xiaohei520321 commented on issue #2012:
URL:
https://github.com/apache/incubator-brpc/issues/2012#issuecomment-1327068423
还有connection_count也是一样,自定义LatencyRecord会导出,server默认的LatencyRecord不导出
--
This is an automated message from the Apache Git Service.
To respond to the message, please
wwbmmm closed issue #1615: urllib.urlopen().read() 访问http service卡住
URL: https://github.com/apache/incubator-brpc/issues/1615
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
wwbmmm closed issue #1629: V99 1207 18:47:09.863345 1326
/home/xiaoju/src/brpc/src/brpc/policy/thrift_protocol.cpp:379]
version=795373312 doesn't match THRIFT_VERSION=2147549184
URL: https://github.com/apache/incubator-brpc/issues/1629
--
This is an automated message from the Apache Git
wwbmmm closed issue #1647: brpc不支持Content-Type=application/octet-stream的请求
URL: https://github.com/apache/incubator-brpc/issues/1647
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific
yanglimingcn commented on issue #2013:
URL:
https://github.com/apache/incubator-brpc/issues/2013#issuecomment-1326971733
那我这边提一个PR吧,到时候你来看看。
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the
wwbmmm commented on issue #2000:
URL:
https://github.com/apache/incubator-brpc/issues/2000#issuecomment-1326939893
独立的io线程性能不一定会好,因为存在io线程和process线程之间同步的开销
你这里减小pthread worker的数量(最小是4),应该会好一些
--
This is an automated message from the Apache Git Service.
To respond to the message,
wwbmmm merged PR #2015:
URL: https://github.com/apache/incubator-brpc/pull/2015
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
serverglen merged PR #2014:
URL: https://github.com/apache/incubator-brpc/pull/2014
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
serverglen commented on PR #2014:
URL: https://github.com/apache/incubator-brpc/pull/2014#issuecomment-1326927755
LGTM
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To
guodongxiaren commented on PR #2016:
URL: https://github.com/apache/incubator-brpc/pull/2016#issuecomment-1326688916
会有失败的case
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific
guodongxiaren opened a new pull request, #2016:
URL: https://github.com/apache/incubator-brpc/pull/2016
### What problem does this PR solve?
Support Unittest for Linux workflow
Issue Number:
Problem Summary:
### What is changed and the side effects?
Changed:
guodongxiaren opened a new pull request, #2015:
URL: https://github.com/apache/incubator-brpc/pull/2015
### What problem does this PR solve?
Support Clang compile for Linux workflow
Issue Number:
Problem Summary:
### What is changed and the side effects?
Changed:
ketor opened a new pull request, #2014:
URL: https://github.com/apache/incubator-brpc/pull/2014
### What problem does this PR solve?
Issue Number:
Problem Summary:
There is issue of markdown format in doc/cn/getting_started.md
[docker][#docker] =>
Tuvie commented on PR #2005:
URL: https://github.com/apache/incubator-brpc/pull/2005#issuecomment-1326354421
@wwbmmm is this PR qualified to be merged? It seems that the Travis often
fails due to some irrelevant reason.
--
This is an automated message from the Apache Git Service.
To
zhouyanghere commented on issue #2000:
URL:
https://github.com/apache/incubator-brpc/issues/2000#issuecomment-1326336262
>
那感觉是不是还是像其他网络框架那样,io线程和socket是绑定的,只是rpc
process那里用bthread去做负载均衡要好一点呢?这样也可以避免长尾请求的阻塞效应。
--
This is an automated message from the Apache Git Service.
To
wwbmmm commented on issue #2000:
URL:
https://github.com/apache/incubator-brpc/issues/2000#issuecomment-1326332240
> write和read是在ProcessEvent内部执行的,还在PthreadA上执行。
这个请求是在pthread A 执行,下一个请求就到pthread B(接收了epoll
bthread的那个pthread)执行了呀。每个请求都换pthread,这样效率就降低了
--
This is an automated
zhouyanghere commented on issue #2000:
URL:
https://github.com/apache/incubator-brpc/issues/2000#issuecomment-1326254330
还有一个原因可能是因为同1个socket的write调用一直在不同的线程里切换导致开销变高?因为看其他的网络框架,一般socket会绑到一个固定io线程处理。
--
This is an automated message from the Apache Git Service.
To respond to the
fusying-hwang commented on issue #1918:
URL:
https://github.com/apache/incubator-brpc/issues/1918#issuecomment-1326140568
it's already resolved close now.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL
fusying-hwang closed issue #1918: Add example for build brpc as a third party
dependency with bazel?
URL: https://github.com/apache/incubator-brpc/issues/1918
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above
fusying-hwang commented on issue #1918:
URL:
https://github.com/apache/incubator-brpc/issues/1918#issuecomment-1326138637
alright, it turns out to be my problem. They do support building as an
external lib from source. My humble bazel knowledge limits my ability to
appreciate this
zhouyanghere commented on issue #2000:
URL:
https://github.com/apache/incubator-brpc/issues/2000#issuecomment-1326116094
>
它处理过程是这样的。Pthread
A上运行bthreadA(epoll_wait),然后处理epoll事件,即ProcessEvent过程,其为bthread
B,PthreadA会去执行bthread B,而将bthread
wwbmmm commented on issue #2000:
URL:
https://github.com/apache/incubator-brpc/issues/2000#issuecomment-1325979641
>
不不,我的重点不是当前bthread挂起并signal其它的worker来继续执行这个bthread带来的性能损耗,我的重点是经过bthread的ProcessEvent并没有切换pthread,为何系统调用函数write和read的开销高于直接调用ProcessEvent时的系统调用函数write和read的开销。
zhouyanghere commented on issue #2000:
URL:
https://github.com/apache/incubator-brpc/issues/2000#issuecomment-1325952578
不不,我的重点不是当前bthread挂起并signal其它的worker来继续执行这个bthread带来的性能损耗,我的重点是经过bthread的ProcessEvent并没有切换pthread,为何系统调用函数write和read的开销高于直接调用ProcessEvent时的系统调用函数write和read的开销。
--
wwbmmm commented on issue #2000:
URL:
https://github.com/apache/incubator-brpc/issues/2000#issuecomment-1325927072
bthread_start_urgent会把当前pthread的栈切到ProcessEvent函数,同时也会将当前bthread挂起并signal其它的worker来继续执行这个bthread,所以会对性能带来一定影响。并不是jumpstack带来的影响。
--
This is an automated message from the
zhouyanghere commented on issue #2000:
URL:
https://github.com/apache/incubator-brpc/issues/2000#issuecomment-1325919196
如果将进程用taskset绑定到固定的CPU上,也没有此差异。是不是brpc
jumpstack的过程导致线程pthread切换了CPU,导致了cache miss,从而影响系统调用的开销。
--
This is an automated message from the Apache Git Service.
To
zhouyanghere commented on issue #2000:
URL:
https://github.com/apache/incubator-brpc/issues/2000#issuecomment-1325914657
Socket::StartInputEvent 里如果直接调用ProcessEvent函数,也是函数调用过程和栈切换,为何brpc
jumpstack就能影响系统调用。是否和jumpstack的处理有关系?jumpstack是汇编代码,能力有限,看不大懂。
--
This is an automated message from
zhouyanghere commented on issue #2000:
URL:
https://github.com/apache/incubator-brpc/issues/2000#issuecomment-1325909263
可能我描述的不是很清楚,和多线程应该没有直接关系。其他的类redis数据库,比如DragonFly DB,延时也很低。
我对ProcessEvent加了bvar统计,如果ProcessEvent走bthread,则ProcessEvent函数的处理平均耗时为26us。如果不走,则ProcessEvent只有6us。单链接下最明显。
wwbmmm commented on issue #2013:
URL:
https://github.com/apache/incubator-brpc/issues/2013#issuecomment-1325908659
我觉得可以
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
yanglimingcn opened a new issue, #2013:
URL: https://github.com/apache/incubator-brpc/issues/2013
brpc有const concurrenty limiter 和 auto concurrency limiter,用于限制服务的并发度,const
concurrency limiter需要实际对服务压测配置,运营起来比较繁琐。auto concurrency limiter比较灵活,基于little’s
low法则,但是在使用过程中发现会经常报ELIMIT。
372046933 commented on code in PR #2009:
URL: https://github.com/apache/incubator-brpc/pull/2009#discussion_r1031004002
##
src/butil/iobuf.cpp:
##
@@ -1211,6 +1211,9 @@ int IOBuf::append_user_data(void* data, size_t size, void
(*deleter)(void*)) {
LOG(FATAL) <<
wwbmmm commented on code in PR #2009:
URL: https://github.com/apache/incubator-brpc/pull/2009#discussion_r1031000482
##
src/butil/iobuf.cpp:
##
@@ -1211,6 +1211,9 @@ int IOBuf::append_user_data(void* data, size_t size, void
(*deleter)(void*)) {
LOG(FATAL) <<
372046933 commented on code in PR #2009:
URL: https://github.com/apache/incubator-brpc/pull/2009#discussion_r1030998420
##
src/butil/iobuf.cpp:
##
@@ -1211,6 +1211,9 @@ int IOBuf::append_user_data(void* data, size_t size, void
(*deleter)(void*)) {
LOG(FATAL) <<
Tuvie commented on issue #1995:
URL:
https://github.com/apache/incubator-brpc/issues/1995#issuecomment-1325850024
@372046933 Thanks for your feedback. About the return value problem, maybe
@wwbmmm can share your opinion.
--
This is an automated message from the Apache Git Service.
To
372046933 commented on issue #1995:
URL:
https://github.com/apache/incubator-brpc/issues/1995#issuecomment-1325848850
@Tuvie I have tested `append_user_data_with_meta`. It works as expected. As
for the return value of `append_user_data` when size is 0. I added a pre-check
in application
zyearn merged PR #2008:
URL: https://github.com/apache/incubator-brpc/pull/2008
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
xiaohei520321 opened a new issue, #2012:
URL: https://github.com/apache/incubator-brpc/issues/2012
**Is your feature request related to a problem? (你需要的功能是否与某个问题有关?)**
最近在用brpc server,使用brpc_metrics将数据导入到prometheus上, 但发现没法获取平均latency, 看了代码
372046933 commented on code in PR #2009:
URL: https://github.com/apache/incubator-brpc/pull/2009#discussion_r1030114659
##
src/butil/iobuf.cpp:
##
@@ -1211,6 +1211,9 @@ int IOBuf::append_user_data(void* data, size_t size, void
(*deleter)(void*)) {
LOG(FATAL) <<
wwbmmm closed issue #1702: brpc访问grpc server
URL: https://github.com/apache/incubator-brpc/issues/1702
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
Tuvie commented on issue #2010:
URL:
https://github.com/apache/incubator-brpc/issues/2010#issuecomment-1324662026
在这个benchmark里面,你发送这么大的数据,瓶颈不在cpu,也不在内存拷贝,物理带宽限制才是原因。针对你的IO密集型服务,你也需要看瓶颈是什么。如果cpu不是瓶颈,瓶颈是物理带宽打满,那么带来收益的只可能是你换更高速率的网卡接入。但是我理解通常不会出现带宽优先打满的情况。
--
This is an automated message
wwbmmm closed issue #1929: stream protocol 多streams用同一个socket
URL: https://github.com/apache/incubator-brpc/issues/1929
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To
wwbmmm commented on issue #1929:
URL:
https://github.com/apache/incubator-brpc/issues/1929#issuecomment-1324641789
Fixed by https://github.com/apache/incubator-brpc/pull/1958
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
zoulei commented on issue #2010:
URL:
https://github.com/apache/incubator-brpc/issues/2010#issuecomment-1324634492
> 这个测试case里面,带宽是瓶颈,都达到25G满带宽了。时延不会有啥差距的。
我大概有点明白了,在一次请求的数据传输过程中,内存的拷贝和数据通过网络传输,并不是串行的,而是并行的,拷贝了一部分之后,就会开始并行的拷贝与传输,而内存的拷贝速度要远快于网络传输的速度,从而导致出现延迟并没有下降的结果。请问大佬们,是这样的吗?
zoulei commented on issue #2010:
URL:
https://github.com/apache/incubator-brpc/issues/2010#issuecomment-1324622578
> 这个测试case里面,带宽是瓶颈,都达到25G满带宽了。时延不会有啥差距的。
但是我看RDMA的理论介绍中,RMDA能减少一次内存的拷贝,减少这一次拷贝却没有降低延迟是因为和网络传输相比,内存的拷贝速度太快了吗?
--
This is an automated message from the Apache Git
wwbmmm commented on issue #1945:
URL:
https://github.com/apache/incubator-brpc/issues/1945#issuecomment-1324610024
rdma分支已不再维护,请使用master分支
这个先close了,如果master有问题,可以再提issue
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
wwbmmm closed issue #1945: rdma分支存在内存泄漏问题
URL: https://github.com/apache/incubator-brpc/issues/1945
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
wwbmmm commented on issue #2000:
URL:
https://github.com/apache/incubator-brpc/issues/2000#issuecomment-1324605671
brpc是多线程的,存在多线程调度延时,而原生redis service是单线程的,没有这个问题,所以没有可比性。
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and
wwbmmm commented on PR #1947:
URL: https://github.com/apache/incubator-brpc/pull/1947#issuecomment-1324596522
@Aaron 这个PR和master有冲突了,能否解决一下。并补充单测,谢谢!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above
wwbmmm commented on PR #1313:
URL: https://github.com/apache/incubator-brpc/pull/1313#issuecomment-1324594460
There is a better solution:
https://github.com/apache/incubator-brpc/pull/1777
--
This is an automated message from the Apache Git Service.
To respond to the message, please log
wwbmmm commented on PR #1777:
URL: https://github.com/apache/incubator-brpc/pull/1777#issuecomment-1324594007
> > Thanks for reporting the issue and submitting this PR! I think it's
better to add some comments to explain the code, or add some test cases, to
prevent future modification from
wwbmmm commented on PR #2009:
URL: https://github.com/apache/incubator-brpc/pull/2009#issuecomment-1324582053
LGTM
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To
guoyuhong opened a new issue, #2011:
URL: https://github.com/apache/incubator-brpc/issues/2011
**Is your feature request related to a problem? (你需要的功能是否与某个问题有关?)**
NO
**Describe the solution you'd like (描述你期望的解决方法)**
**Describe alternatives you've considered
372046933 commented on code in PR #2009:
URL: https://github.com/apache/incubator-brpc/pull/2009#discussion_r1030001048
##
src/butil/iobuf.cpp:
##
@@ -1211,6 +1211,9 @@ int IOBuf::append_user_data(void* data, size_t size, void
(*deleter)(void*)) {
LOG(FATAL) <<
Tuvie commented on issue #2010:
URL:
https://github.com/apache/incubator-brpc/issues/2010#issuecomment-1324513623
这个测试case里面,带宽是瓶颈,都达到25G满带宽了。时延不会有啥差距的。
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above
Tuvie commented on code in PR #2009:
URL: https://github.com/apache/incubator-brpc/pull/2009#discussion_r1029986626
##
src/butil/iobuf.cpp:
##
@@ -1211,6 +1211,9 @@ int IOBuf::append_user_data(void* data, size_t size, void
(*deleter)(void*)) {
LOG(FATAL) <<
Tuvie commented on code in PR #2005:
URL: https://github.com/apache/incubator-brpc/pull/2005#discussion_r1029985331
##
src/butil/iobuf.cpp:
##
@@ -1206,7 +1211,14 @@ int IOBuf::appendv(const const_iovec* vec, size_t n) {
return 0;
}
-int IOBuf::append_user_data(void*
zoulei opened a new issue, #2010:
URL: https://github.com/apache/incubator-brpc/issues/2010
![6b0b37ede60da86a9e89a6ead894ee5](https://user-images.githubusercontent.com/11936655/203461582-792c135e-8b74-4295-baa3-de9dbee82a2a.jpg)
wwbmmm commented on PR #2005:
URL: https://github.com/apache/incubator-brpc/pull/2005#issuecomment-1324465314
LGTM
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To
372046933 commented on code in PR #2009:
URL: https://github.com/apache/incubator-brpc/pull/2009#discussion_r1029969953
##
src/butil/iobuf.cpp:
##
@@ -1211,6 +1211,9 @@ int IOBuf::append_user_data(void* data, size_t size, void
(*deleter)(void*)) {
LOG(FATAL) <<
372046933 commented on code in PR #2009:
URL: https://github.com/apache/incubator-brpc/pull/2009#discussion_r1029969953
##
src/butil/iobuf.cpp:
##
@@ -1211,6 +1211,9 @@ int IOBuf::append_user_data(void* data, size_t size, void
(*deleter)(void*)) {
LOG(FATAL) <<
wwbmmm commented on code in PR #2009:
URL: https://github.com/apache/incubator-brpc/pull/2009#discussion_r1029963007
##
src/butil/iobuf.cpp:
##
@@ -1211,6 +1211,9 @@ int IOBuf::append_user_data(void* data, size_t size, void
(*deleter)(void*)) {
LOG(FATAL) <<
372046933 commented on code in PR #2005:
URL: https://github.com/apache/incubator-brpc/pull/2005#discussion_r1029957001
##
src/butil/iobuf.cpp:
##
@@ -1206,7 +1211,14 @@ int IOBuf::appendv(const const_iovec* vec, size_t n) {
return 0;
}
-int
guodongxiaren commented on PR #2008:
URL: https://github.com/apache/incubator-brpc/pull/2008#issuecomment-1323816098
@zyearn 看下
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific
372046933 opened a new pull request, #2009:
URL: https://github.com/apache/incubator-brpc/pull/2009
### What problem does this PR solve?
Issue Number:
Problem Summary:
### What is changed and the side effects?
Changed:
Side effects:
- Performance
wwbmmm commented on PR #2008:
URL: https://github.com/apache/incubator-brpc/pull/2008#issuecomment-1323020485
LGTM
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To
Tuvie commented on code in PR #2005:
URL: https://github.com/apache/incubator-brpc/pull/2005#discussion_r1028788588
##
src/butil/iobuf.cpp:
##
@@ -1219,11 +1227,23 @@ int IOBuf::append_user_data(void* data, size_t size,
void (*deleter)(void*)) {
deleter = ::free;
Tuvie commented on code in PR #2005:
URL: https://github.com/apache/incubator-brpc/pull/2005#discussion_r1028788462
##
test/brpc_rdma_unittest.cpp:
##
@@ -80,6 +80,7 @@ extern bool g_skip_rdma_init;
}
}
+static const size_t MAX_USER_MRS = 16;
Review Comment:
should
vinllen closed issue #1997: 程序内部core但是core栈不完整
URL: https://github.com/apache/incubator-brpc/issues/1997
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe,
guodongxiaren commented on PR #2008:
URL: https://github.com/apache/incubator-brpc/pull/2008#issuecomment-1322427935
测试过程中commit比较多,合并的时候可以选择 squash and merge @zyearn @wwbmmm
unittest在Linux上也有失败。后面单独提一个PR来让大家确认问题。
bazel编译,以及gcc clang分别编译,本周再提几次。
--
This is an
guodongxiaren commented on issue #1963:
URL:
https://github.com/apache/incubator-brpc/issues/1963#issuecomment-1322414866
sorry, I missed this message last week. I have submitted one pull request
for two compile task.
@zyearn
--
This is an automated message from the Apache Git
guodongxiaren opened a new pull request, #2008:
URL: https://github.com/apache/incubator-brpc/pull/2008
### What problem does this PR solve?
Issue Number:
Problem Summary:
### What is changed and the side effects?
Changed:
Side effects:
- Performance
hongliuliao opened a new pull request, #2007:
URL: https://github.com/apache/incubator-brpc/pull/2007
### What problem does this PR solve?
Issue Number: NONE
Problem Summary: make brpc user know the graph means
### What is changed and the side effects?
Changed:
wwbmmm commented on code in PR #2005:
URL: https://github.com/apache/incubator-brpc/pull/2005#discussion_r1027626417
##
src/butil/iobuf.h:
##
@@ -66,6 +66,7 @@ friend class IOBufCutter;
public:
static const size_t DEFAULT_BLOCK_SIZE = 8192;
static const size_t
Tuvie commented on issue #1995:
URL:
https://github.com/apache/incubator-brpc/issues/1995#issuecomment-1321491547
> Thanks a million. Is `append_user_data_with_meta` in the PR? I couldn't
find the definition in
serverglen commented on issue #2002:
URL:
https://github.com/apache/incubator-brpc/issues/2002#issuecomment-1321422814
指的是ServerOptions中的`const Authenticator* auth;`?,按照如下方式设置试试:
```c++
brpc::ServerOptions options;
options.auth = nullptr;
if (server.Start(point, ) != 0) {
serverglen commented on PR #2003:
URL: https://github.com/apache/incubator-brpc/pull/2003#issuecomment-1321408977
Could you please submit the application scenario of brpc in your company?
See #1640
--
This is an automated message from the Apache Git Service.
To respond to the message,
wwbmmm commented on PR #2005:
URL: https://github.com/apache/incubator-brpc/pull/2005#issuecomment-1321385642
`append_user_data_with_meta` not found in files changed
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
wwbmmm merged PR #2004:
URL: https://github.com/apache/incubator-brpc/pull/2004
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
vinllen commented on issue #1997:
URL:
https://github.com/apache/incubator-brpc/issues/1997#issuecomment-1321346714
> > > 这个是你业务代码抛出了异常,没有catch,异常逃逸到了brpc框架内。
> >
> >
> > 业务代码抛异常为什么会跑到brpc的 SocketDeleter::operator() 函数里呢?
> > 这种堆栈以前也见过,但想不明白为啥会core在这里
>
> ```c++
>
372046933 commented on code in PR #1996:
URL: https://github.com/apache/incubator-brpc/pull/1996#discussion_r1027297652
##
example/build_with_bazel/brpc_workspace.bzl:
##
@@ -74,10 +74,15 @@ def brpc_workspace():
],
)
-native.new_local_repository(
-
fansehep commented on PR #1996:
URL: https://github.com/apache/incubator-brpc/pull/1996#issuecomment-1321134546
> >
>
> Ok, I will try it later.
@Huixxi Have you test successful in ```MacOS```?
Here is a simple demo in this pull request.
```bash
cd
chenBright opened a new issue, #2006:
URL: https://github.com/apache/incubator-brpc/issues/2006
了解了一下协议栈的[close逻辑](https://github.com/torvalds/linux/blob/master/net/ipv4/tcp.c#L2904):
1. 如果接收缓冲区没有数据,则会将fin插入到发送缓冲区最后面,先将数据发送完,再发fin,进行四次挥手。
2. 如果接收缓冲区还有数据,则直接发送rst,不会进行四次挥手。
372046933 commented on code in PR #1996:
URL: https://github.com/apache/incubator-brpc/pull/1996#discussion_r1027274956
##
example/build_with_bazel/brpc_workspace.bzl:
##
@@ -0,0 +1,94 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor
372046933 commented on issue #1995:
URL:
https://github.com/apache/incubator-brpc/issues/1995#issuecomment-1321077817
Thanks a million. Is `append_user_data_with_meta` in the PR? I couldn't find
the definition in
Tuvie commented on issue #1995:
URL:
https://github.com/apache/incubator-brpc/issues/1995#issuecomment-1321032110
@372046933 Pls try this PR:
https://github.com/apache/incubator-brpc/pull/2005
A new function named append_user_data_with_meta is provided. You should give
the lkey
Tuvie opened a new pull request, #2005:
URL: https://github.com/apache/incubator-brpc/pull/2005
### What problem does this PR solve?
Issue Number: [1995](https://github.com/apache/incubator-brpc/issues/1995)
Problem Summary: When using IOBuf::append_user_data, if the
Tuvie closed pull request #1999: allow IOBuf::append_user_data specifying any
valid address in registered rdma memory region
URL: https://github.com/apache/incubator-brpc/pull/1999
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
guodongxiaren commented on issue #1997:
URL:
https://github.com/apache/incubator-brpc/issues/1997#issuecomment-1320943345
> > 这个是你业务代码抛出了异常,没有catch,异常逃逸到了brpc框架内。
>
>
>
> 业务代码抛异常为什么会跑到brpc的 SocketDeleter::operator() 函数里呢?
>
> 这种堆栈以前也见过,但想不明白为啥会core在这里
```cpp
guodongxiaren commented on PR #1996:
URL: https://github.com/apache/incubator-brpc/pull/1996#issuecomment-1320817221
> >
>
> Ok, I will try it later.
Mac能生效吗?
@Huixxi
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
guodongxiaren opened a new pull request, #2004:
URL: https://github.com/apache/incubator-brpc/pull/2004
### What is changed and the side effects?
Changed: fix bad smell of C++ code
Side effects:
- Performance effects(性能影响): no
- Breaking backward
changchengx commented on PR #2003:
URL: https://github.com/apache/incubator-brpc/pull/2003#issuecomment-1319715850
@wwbmmm No. From semantic perspective, memset doesn't consider variable type
at all and guarantee atomic operation.
--
This is an automated message from the Apache Git
wwbmmm commented on PR #2003:
URL: https://github.com/apache/incubator-brpc/pull/2003#issuecomment-1319702633
@changchengx Do you have any problem using memset ? (eg. compile error or
unexpected behavior at runtime)
--
This is an automated message from the Apache Git Service.
To respond
changchengx opened a new pull request, #2003:
URL: https://github.com/apache/incubator-brpc/pull/2003
1. memset assign value to var without considering var type.
2. memset can't guarantee atomic operation
Signed-off-by: Liu, Changcheng
--
This is an automated message from
Lowpower opened a new issue, #2002:
URL: https://github.com/apache/incubator-brpc/issues/2002
**Is your feature request related to a problem? (你需要的功能是否与某个问题有关?)**
**Describe the solution you'd like (描述你期望的解决方法)**
**Describe alternatives you've considered (描述你想到的折衷方案)**
shukean opened a new issue, #2001:
URL: https://github.com/apache/incubator-brpc/issues/2001
**Describe the bug (描述bug)**
当重试次数>=3时, 开启 backup request 以及 retry policy 时,
会触发 CHECK_EQ(current_id(), info.id) << "error_code=" << _error_code;
**To Reproduce (复现方法)**
1) 发送一次rpc:
zhouyanghere commented on issue #2000:
URL:
https://github.com/apache/incubator-brpc/issues/2000#issuecomment-1318310432
有没有办法在低压力下提供低延时又在高压力下高吞吐。
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go
zhouyanghere opened a new issue, #2000:
URL: https://github.com/apache/incubator-brpc/issues/2000
**Describe the bug (描述bug)**
使用redis-benchmark -t set -c 1 -n 10单链接无压力测试,对比redis
server的20+us的单个请求延时,brpc redis server一般达到了40us+
**To Reproduce (复现方法)**
redis-benchmark -t set
372046933 commented on issue #1995:
URL:
https://github.com/apache/incubator-brpc/issues/1995#issuecomment-1318280013
Yes, we can manage `LKey`. Does adding parameter to `append_user_data `
breaks backward compatibility?
--
This is an automated message from the Apache Git Service.
To
Tuvie commented on issue #1995:
URL:
https://github.com/apache/incubator-brpc/issues/1995#issuecomment-1318269408
> @Tuvie We use `append_user_data` in tensorflow's
`AsyncOpKernel::ComputeAsync`. The memory comes from tensor in TensorFlow OP.
It's allocated by the framework on each run.
wwbmmm commented on code in PR #1999:
URL: https://github.com/apache/incubator-brpc/pull/1999#discussion_r1024872546
##
src/brpc/rdma/rdma_helper.cpp:
##
@@ -99,37 +99,41 @@ static ibv_device** g_devices = NULL;
static ibv_context* g_context = NULL;
static SocketId
372046933 commented on issue #1995:
URL:
https://github.com/apache/incubator-brpc/issues/1995#issuecomment-1318247334
@Tuvie We use `append_user_data` in tensorflow's
`AsyncOpKernel::ComputeAsync`. The memory comes from tensor in TensorFlow OP.
It's allocated by the framework on each run.
401 - 500 of 7562 matches
Mail list logo