Symptoms:
NSX Edge 6.3.5 experiences crash with following messages
vserrdd[1729]: [daemon.warning] Memory overloaded: 96.29 used, threshhold: 90.00
Crash dump
[DR Back End]: [kern.warning] log_cleanup.sh invoked oom-killer: gfp_mask=0x26000c0, order=2, oom_score_adj=0
[DR Back End]: [kern.info] log_cleanup.sh cpuset=/ mems_allowed=0
[kern.warning] CPU: 1 PID: 29614 Comm: log_cleanup.sh Tainted: G O 4.4.57 #1
[DR Back End]: [kern.warning] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, B
IOS 6.00 09/21/2015
[DR Back End]: [kern.warning] 0000000000000000 ffff880013923b58 ffffffff81310844 0000000000000007
[DR Back End]: [kern.warning] 0000000000000000 ffff88003afd8000 ffff880013923bb8 ffffffff8117c8e6
[DR Back End]: [kern.warning] ffff88002888b180 0000000000000206 000000000005e154 ffff880013923b90
[DR Back End]: [kern.warning] Call Trace:
[DR Back End]: [kern.warning] [<ffffffff81310844>] dump_stack+0x58/0x84
[DR Back End]: [kern.warning] [<ffffffff8117c8e6>] dump_header.isra.10+0x54/0x19e
[DR Back End]: [kern.warning] [<ffffffff8161089e>] ? _raw_spin_unlock_irqrestore+0xe/0x10
[DR Back End]: [kern.warning] [<ffffffff81139a83>] oom_kill_process+0x203/0x420
[DR Back End]: [kern.warning] [<ffffffff81139f9f>] out_of_memory+0x29f/0x2e0
[DR Back End]: [kern.warning] [<ffffffff8113e91e>] __alloc_pages_nodemask+0x8de/0x930
[DR Back End]: [kern.warning] [<ffffffff81181ecc>] ? get_empty_filp+0x5c/0x1c0
[DR Back End]: [kern.warning] [<ffffffff8113ebbb>] alloc_kmem_pages_node+0x1b/0x20
[DR Back End]: [kern.warning] [<ffffffff81056232>] copy_process+0x162/0x1ae0
[DR Back End]: [kern.warning] [<ffffffff8117a2b0>] ? kmem_cache_alloc+0x140/0x150
[DR Back End]: [kern.warning] [<ffffffff81057d21>] _do_fork+0x71/0x350
[DR Back End]: [kern.warning] [<ffffffff81067fb1>] ? sigprocmask+0x51/0x80
VMware NSX for vSphere 6.4.x
VMware NSX for vSphere 6.3.x
Memory issue within nagios process on NSX edge causes appliance to crash
This issue affects both 6.3.x and 6.4.x. The solution is expected to be part of 6.4.2 & possibly, 6.3.7.
Workaround:
Work around:
Change L4 LB to L7 LB by setting the "Acceleration Status" to disabled at the Global Configuration page under the Load Balancer tab.