Sysadmin Nginx 100% use Cpu

Two weeks ago one of the VMs completely went into a freeze state.

Connecting with kvm (ssh was timeout) this is what we saw.



Done at the provider, the day after he planned an intervention on that host hosting the vm, and updated kernel and librv.

Nevertheless, nginx did not resolve. There have no longer been complete freezes.
For now on the VM with brotli disabled, everything is OK.
before singing victory, we wait 48 hours.

I wonder if it was really brotli, because a simple compression can give so many problems?

 

I confirm, disabling brotli, everything works great.
50 hours of 100% continuous uptime.

I understand that google does not touch the brotli code for years, but is this problem possible?

 

Well ... there's definitely something wrong!
New dedicated server, new data center, new clean installation of centos 7 with centminmod with the following configuration:

Code:

PHP_PGO = 'y'
LETSENCRYPT_DETECT = 'y'
NGINX_IPV = 'y'
NGXDYNAMIC_BROTLI = 'y'
NGINX_LIBBROTLI = 'y' # https://github.com/eustas/ngx_brotli
NGINX_LIBBROTLISTATIC = 'y'
NGXDYNAMIC_NGXPAGESPEED = 'n'
NGINX_DEVTOOLSETGCC = 'y'
GENERAL_DEVTOOLSETGCC = 'y'
CLANG = 'n'
LIBRESSL_SWITCH = 'n'
NGINX_UPDATEMAINTENANCE = 'n'
PHP_UPDATEMAINTENANCE = 'n'
MARIADB_UPDATEMAINTENANCE = 'y'
CUSTOMSERVERNAME = 'y'
CUSTOMSERVERSTRING = 'nginx'
DUALCERTS = 'y' #dual cert RSA + ECDSA
NGINX_DYNAMICTLS = 'n'
CLOUDFLARE_ZLIB = 'y'
NGINX_HPACK = 'y'
NGINX_ZLIBCUSTOM = 'y'

-----------

Code:

cPUInfo;
processor: 0
vendor_id: GenuineIntel
cpu family: 6
model: 60
model name: Intel (R) Xeon (R) CPU E3-1271 v3 @ 3.60GHz
stepping: 3
microcode: 0x25
cpu MHz: 4000.122
cache size: 8192 KB
physical id: 0
siblings: 8
core id: 0
cpu cores: 4
apicid: 0
initial apicid: 0
fpu: yes
fpu_exception: yes
cpuid level: 13
wp: yes
flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx smx east tm2 SSSE3 sdbg FMA CX16 XTPR PDCM pcid sse4_1 sse4_2 X2apic movbe POPCNT tsc_deadline_timer aes xsave AVX f16c rdrand lahf_lm aBM epb ssbd IBRS ibpb stibp tpr_shadow vnmi FlexPriority ept vpid fsgsbase tsc_adjust BMI1 AVX2 SMEP bmi2 erms invpcid xsaveopt dtherm ida arat pln pts spec_ctrl intel_stibp flush_l1d
bogomips: 7200.45
clflush size: 64
cache_alignment: 64
address sizes: 39 bits physical, 48 bits virtual
power management:

Code:

uname -r
3.10.0-957.1.3.el7.x86_64

NGINX still goes 100% on 1 single core.
I have a single site on this server that does very little traffic.
there is a BUG, as big as a house, to be solved.

 

Hello, I tried even just with these two, it does not go anyway.
The sites are always different than basic wordpress, at this time 1 only vhost configured with wordpress + redis.

 

Sure!

I'm doing further tests.