If I wanted to get support for another filesystem type added, and I had the code changes necessary to do it, what should I do to go about getting the change incorporated into the default builds?
Thanks,
Dave
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Which filesystem do you have in mind? Generally, you just need to add the filesystem type to a function checking whether this filesystem supports quota. I can do that for you if the filesystem is reasonable enough ;).
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It's for EMC's Multi-Path File System (MPFS). For the most part, it's exactly like NFS. The only difference is that it will service read and write I/O requests directly over Fibre Channel or iSCSI. All other requests get passed on to the NFS stack. MPFS is currently a proprietary implementation of the IETF draft for block pNFS, proposed to be included in NFS 4.1. http://www.emc.com/products/software/celerra_mpfs/
Sorry about the line wraps, if that happened in the post. I could send you the patch files if you want.
I also have a patch that seems to work with 3.12 and existing RHEL4 systems, but since nfs_fstype() function, that patch is a little larger. Could I upload it somewhere for you?
Thanks,
Dave
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
If I wanted to get support for another filesystem type added, and I had the code changes necessary to do it, what should I do to go about getting the change incorporated into the default builds?
Thanks,
Dave
Which filesystem do you have in mind? Generally, you just need to add the filesystem type to a function checking whether this filesystem supports quota. I can do that for you if the filesystem is reasonable enough ;).
Hi Jan,
It's for EMC's Multi-Path File System (MPFS). For the most part, it's exactly like NFS. The only difference is that it will service read and write I/O requests directly over Fibre Channel or iSCSI. All other requests get passed on to the NFS stack. MPFS is currently a proprietary implementation of the IETF draft for block pNFS, proposed to be included in NFS 4.1. http://www.emc.com/products/software/celerra_mpfs/
Here's the patch that seems to work for 3.14:
--- quotasys.c.orig 2007-07-16 06:31:02.000000000 -0400
+++ quotasys.c 2007-07-16 07:16:40.000000000 -0400
@@ -43,7 +43,7 @@
*/
int nfs_fstype(char *type)
{
- return !strcmp(type, MNTTYPE_NFS) || !strcmp(type, MNTTYPE_NFS4);
+ return !strcmp(type, MNTTYPE_NFS) || !strcmp(type, MNTTYPE_NFS4) || !strcmp(type, MNTTYPE_MPFS);
}
/*
@@ -68,6 +68,7 @@
!strcmp(type, MNTTYPE_REISER) ||
!strcmp(type, MNTTYPE_XFS) ||
!strcmp(type, MNTTYPE_NFS) ||
+ !strcmp(type, MNTTYPE_MPFS) ||
!strcmp(type, MNTTYPE_NFS4)) {
free(mtype);
return 1;
--- mntopt.h.orig 2007-07-16 06:25:27.000000000 -0400
+++ mntopt.h 2007-07-16 06:34:18.000000000 -0400
@@ -14,6 +14,7 @@
#define MNTTYPE_AUTOFS "autofs" /* Automount mountpoint */
#define MNTTYPE_JFS "jfs" /* JFS file system */
#define MNTTYPE_NFS4 "nfs4" /* NFSv4 filesystem */
+#define MNTTYPE_MPFS "mpfs" /* EMC Celerra MPFS filesystem - uses NFS-style RPC quotas*/
/* mount options */
#define MNTOPT_NOQUOTA "noquota" /* don't enforce quota */
Sorry about the line wraps, if that happened in the post. I could send you the patch files if you want.
I also have a patch that seems to work with 3.12 and existing RHEL4 systems, but since nfs_fstype() function, that patch is a little larger. Could I upload it somewhere for you?
Thanks,
Dave
Hi Jan,
I went ahead and posted the patch files to Tracker -> Patches.
Hope this helps.
Thanks,
Dave Boone
OK, I've added the change to quota tools.