> this output format looks like from the stack watchdog, not UBSAN. UBSAN its
You're right, it's the user stack from the watchdog.

The UBSAN error was:

 SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior 
src/kudu/client/master_proxy_rpc.cc:188:13 in

And the log cuts off there.

Though looking at another, it SIGSEGVed with

*** Aborted at 1546650883 (unix time) try "date -d @1546650883" if you are 
using GNU date ***
PC: @     0x7f48d23ce156 __dynamic_cast
*** SIGSEGV (@0xfffffffffffffff0) received by PID 13824 (TID 0x7f48c8fce700) 
from PID 18446744073709551600; stack trace: ***
    @     0x7f48d9865330 (unknown) at ??:0
    @     0x7f48d23ce156 __dynamic_cast at ??:0
    @           0x549184 __ubsan::checkDynamicType() at 
/home/jenkins-slave/workspace/kudu-master/2/thirdparty/src/llvm-6.0.0.src/projects/compiler-rt/lib/ubsan/ubsan_type_hash_itanium.cc:234
 (discriminator 2)
    @           0x547083 HandleDynamicTypeCacheMiss() at 
/home/jenkins-slave/workspace/kudu-master/2/thirdparty/src/llvm-6.0.0.src/projects/compiler-rt/lib/ubsan/ubsan_handlers_cxx.cc:37
    @           0x5479a3 __ubsan_handle_dynamic_type_cache_miss_abort at 
/home/jenkins-slave/workspace/kudu-master/2/thirdparty/src/llvm-6.0.0.src/projects/compiler-rt/lib/ubsan/ubsan_handlers_cxx.cc:93
    @     0x7f48e3bf76b3 
_ZZN4kudu6client8internal20AsyncLeaderMasterRpcINS_6master23GetTableSchemaRequestPBENS3_24GetTableSchemaResponsePBEE27RetryOrReconnectIfNecessaryEPNS_6StatusEENKUlvE_clEv
 at ??:0
    @     0x7f48e3bcabfc 
_ZN4kudu13ScopedCleanupIZNS_6client8internal20AsyncLeaderMasterRpcINS_6master23GetTableSchemaRequestPBENS4_24GetTableSchemaResponsePBEE27RetryOrReconnectIfNecessaryEPNS_6StatusEEUlvE_ED2Ev
 at ??:0
    @     0x7f48e3bca4cc 
kudu::client::internal::AsyncLeaderMasterRpc<>::RetryOrReconnectIfNecessary() 
at ??:0
    @     0x7f48e3bc8d7a 
kudu::client::internal::AsyncLeaderMasterRpc<>::SendRpcCb() at ??:0
    @     0x7f48e3bf6af5 
_ZNSt5_BindIFSt7_Mem_fnIMN4kudu6client8internal20AsyncLeaderMasterRpcINS1_6master23GetTableSchemaRequestPBENS5_24GetTableSchemaResponsePBEEEFvRKNS1_6StatusEEEPS8_S9_EE6__callIvJEJLm0ELm1EEEET_OSt5tupleIJDpT0_EESt12_Index_tupleIJXspT1_EEE
 at ??:0
    @     0x7f48e3bf698f 
_ZNSt5_BindIFSt7_Mem_fnIMN4kudu6client8internal20AsyncLeaderMasterRpcINS1_6master23GetTableSchemaRequestPBENS5_24GetTableSchemaResponsePBEEEFvRKNS1_6StatusEEEPS8_S9_EEclIJEvEET0_DpOT_
 at ??:0
    @     0x7f48de83daac boost::function0<>::operator()() at ??:0
    @     0x7f48de838cf2 kudu::rpc::OutboundCall::CallCallback() at ??:0
    @     0x7f48de83a379 kudu::rpc::OutboundCall::SetTimedOut() at ??:0
    @     0x7f48de7e4658 kudu::rpc::Connection::HandleOutboundCallTimeout() at 
??:0
    @     0x7f48de7e43ad 
kudu::rpc::Connection::CallAwaitingResponse::HandleTimeout() at ??:0
    @     0x7f48d4050606 ev_invoke_pending at ??:0
    @     0x7f48de84e2ca kudu::rpc::ReactorThread::InvokePendingCb() at ??:0
    @     0x7f48d40514f8 ev_run at ??:0
    @     0x7f48de84e619 kudu::rpc::ReactorThread::RunThread() at ??:0
    @     0x7f48de865bd3 boost::_bi::bind_t<>::operator()() at ??:0
    @     0x7f48de83daac boost::function0<>::operator()() at ??:0
    @     0x7f48d62b229b kudu::Thread::SuperviseThread() at ??:0
    @     0x7f48d985d184 start_thread at ??:0
    @     0x7f48d1e8fffd clone at ??:0

which also hits HandleDynamicTypeCacheMiss. So seems like it may not be as 
simple as I described.


http://gerrit.cloudera.org:8080/#/c/12170/1//COMMIT_MSG@30
PS1, Line 30: From this, it seems that KLOG_EVERY_N_SECS doesn't pair well with
            : ScopedCleanup.
> But do we know why?
Looking into it more.



--
To view, visit http://gerrit.cloudera.org:8080/12170
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I8cefd9613018247a1a25d17adedc021e8be166f6
Gerrit-Change-Number: 12170
Gerrit-PatchSet: 1
Gerrit-Owner: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <a...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <aser...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Todd Lipcon <t...@apache.org>
Gerrit-Comment-Date: Mon, 07 Jan 2019 19:03:11 +0000
Gerrit-HasComments: Yes

Reply via email to