I didn't found the attachment here. Can you put it somewhere. Maybe as
github gist?
On Monday, May 9, 2022 at 4:18:52 AM UTC+2 garenchan wrote:
> /proc//limits of the two processes are the same.
> But /proc//smaps of the two processes are quite different, I have
> uploaded them to the attachme
/proc//limits of the two processes are the same.
But /proc//smaps of the two processes are quite different, I have
uploaded them to the attachment.
在2022年5月8日星期日 UTC+8 16:08:50 写道：
> I cannot say what the difference is, but I would do two things to solve
> the mystery:
>
> 1) Compare /proc//li
I cannot say what the difference is, but I would do two things to solve the
mystery:
1) Compare /proc//limits for differences.
2) Compare the output of egrep '^Rss:|^[0-9a-f]' /proc//smaps for the
processes on the different machines.
Report the results and share the outputs.
On Friday, May 6,
https://www.arp242.net/static-go.html
> On May 5, 2022, at 11:27 PM, garenchan wrote:
>
> The binary file is not a dynamic executable. So I don't think that will
> happen.
>
> 在2022年5月6日星期五 UTC+8 11:54:44 写道：
>> Are you certain that on the low memory instance there isn’t another process
>>
The binary file is not a dynamic executable. So I don't think that will
happen.
在2022年5月6日星期五 UTC+8 11:54:44 写道：
> Are you certain that on the low memory instance there isn’t another
> process - maybe Go - that hasn’t already loaded the shared libraries - and
> on the other it hasn’t - so th
Are you certain that on the low memory instance there isn’t another process -
maybe Go - that hasn’t already loaded the shared libraries - and on the other
it hasn’t - so the RSS is different. Compare the VSZ sizes and see if they are
the same.
> On May 5, 2022, at 10:28 PM, garenchan wrote:
@jake, I tried 1.18 version of Go, but the problem still exists.
I also tried using C to write program that includes memory allocation and
usage, and this problem does not occur.
在2022年5月5日星期四 UTC+8 21:25:20 写道：
> Since no one has responded with concrete ideas, I'll throw out two
> suggestions
Since no one has responded with concrete ideas, I'll throw out two
suggestions. They may seem obvious.
First have you tries the latest version of Go? and do you get the same
results?
Second have you run the experiment with a small binaries not from Go? I
would suggest something that does all
Both hosts have 8 cores and 16GB RAM.
在2022年4月30日星期六 UTC+8 00:19:44 写道：
> *What version of Go are you using (go version)?*
>
> $ go version
> go version go1.17.6 linux/amd64
>
> *Does this issue reproduce with the latest release?*
>
> uncertain
>
> *What operating system and processor architectur
