Re: [Openais] [openais-whitetank PATCH 2/2] IPC: if shmget() or semget() return ENOSPC then exit the while loop.

As discussed on IRC, this is same patch as: [Openais] [PATCH 1/1] whitetank: Return error if there is no space for semaphore Angus Salkeld napsal(a): > I noticed this while testing the high FD usage patch. > It can get stuck in the while loop. > > Signed-off-by: Angus Salkeld > --- > lib/uti

Re: [Openais] [openais-whitetank PATCH 1/2] IPC: handle maximum use of file descriptors gracefully.

Reviewed-by: Jan Friesse Angus Salkeld napsal(a): > If we get EMFILE from accept() then withdraw the published > server listening socket. Then when a connection closes > see if we need to re-publish the server sockect. > > Signed-off-by: Angus Salkeld > --- > exec/ipc.c | 161

[Openais] [openais-whitetank PATCH 1/2] IPC: handle maximum use of file descriptors gracefully.

If we get EMFILE from accept() then withdraw the published server listening socket. Then when a connection closes see if we need to re-publish the server sockect. Signed-off-by: Angus Salkeld --- exec/ipc.c | 161 +-- 1 files changed, 101

[Openais] [openais-whitetank PATCH 2/2] IPC: if shmget() or semget() return ENOSPC then exit the while loop.

I noticed this while testing the high FD usage patch. It can get stuck in the while loop. Signed-off-by: Angus Salkeld --- lib/util.c |6 ++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/lib/util.c b/lib/util.c index f285eb5..76b3891 100644 --- a/lib/util.c +++ b/lib/uti

Re: [Openais] [PATCH 1/1] whitetank: Return error if there is no space for semaphore

On Thu, Mar 17, 2011 at 05:56:46PM +0100, Jan Friesse wrote: > open_ais_service_connect creates SYS V semaphores and shms. > If system limit for semaphores/shms is exceeded code looped > in endless cycle. > Reviewed-by: Angus Salkeld > Now ENOSPC is correctly handled and SA_AIS_ERR_NO_SPACE is