使用ray 2.2.0版本,运行出错!

【Ray使用环境】生产/测试/POC
测试DEMO
#include<ray/api.h>
int Plus(int x, int y) {return x + y;}

RAY_REMOTE(Plus);
int main()
{
ray::init();
auto object = ray::Plus(100);
auto put_get = *(ray::Get(object));
std::cout << "put_get = " << put_get << std::endl;
ray::Shutdowm();
return 0;
}

出现问题:
terminal 信息:
logging.cc:361: ***** SIGABRT received at time=1675389650 on cpu 120 ***

gcs_server.out 信息:
Reporting worker exit, exit_type = SYSTEM_ERROR, exit_detail = Worker unexpectedly exits with a connection error code 2. End of file. There are some potential root causes.(1)The process is killed by SIGKILL by OOM killer due to high memory usage. (2) ray stop --force is called.(3) The worker is crashed unexpectedly due to SIGSEGV or other unexpected errors…

free -h 信息:
total used free shared buff/cache available
Mem: 1.0Ti 4.9Gi 996Gi 2.1Gi 3.2Gi 993Gi
Swap: 4.0Gi 0B 4.0Gi

【Ray版本和类库】
ray 2.2.0版本

感谢使用Ray C++接口。

上面贴的代码问题挺多的,在我们的环境里编译有多个问题。我改了一下可以跑:

#include<ray/api.h>
int Plus(int x, int y) {return x + y;}

RAY_REMOTE(Plus);
int main()
{
ray::Init();
auto object = ray::Task(Plus).Remote(1, 2);
auto put_get = *(ray::Get(object));
std::cout << "put_get = " << put_get << std::endl;
ray::Shutdown();
return 0;
}

代码是手敲过来的,
运行环境是CPU: ARM
OS : openEuler
是否同环境有关?

代码更改后报了同样的错误。
运行环境:arm服务器。
ray安装方式:pip install -U ray[cpp]

看一下log,找找worker日志的报错。日志在 /tmp/ray/session_latest/logs

:grinning:麻烦帮我分析一下:

cpp-core-driver-xxx.log 文件信息:
SIGSEGV received at time=1675404727 on cpu 48 ***
logging.cc:361: std::string::assign()
logging.cc:361: absl::lts_20211102::AbslFailureSignalHandler()
logging.cc:361: (unknown)
logging.cc:361: ray::internal::BuildInvocationSpec1()
logging.cc:361: ray::internal::AbstractRayRuntime::Call()
logging.cc:361: ray::internal::TaskCaller<>::Remote<>()
logging.cc:361: main
logging.cc:361: (unknown)
logging.cc:361: __libc_start_main

dashboard.log 文件信息:
usage_stats_head.py:137 – Usage report request failed. HTTPSConnectionPool(host=‘usage-stats.ray.io’,port=443):Max retries exceeded with url:
(Caused by SSLError(1, ‘[SSL: WRONG_VERSION_NUMBER] wrong version number (_ssl.c:1129)’))

我怀疑还是代码写的有问题,然后ray底层的failure handling做的不太好导致报错不友好。你的代码再贴一下?