[MESOS-8594] Mesos master stack overflow in libprocess socket send loop.

• Type: Bug
• Status: Resolved
• Priority: Blocker
• Resolution: Fixed
• Affects Version/s: 1.3.2, 1.4.1, 1.5.0, 1.6.0
• Fix Version/s: 1.5.1, 1.6.0
• Component/s: master
• Labels: reliability
• Target Version/s: 1.5.1, 1.6.0
• Sprint: Mesosphere Sprint 79
• Story Points: 1
• Assignee: Benjamin Mahler
• Reporter: A. Dukhovniy
• Created: 19/Feb/18 13:52
• Updated: 04/Jul/18 00:08
• Resolved: 01/May/18 00:45

Description

Mesos master crashes under load. Attached are some infos from the `lldb`:

Process 41933 resuming
Process 41933 stopped
* thread #10, stop reason = EXC_BAD_ACCESS (code=2, address=0x7000089ecff8)
frame #0: 0x000000010c30ddb6 libmesos-1.6.0.dylib`::_Some() at some.hpp:35
32 template <typename T>
33 struct _Some
34 {
-> 35 _Some(T _t) : t(std::move(_t)) {}
36
37 T t;
38 };
Target 0: (mesos-master) stopped.
(lldb)

To quote abudnik

it’s the stack overflow bug in libprocess due to the way `internal::send()` and `internal::_send()` are implemented in `process.cpp`

Issue Links

• duplicates: MESOS-8834 Indirect recursion between `send` and `_send` in libprocess may cause stack overflow.
• is related to: MESOS-9024 Mesos master segfaults with stack overflow under load.
• relates to: MESOS-8852 process::loop does not guarantee stack overflow prevention.