Storaged crash when scanEdge called

[kikimo]

Please check the FAQ documentation before raising an issue

Storaged crash when scan edge with the following parameter:

	req := &storage.ScanEdgeRequest{
		SpaceID: globalSpaceID,
		Parts: map[int32]*storage.ScanCursor{
			globalPartitionID: {
				HasNext:    true,
				NextCursor: nil,
			},
		},
...

the crash stack:

[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
--Type <RET> for more, q to quit, c to continue without paging--c
Core was generated by `/root/src/nebula/build/bin/nebula-storaged --flagfile /root/nebula-chaos-cluste'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x0000000004f15958 in std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >::basic_string(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) ()
[Current thread is 1 (Thread 0x7f873f1ff700 (LWP 154))]
(gdb) bt
#0  0x0000000004f15958 in std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >::basic_string(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) ()
#1  0x0000000002d13411 in nebula::storage::ScanEdgeProcessor::runInSingleThread (this=0x7f8770e7d600, req=...) at /root/src/nebula/src/storage/query/ScanEdgeProcessor.cpp:139
#2  0x0000000002d125cd in nebula::storage::ScanEdgeProcessor::doProcess (this=0x7f8770e7d600, req=...) at /root/src/nebula/src/storage/query/ScanEdgeProcessor.cpp:53
#3  0x0000000002d11e81 in nebula::storage::ScanEdgeProcessor::<lambda()>::operator()(void) const (__closure=0x7f8770e211c0) at /root/src/nebula/src/storage/query/ScanEdgeProcessor.cpp:19
#4  0x0000000002d19073 in folly::detail::function::FunctionTraits<void()>::callBig<nebula::storage::ScanEdgeProcessor::process(const nebula::storage::cpp2::ScanEdgeRequest&)::<lambda()> >(folly::detail::function::Data &) (p=...)
    at /root/src/nebula/build/third-party/install/include/folly/Function.h:385
#5  0x000000000457b8a7 in virtual thunk to apache::thrift::concurrency::FunctionRunner::run() ()
#6  0x00000000046b0ca3 in apache::thrift::concurrency::ThreadManager::Impl::Worker::run() ()
#7  0x00000000046b48fd in apache::thrift::concurrency::PthreadThread::threadMain(void*) ()
#8  0x00007f877f03f609 in start_thread (arg=<optimized out>) at pthread_create.c:477
#9  0x00007f877ef66293 in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
(gdb) up
#1  0x0000000002d13411 in nebula::storage::ScanEdgeProcessor::runInSingleThread (this=0x7f8770e7d600, req=...) at /root/src/nebula/src/storage/query/ScanEdgeProcessor.cpp:139
139	    auto ret = plan.go(partId, cursor.get_has_next() ? *cursor.get_next_cursor() : "");
(gdb) down
#0  0x0000000004f15958 in std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >::basic_string(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) ()
(gdb) up
#1  0x0000000002d13411 in nebula::storage::ScanEdgeProcessor::runInSingleThread (this=0x7f8770e7d600, req=...) at /root/src/nebula/src/storage/query/ScanEdgeProcessor.cpp:139
139	    auto ret = plan.go(partId, cursor.get_has_next() ? *cursor.get_next_cursor() : "");
(gdb) p partId
$1 = 1
(gdb) p cursor.get_has_next()
You can't do that without a process to debug.
(gdb) p cursor
$2 = {static __fbthrift_cpp2_gen_json = false, static __fbthrift_cpp2_gen_nimble = false, static __fbthrift_cpp2_gen_has_thrift_uri = false, static __fbthrift_cpp2_is_union = false, has_next = true, next_cursor = {static npos = 18446744073709551615,
    _M_dataplus = {<std::allocator<char>> = {<__gnu_cxx::new_allocator<char>> = {<No data fields>}, <No data fields>}, _M_p = 0x7f873f1fc828 ""}, _M_string_length = 0, {_M_local_buf = '\000' <repeats 15 times>, _M_allocated_capacity = 0}}, __isset = {has_next = true,
    next_cursor = false}}
(gdb)

Your Environments (required)

• OS: uname -a
• Compiler: g++ --version or clang++ --version
• CPU: lscpu
• Commit id (e.g. a3ffc7d8)

How To Reproduce(required)

Steps to reproduce the behavior:

1. Step 1
2. Step 2
3. Step 3

Expected behavior

Additional context

[Shylock-Hg]

When has_next set to true, the next_cursor can't be null.

[Shylock-Hg]

A fix in #3427