Attachment 577113 Details for Bug 810917 – 6.2.z-bz810917-1-Unlink-shm-buffers-if-init-fails

[patch] 6.2.z-bz810917-1-Unlink-shm-buffers-if-init-fails

bz810917-1-Unlink-shm-buffers-if-init-fails.patch (text/plain), 1.49 KB, created by Jan Friesse on 2012-04-12 15:35:55 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Jan Friesse

Created: 2012-04-12 15:35:55 UTC

Size: 1.49 KB

patch

obsolete

>Author: Jan Friesse <jfriesse@redhat.com>
>Date:   Thu Apr 12 17:33:51 2012 +0200
>
>    Unlink shm buffers if init fails
>    
>    If ipc init failed, buffers was not unlinked nether by client (lib) side
>    nor server (corosync) side. This may lead to fill all available space,
>    resulting in no accept of other connection. Typical example can be user
>    running any corosync ipc binary (like corosync-objctl), without correct
>    uid/gid entry in corosync configuration, resulting in DOS.
>    
>    Signed-off-by: Jan Friesse <jfriesse@redhat.com>
>    Reviewed-by: Steven Dake <sdake@redhat.com>
>    (Backported for 6.2.z)
>
>diff --git a/lib/coroipcc.c b/lib/coroipcc.c
>index b7a3db9..16cc2b3 100644
>--- a/lib/coroipcc.c
>+++ b/lib/coroipcc.c
>@@ -742,7 +742,7 @@ coroipcc_service_connect (
> 
> 	ipc_instance->fd = request_fd;
> 
>-	if (res_setup.error == CS_ERR_TRY_AGAIN) {
>+	if (res_setup.error != CS_OK) {
> 		res = res_setup.error;
> 		goto error_exit;
> 	}
>@@ -764,12 +764,16 @@ error_exit:
> 		semctl (ipc_instance->control_buffer->semid, 0, IPC_RMID);
> #endif
> 	memory_unmap (ipc_instance->dispatch_buffer, dispatch_size);
>+	unlink (dispatch_map_path);
> error_dispatch_buffer:
> 	memory_unmap (ipc_instance->response_buffer, response_size);
>+	unlink (response_map_path);
> error_response_buffer:
> 	memory_unmap (ipc_instance->request_buffer, request_size);
>+	unlink (request_map_path);
> error_request_buffer:
> 	memory_unmap (ipc_instance->control_buffer, 8192);
>+	unlink (control_map_path);
> error_connect:
> 	close (request_fd);
>

Actions: View | Diff

Attachments on bug 810917: 577113