Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

100% CPU when the _SC_OPEN_MAX is a big number #147

Closed
whywhy36 opened this issue Apr 4, 2024 · 3 comments
Closed

100% CPU when the _SC_OPEN_MAX is a big number #147

whywhy36 opened this issue Apr 4, 2024 · 3 comments

Comments

@whywhy36
Copy link

whywhy36 commented Apr 4, 2024 •
edited
Loading

We observed that it took very long time to open a JetBrains terminal in the new Amazon AL2023 VM. The CPU usage is 100% for process PtyProcess Reap (the full name of that thread is PtyProcess Reaper for [/bin/bash, --rcfile, /home/owner/.cache/JetBrains/RemoteDev/dist/ideaIU-2022.3.2/plugins/terminal/jediterm-bash.in, -i]).

strace -p PID indicates that JNA call was trying to close a huge range of FDs, which should be defined in exe_pty.c. If _SC_OPEN_MAX holds a big value, the loop would degrade the performance.

The similar issue has been reported in other projects, such as

Wondering should we apply a similar patch as netdata/netdata#14213 to the impacted codes.
@whywhy36
Copy link
Author

whywhy36 commented Apr 4, 2024

The issue should also be fixed by PR #124

@bigfoot547 bigfoot547 mentioned this issue Jun 11, 2024
Closed
@morgwai
Copy link

morgwai commented Jun 16, 2024 •
edited
Loading

For Linux users affected by this problem and looking for a workaround until #149 is released:
Decreasing the limits of maximum open files mitigates the problem. See this stackExchange post explaining how to set global and per process limits.
For me setting fs.file-max=10485760 in sysctl and * hard nofile 1048576 in limits.conf was sufficiently low to make it work.

segrey added a commit that referenced this issue Jun 16, 2024
@segrey
speed up closing all file descriptors in the child process (Fixes #125 …
04685d8 
…#147)
@segrey
Copy link
Collaborator

segrey commented Jun 17, 2024

Thank you for reporting and sorry for the delay. Should be fixed by 04685d8 and available since pty4j version 0.12.27.

@segrey segrey closed this as completed Jun 17, 2024
@adangel adangel mentioned this issue Jun 21, 2024
Closed
adangel added a commit to adangel/cdt that referenced this issue Jun 22, 2024
@adangel @segrey
Speed up closing all file descriptors in the child process (Fixes ecl…
adef8eb 
…ipse-cdt#855)

When _SC_OPEN_MAX (max nr of open files limit per process) is a very
big number, then closing all possible file handles can take a while.
This change first tries to use the syscall close_range() if available,
falling back to use /proc/self/fd to close only open file handles,
and lastly falling back to the old way of closing all possible handles
one after another.
In general, the first or second approach should be available
which speeds up the pty spawning.

Refs JetBrains/pty4j#147
Copied from JetBrains/pty4j@04685d8
(which is EPL 1.0)

Co-authored-by: Sergey Simonchik <sergey.simonchik@jetbrains.com>
@adangel adangel mentioned this issue Jun 22, 2024
Merged
adangel added a commit to adangel/cdt that referenced this issue Jun 22, 2024
@adangel @segrey
Speed up closing all file descriptors in the child process (Fixes ecl…
23af0ec 
…ipse-cdt#835)

When _SC_OPEN_MAX (max nr of open files limit per process) is a very
big number, then closing all possible file handles can take a while.
This change first tries to use the syscall close_range() if available,
falling back to use /proc/self/fd to close only open file handles,
and lastly falling back to the old way of closing all possible handles
one after another.
In general, the first or second approach should be available
which speeds up the pty spawning.

Refs JetBrains/pty4j#147
Copied from JetBrains/pty4j@04685d8
(which is EPL 1.0)

Co-authored-by: Sergey Simonchik <sergey.simonchik@jetbrains.com>
jonahgraham pushed a commit to eclipse-cdt/cdt that referenced this issue Jun 23, 2024
@adangel @segrey @jonahgraham
Speed up closing all file descriptors in the child process (Fixes #835)
7bd8d52 
When _SC_OPEN_MAX (max nr of open files limit per process) is a very
big number, then closing all possible file handles can take a while.
This change first tries to use the syscall close_range() if available,
falling back to use /proc/self/fd to close only open file handles,
and lastly falling back to the old way of closing all possible handles
one after another.
In general, the first or second approach should be available
which speeds up the pty spawning.

Refs JetBrains/pty4j#147
Copied from JetBrains/pty4j@04685d8
(which is EPL 1.0)

Co-authored-by: Sergey Simonchik <sergey.simonchik@jetbrains.com>
jonahgraham pushed a commit to jonahgraham/cdt that referenced this issue Jun 27, 2024
@adangel @segrey @jonahgraham
Speed up closing all file descriptors in the child process (Fixes ecl…
749f0e6 
…ipse-cdt#835)

_This commit adds to eclipse-cdt#840 7bd8d52 to apply the same fix to
another place that does the same operation on all fds._

When _SC_OPEN_MAX (max nr of open files limit per process) is a very
big number, then closing all possible file handles can take a while.
This change first tries to use the syscall close_range() if available,
falling back to use /proc/self/fd to close only open file handles,
and lastly falling back to the old way of closing all possible handles
one after another.
In general, the first or second approach should be available
which speeds up the pty spawning.

Refs JetBrains/pty4j#147
Copied from JetBrains/pty4j@04685d8
(which is EPL 1.0)

Co-authored-by: Sergey Simonchik <sergey.simonchik@jetbrains.com>
@jonahgraham jonahgraham mentioned this issue Jun 27, 2024
Merged
jonahgraham pushed a commit to eclipse-cdt/cdt that referenced this issue Jun 28, 2024
@adangel @segrey @jonahgraham
Speed up closing all file descriptors in the child process (Fixes #835)
3875408 
_This commit adds to #840 7bd8d52 to apply the same fix to
another place that does the same operation on all fds._

When _SC_OPEN_MAX (max nr of open files limit per process) is a very
big number, then closing all possible file handles can take a while.
This change first tries to use the syscall close_range() if available,
falling back to use /proc/self/fd to close only open file handles,
and lastly falling back to the old way of closing all possible handles
one after another.
In general, the first or second approach should be available
which speeds up the pty spawning.

Refs JetBrains/pty4j#147
Copied from JetBrains/pty4j@04685d8
(which is EPL 1.0)

Co-authored-by: Sergey Simonchik <sergey.simonchik@jetbrains.com>
jonahgraham pushed a commit to jonahgraham/cdt that referenced this issue Jun 28, 2024
@adangel
Speed up closing all file descriptors in the child process (Fixes ecl…
9fc1e02 
…ipse-cdt#835)

When _SC_OPEN_MAX (max nr of open files limit per process) is a very
big number, then closing all possible file handles can take a while.
This change first tries to use the syscall close_range() if available,
falling back to use /proc/self/fd to close only open file handles,
and lastly falling back to the old way of closing all possible handles
one after another.
In general, the first or second approach should be available
which speeds up the pty spawning.

Refs JetBrains/pty4j#147
Copied from JetBrains/pty4j@04685d8
(which is EPL 1.0)

Co-authored-by: Sergey Simonchik <sergey.simonchik@jetbrains.com>
(cherry picked from commit 7bd8d52)
jonahgraham pushed a commit to jonahgraham/cdt that referenced this issue Jun 28, 2024
@adangel
Speed up closing all file descriptors in the child process (Fixes ecl…
b017ae0 
…ipse-cdt#835)

_This commit adds to eclipse-cdt#840 7bd8d52 to apply the same fix to
another place that does the same operation on all fds._

When _SC_OPEN_MAX (max nr of open files limit per process) is a very
big number, then closing all possible file handles can take a while.
This change first tries to use the syscall close_range() if available,
falling back to use /proc/self/fd to close only open file handles,
and lastly falling back to the old way of closing all possible handles
one after another.
In general, the first or second approach should be available
which speeds up the pty spawning.

Refs JetBrains/pty4j#147
Copied from JetBrains/pty4j@04685d8
(which is EPL 1.0)

Co-authored-by: Sergey Simonchik <sergey.simonchik@jetbrains.com>
(cherry picked from commit 3875408)
jonahgraham pushed a commit to jonahgraham/cdt that referenced this issue Jun 28, 2024
@adangel @jonahgraham
Speed up closing all file descriptors in the child process (Fixes ecl…
af4dd0a 
…ipse-cdt#835)

When _SC_OPEN_MAX (max nr of open files limit per process) is a very
big number, then closing all possible file handles can take a while.
This change first tries to use the syscall close_range() if available,
falling back to use /proc/self/fd to close only open file handles,
and lastly falling back to the old way of closing all possible handles
one after another.
In general, the first or second approach should be available
which speeds up the pty spawning.

Refs JetBrains/pty4j#147
Copied from JetBrains/pty4j@04685d8
(which is EPL 1.0)

Co-authored-by: Sergey Simonchik <sergey.simonchik@jetbrains.com>
cherry picked and squashed from commits:
- 7bd8d52
- 24d9bd1
- 3875408
jonahgraham pushed a commit to jonahgraham/cdt that referenced this issue Jun 29, 2024
@adangel @jonahgraham
Speed up closing all file descriptors in the child process (Fixes ecl…
3588370 
…ipse-cdt#835)

When _SC_OPEN_MAX (max nr of open files limit per process) is a very
big number, then closing all possible file handles can take a while.
This change first tries to use the syscall close_range() if available,
falling back to use /proc/self/fd to close only open file handles,
and lastly falling back to the old way of closing all possible handles
one after another.
In general, the first or second approach should be available
which speeds up the pty spawning.

Refs JetBrains/pty4j#147
Copied from JetBrains/pty4j@04685d8
(which is EPL 1.0)

Co-authored-by: Sergey Simonchik <sergey.simonchik@jetbrains.com>
cherry picked and squashed from commits:
- 7bd8d52
- 24d9bd1
- 3875408
jonahgraham pushed a commit to eclipse-cdt/cdt that referenced this issue Jun 29, 2024
@adangel @jonahgraham
Speed up closing all file descriptors in the child process (Fixes #835)
f40c481 
When _SC_OPEN_MAX (max nr of open files limit per process) is a very
big number, then closing all possible file handles can take a while.
This change first tries to use the syscall close_range() if available,
falling back to use /proc/self/fd to close only open file handles,
and lastly falling back to the old way of closing all possible handles
one after another.
In general, the first or second approach should be available
which speeds up the pty spawning.

Refs JetBrains/pty4j#147
Copied from JetBrains/pty4j@04685d8
(which is EPL 1.0)

Co-authored-by: Sergey Simonchik <sergey.simonchik@jetbrains.com>
cherry picked and squashed from commits:
- 7bd8d52
- 24d9bd1
- 3875408
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

close all FDs in child faster(er) bigfoot547/pty4j
3 participants
@whywhy36 @segrey @morgwai