>From Malahal :
Malahal has uploaded a new change for review.
https://review.gerrithub.io/249076
Change subject: Avoid using strmaxcpy for copying krb5_principal.realm.data
..
Avoid using strmaxcpy for copying krb5_principal.
>From Matt Benjamin :
Matt Benjamin has uploaded a new change for review.
https://review.gerrithub.io/249078
Change subject: CEPH: fallback to ceph_ll_lookup_inode if ceph_ll_get_inode
fails
..
CEPH: fallback to ceph_ll_look
>From Matt Benjamin :
Matt Benjamin has uploaded a new change for review.
https://review.gerrithub.io/249077
Change subject: ceph: actually use CEPH_PREFIX
..
ceph: actually use CEPH_PREFIX
Look for Ceph headers and librarie
>From Frank Filz :
Frank Filz has uploaded a new change for review.
https://review.gerrithub.io/249057
Change subject: Bring Protocols/XDR/nfs4.x into line with RFC 7531
..
Bring Protocols/XDR/nfs4.x into line with RFC 7531
>From Frank Filz :
Frank Filz has uploaded a new change for review.
https://review.gerrithub.io/249058
Change subject: Fix minor spelling errors in license text
..
Fix minor spelling errors in license text
Change-Id: I54ae64
>From Frank Filz :
Frank Filz has uploaded a new change for review.
https://review.gerrithub.io/249056
Change subject: Remove obsolete tools/maketest
..
Remove obsolete tools/maketest
Change-Id: Ia50982b64897d94dfcf68cff0e5c
>From Daniel Gryniewicz :
Daniel Gryniewicz has uploaded a new change for review.
https://review.gerrithub.io/249045
Change subject: Fix RDMA build
..
Fix RDMA build
Add a FindRDMA cmake module and use it to properly link to
RDMA building should be fixed in head of ntirpc. I've sent a review
request to fix RDMA in ganesha. It also allows overriding the ntirpc
tag on the command line, so those few doing ntirpc development can do
it that way if they want.
Dan
On Mon, Oct 5, 2015 at 11:15 AM, William Allen Simpson
wr
Nilesh Chate [chatenil...@gmail.com] wrote:
>Hi,
>I want to create a scenario where :
>1. Server will export a path (say /home/xyz).
>2. Client mounts this path on his machine (say /mnt/ganesha_mounted).
>3. Client writes on the mounted path (say #touch
>/mnt/ganesha_mounted
Hmm, that’s going to be very tricky. You need to handle the condition where the
client is writing to an arbitrary file handle (because the server rebooted or
the server’s namespace cache has flushed since the client did a LOOKUP). How
will you translate an arbitrary file handle to a replica file
>From Frederic Saunier :
Frederic Saunier has uploaded a new change for review.
https://review.gerrithub.io/249033
Change subject: Add include and link cmake hints for libraries in non-standard
locations
..
Add include and l
On 10/5/15 8:30 AM, Daniel Gryniewicz wrote:
> Nothing is overwritten in your source tree, even if you build ganesha
> in your source tree. The ntirpc checkout happens in the *build* tree,
> in a subdir specific to building ntirpc.
>
OK. I was going by my log of Oct 01, 10:05 am, where my
/home/b
My bad! I scrolled too quickly through the wiki page and missed this section.
Thanks for redirecting me.
The fix is now submitted through the standard flow.
Sorry for the churn.
-- Frederic
-Original Message-
From: Dominique Martinet [mailto:dominique.marti...@cea.fr]
Sent: Monday, O
Hi,
SAUNIER, FREDERIC wrote on Mon, Oct 05, 2015 at 01:45:53PM +:
> I recently got compilation issues where build was no longer able to find out
> the locations of
> ibverbs/rdmacm/mooshika/glusterfs/ceph libraries, though the related
> _PREFIX
> were passed to cmake.
>
> I'm submitting
Hi,
I recently got compilation issues where build was no longer able to find out
the locations of
ibverbs/rdmacm/mooshika/glusterfs/ceph libraries, though the related _PREFIX
were passed to cmake.
I'm submitting this patch (https://review.gerrithub.io/249025) that allows to
compile with all
Hi,
This week, coverity analysis reports 2 eliminated defects and 2 new defects.
All of them are available with online Coverity results.
Cheers,
-- Frederic
--
___
Nfs-ganes
Nothing is overwritten in your source tree, even if you build ganesha
in your source tree. The ntirpc checkout happens in the *build* tree,
in a subdir specific to building ntirpc.
In addition, a *specific* commit hash is checked out, not the head of any tree.
Dan
On Sun, Oct 4, 2015 at 7:32 PM
Hi,
I want to create a scenario where :
1. Server will export a path (say /home/xyz).
2. Client mounts this path on his machine (say /mnt/ganesha_mounted).
3. Client writes on the mounted path (say #touch
/mnt/ganesha_mounted/test.txt).
4. By default test.txt will be written to /home/xyz on server
18 matches
Mail list logo
