When deploying a Guaranteed QoS in k8s, Envoy occasionally restarts

[hushuai1559]

my k8s version is 1.28.9，I tried using Envoy versions from 1.17.4 to 1.27.1, but encountered the error "Caught Segmentation fault" in all cases.
The command I am running is envoy -c /ambassador/bootstrap-ads.json --base-id 0 --drain-time-s 600 -l error.
The detailed error is as follows:

[2024-12-10 02:10:43.598][180][critical][backtrace] [./source/server/backtrace.h:104] Caught Segmentation fault, suspect faulting address 0x0
2024-12-10T10:10:43.599081709+08:00 [2024-12-10 02:10:43.599][180][critical][backtrace] [./source/server/backtrace.h:91] Backtrace (use tools/stack_decode.py to get line numbers):
2024-12-10T10:10:43.599087228+08:00 [2024-12-10 02:10:43.599][180][critical][backtrace] [./source/server/backtrace.h:92] Envoy version: 6637fd1bab315774420f3c3d97488fedb7fc710f/1.27.2/Clean/RELEASE/BoringSSL
[2024-12-10 02:10:43.599][180][critical][backtrace] [./source/server/backtrace.h:98] #0: [0x7f27578f7320]
2024-12-10T10:10:43.599445995+08:00 [2024-12-10 02:10:43.599][180][critical][backtrace] [./source/server/backtrace.h:98] https://github.com/emissary-ingress/emissary/issues/1: [0x556b85fa1888]
2024-12-10T10:10:43.599467253+08:00 [2024-12-10 02:10:43.599][180][critical][backtrace] [./source/server/backtrace.h:98] https://github.com/emissary-ingress/emissary/issues/2: [0x556b86f81087]
2024-12-10T10:10:43.599491208+08:00 [2024-12-10 02:10:43.599][180][critical][backtrace] [./source/server/backtrace.h:98] https://github.com/emissary-ingress/emissary/issues/3: [0x556b879b487a]
[2024-12-10 02:10:43.599][180][critical][backtrace] [./source/server/backtrace.h:98] https://github.com/emissary-ingress/emissary/issues/4: [0x556b879b3491]
2024-12-10T10:10:43.599539289+08:00 [2024-12-10 02:10:43.599][180][critical][backtrace] [./source/server/backtrace.h:98] https://github.com/emissary-ingress/emissary/issues/5: [0x556b85fc4b2b]
2024-12-10T10:10:43.599554492+08:00 [2024-12-10 02:10:43.599][180][critical][backtrace] [./source/server/backtrace.h:98] https://github.com/emissary-ingress/emissary/issues/6: [0x556b879e9563]
[2024-12-10 02:10:43.599][180][critical][backtrace] [./source/server/backtrace.h:98] https://github.com/emissary-ingress/emissary/issues/7: [0x7f275794249a]

At the same time, an error occurred on the host machine, and the details are as follows:

[11578293.116244] envoy[199768]: segfault at 245a00000008 ip 0000559ee8b650f1 sp 00007ff54973d2c0 error 6 in envoy[559ee5743000+372c000]
[11578293.116253] Code: e8 a4 76 2c 00 cc cc cc cc 55 48 89 e5 41 57 41 56 41 55 41 54 53 48 83 ec 18 49 89 fe bf 40 00 00 00 e8 92 d5 2c 00 49 89 c5 <c7> 40 08 00 00 00 00 48 8d 05 71 36 5c 00 48 83 c0 10 49 89 45 00

I noticed that Envoy retrieves the number of cores from the host machine when starting, which may conflict with Kubernetes' Guaranteed QoS. Could this be the cause of the issue?

[hushuai1559]

This issue only occurs on the x86_64 architecture, it does not appear on ARM.

[phlax]

@hushuai1559 you might be better raising this with ambassador directly - also the minimum currently supported version of envoy is 1.29.x

[hushuai1559]

@hushuai1559 you might be better raising this with ambassador directly - also the minimum currently supported version of envoy is 1.29.x

This issue doesn't seem related to the ambassador. I have tried replacing the Envoy version with 1.32.3, but the kernel error still occurs.