Apps are increasingly using more than 1024 file descriptors.
See discussion in several distro bug trackers, e.g.
https://bugs.launchpad.net/bugs/663090
https://issues.rpath.com/browse/RPL-2054
Cc: Dan Kegel <[email protected]>
Signed-off-by: Tim Gardner <[email protected]>
---
include/asm-generic/resource.h | 2 +-
include/linux/fs.h | 3 ++-
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/include/asm-generic/resource.h b/include/asm-generic/resource.h
index 587566f..61fa862 100644
--- a/include/asm-generic/resource.h
+++ b/include/asm-generic/resource.h
@@ -78,7 +78,7 @@
[RLIMIT_CORE] = { 0, RLIM_INFINITY }, \
[RLIMIT_RSS] = { RLIM_INFINITY, RLIM_INFINITY }, \
[RLIMIT_NPROC] = { 0, 0 }, \
- [RLIMIT_NOFILE] = { INR_OPEN, INR_OPEN }, \
+ [RLIMIT_NOFILE] = { INR_OPEN_CUR, INR_OPEN_MAX }, \
[RLIMIT_MEMLOCK] = { MLOCK_LIMIT, MLOCK_LIMIT }, \
[RLIMIT_AS] = { RLIM_INFINITY, RLIM_INFINITY }, \
[RLIMIT_LOCKS] = { RLIM_INFINITY, RLIM_INFINITY }, \
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 52f283c..6e1e4e6 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -23,7 +23,8 @@
/* Fixed constants first: */
#undef NR_OPEN
-#define INR_OPEN 1024 /* Initial setting for nfile rlimits */
+#define INR_OPEN_CUR 1024 /* Initial setting for nfile rlimits */
+#define INR_OPEN_MAX 4096 /* Hard limit for nfile rlimits */
#define BLOCK_SIZE_BITS 10
#define BLOCK_SIZE (1<<BLOCK_SIZE_BITS)
--
1.7.0.4
On Thu, 31 Mar 2011 15:26:50 -0600
Tim Gardner <[email protected]> wrote:
> Apps are increasingly using more than 1024 file descriptors.
> See discussion in several distro bug trackers, e.g.
> https://bugs.launchpad.net/bugs/663090
> https://issues.rpath.com/browse/RPL-2054
>
^^ how not to write a changelog.
Here's the useful part, from Dan's bugzilla report:
: You don't want to raise the default soft limit, since that might break
: apps that use select(), but it's safe to raise the default hard limit;
: that way, apps that know they need lots of file descriptors can raise their
: soft limit without needing root, and without user intervention.
:
: Ubuntu is doing this for 11.04 by changing include/asm-generic/resource.h:
:
: - [RLIMIT_NOFILE] = { INR_OPEN, INR_OPEN }, \
: + [RLIMIT_NOFILE] = { INR_OPEN, INR_OPEN*4 }, \
:
: They chose to do this with a kernel change because they have a policy
: of not changing kernel defaults in userland.
:
: While 4096 might not be enough for *all* apps, it seems to be plenty
: for the apps I've seen lately that are unhappy with 1024.
> ---
> include/asm-generic/resource.h | 2 +-
> include/linux/fs.h | 3 ++-
> 2 files changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/include/asm-generic/resource.h b/include/asm-generic/resource.h
> index 587566f..61fa862 100644
> --- a/include/asm-generic/resource.h
> +++ b/include/asm-generic/resource.h
> @@ -78,7 +78,7 @@
> [RLIMIT_CORE] = { 0, RLIM_INFINITY }, \
> [RLIMIT_RSS] = { RLIM_INFINITY, RLIM_INFINITY }, \
> [RLIMIT_NPROC] = { 0, 0 }, \
> - [RLIMIT_NOFILE] = { INR_OPEN, INR_OPEN }, \
> + [RLIMIT_NOFILE] = { INR_OPEN_CUR, INR_OPEN_MAX }, \
> [RLIMIT_MEMLOCK] = { MLOCK_LIMIT, MLOCK_LIMIT }, \
> [RLIMIT_AS] = { RLIM_INFINITY, RLIM_INFINITY }, \
> [RLIMIT_LOCKS] = { RLIM_INFINITY, RLIM_INFINITY }, \
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 52f283c..6e1e4e6 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -23,7 +23,8 @@
>
> /* Fixed constants first: */
> #undef NR_OPEN
> -#define INR_OPEN 1024 /* Initial setting for nfile rlimits */
> +#define INR_OPEN_CUR 1024 /* Initial setting for nfile rlimits */
> +#define INR_OPEN_MAX 4096 /* Hard limit for nfile rlimits */
>
> #define BLOCK_SIZE_BITS 10
> #define BLOCK_SIZE (1<<BLOCK_SIZE_BITS)
> --
> 1.7.0.4
On 03/31/2011 03:35 PM, Andrew Morton wrote:
>
> ^^ how not to write a changelog.
>
It seemed like plagiarism to just copy Dan's argument, but now here you
have it.
rtg
--
Tim Gardner [email protected]
On Fri, Apr 1, 2011 at 1:30 AM, Tim Gardner <[email protected]> wrote:
> On 03/31/2011 03:35 PM, Andrew Morton wrote:
>>
>> ^^ how not to write a changelog.
>
> It seemed like plagiarism to just copy Dan's argument, but now here you have
> it.
I don't mind one bit. Very happy to see this change going in.
- Dan