_rpc_dtablesize: decrease to fix excessive memory usage
Subsystems (2)
cover _rpc_dtablesize: decrease to fix excessive memory usage
In the client code, the function _rpc_dtablesize() is used to determine the memory allocation for the __svc_xports array.
However, some operating systems (including the recent Manjaro OS) can have _SC_OPEN_MAX values as high as 1073741816, which can cause the __svc_xports array to become too large. This results in the process being killed.
There is a limit to the maximum number of files. To avoid this problem, a possible solution is to set the size to the lesser of 1024 and this value to ensure that the array space for open files is not too large, thus preventing the process from terminating.
Also discovered that some users have taken action on the issue. It is necessary to address this for all users. Ultimately, we determined that adjusting the size value of _rpc_dtablesize() and streamlining the existing user code would be the most effective solution.
---
Changes in v1:
Clean up the existing code in user
---
Links:
RFC:https://lore.kernel.org/linux-nfs/[email protected]/T/#u
Zhuohao Bai (2):
_rpc_dtablesize: Decrease the value of size.
_rpc_dtablesize: Cleaning up the existing code
src/rpc_dtablesize.c | 2 ++ src/svc.c | 2 -- 2 files changed, 2 insertions(+), 2 deletions(-)
Signature · 1 line
2.25.1
PATCH 1/2 _rpc_dtablesize: Decrease the value of size.
To fix the bug caused by a Size value that is too large and leads to an array that requires excessive memory, which subsequently results in the failure of rpcbind to start properly.
--- src/rpc_dtablesize.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
src/rpc_dtablesize.c +1 -1
@@ -41,7 +41,7 @@ _rpc_dtablesize(void)static int size;if (size ==0 ) { - size =sysconf (_SC_OPEN_MAX); + size =min (1024 ,sysconf (_SC_OPEN_MAX)); }return (size); }
Signature · 1 line
2.25.1
PATCH 2/2 _rpc_dtablesize: Cleaning up the existing code
In reply to Zhuohao BaiSome users have already taken steps to address this issue. To prevent duplication, we need to remove the modified code and modify the rpc_dtablesize function.
--- src/rpc_dtablesize.c | 4 +++- src/svc.c | 2 -- 2 files changed, 3 insertions(+), 3 deletions(-)
src/rpc_dtablesize.c +3 -1
@@ -41,7 +41,9 @@ _rpc_dtablesize(void)static int size;if (size ==0 ) { - size =min (1024 ,sysconf (_SC_OPEN_MAX)); + size =sysconf (_SC_OPEN_MAX); +if (size >FD_SETSIZE ) + size =FD_SETSIZE ; }return (size); }
src/svc.c +0 -2
@@ -657,8 +657,6 @@ svc_getreqset (readfds)assert (readfds !=NULL ); setsize =_rpc_dtablesize (); -if (setsize >FD_SETSIZE ) - setsize =FD_SETSIZE ; maskp = readfds->fds_bits;for (sock =0 ; sock < setsize; sock +=NFDBITS ) {
Signature · 1 line
2.25.1
Quoted 27 lines · Zhuohao Bai
In the client code, the function
_rpc_dtablesize()is used to determine the memory allocation for the__svc_xportsarray.However, some operating systems (including the recent Manjaro OS) can have
_SC_OPEN_MAXvalues as high as 1073741816, which can cause the__svc_xportsarray to become too large. This results in the process being killed.There is a limit to the maximum number of files. To avoid this problem, a possible solution is to set the size to the lesser of 1024 and this value to ensure that the array space for open files is not too large, thus preventing the process from terminating.
Also discovered that some users have taken action on the issue. It is necessary to address this for all users. Ultimately, we determined that adjusting the size value of
_rpc_dtablesize()and streamlining the existing user code would be the most effective solution.---
Changes in v1:
Clean up the existing code in user---
Links:
RFC:https://lore.kernel.org/linux-nfs/[email protected]/T/#u
Quoted 8 lines
Zhuohao Bai (2):
_rpc_dtablesize: Decrease the value of size.
_rpc_dtablesize: Cleaning up the existing codesrc/rpc_dtablesize.c | 2 ++ src/svc.c | 2 -- 2 files changed, 2 insertions(+), 2 deletions(-)
Both Committed... (tag: libtirpc-1-3-5-rc2)
steved.