On 17/06/2015 18:06, Robert LeBlanc wrote:
Well, I think this has gone well past my ability to implement. Should
this be turned into a BP and see if someone is able to work on it?
Sorry, didn't meant to hijack your thread :-)
It might still be useful to discuss the simpler case of tracking
I am all for it!
Thx
YuriW
- Original Message -
From: Sage Weil s...@newdream.net
To: Yuri Weinstein ywein...@redhat.com
Cc: se...@ceph.com, ceph-devel@vger.kernel.org, Loic Dachary
ldach...@redhat.com, Xinxin Shu xinxin@intel.com, Alfredo Deza
ad...@redhat.com
Sent: Wednesday,
firefly v0.80.10 is ready for publishing (Sage, Alfredo, Loic, Xinxin FYI)
All results details were summarized in http://tracker.ceph.com/issues/11090
Note:
- upgrade/dumpling-firefly-x (to hammer)(distros) - runs out of memory on vps
and unreliable
- #11957 fixed (thanks Ilya, Zack!) and
We've since merged something
that stripes over several small xattrs so that we can keep things inline,
but it hasn't been backported to hammer yet. See
c6cdb4081e366f471b372102905a1192910ab2da.
Hi Sage:
You wrote yet - should we earmark it for hammer backport?
Nathan
--
To unsubscribe
On Thu, Jun 11, 2015 at 7:34 PM, Sage Weil sw...@redhat.com wrote:
* ceph-objectstore-tool should be in the ceph server package (#11376, Ken
Dreyer)
We had a little trouble yum updating from 0.94.1 to 0.94.2:
file /usr/bin/ceph-objectstore-tool from install of
ceph-1:0.94.2-0.el6.x86_64
On Wed, Jun 17, 2015 at 6:04 AM, juncheng bai
baijunch...@unitedstack.com wrote:
Hi.
Yeah, you are right, use the default max_segments, the request size can
be the object size, because the bi_phys_segments of bio could be recount,
there's just a possibility.
I want to fully understand the
On Wed, Jun 17, 2015 at 12:36 PM, juncheng bai
baijunch...@unitedstack.com wrote:
Hi.
For cloned image, it is stateless about the object. when the write
offset is smaller than the parent_ovelap,it need to call
rbd_img_obj_exists_submit to check the object status every time.
I want to save
On 2015/6/17 16:24, Ilya Dryomov wrote:
On Wed, Jun 17, 2015 at 6:04 AM, juncheng bai
baijunch...@unitedstack.com wrote:
Hi.
Yeah, you are right, use the default max_segments, the request size can
be the object size, because the bi_phys_segments of bio could be recount,
there's just a
On Wed, Jun 17, 2015 at 1:02 PM, Nathan Cutler ncut...@suse.cz wrote:
We've since merged something
that stripes over several small xattrs so that we can keep things inline,
but it hasn't been backported to hammer yet. See
c6cdb4081e366f471b372102905a1192910ab2da.
Hi Sage:
You wrote yet -
Hi.
For cloned image, it is stateless about the object. when the write
offset is smaller than the parent_ovelap,it need to call
rbd_img_obj_exists_submit to check the object status every time.
I want to save the status of the object which it is within the scope
of parent_overlap.
The rbd
On Wed, Jun 17, 2015 at 4:28 PM, Douglas Fuller dful...@redhat.com wrote:
On Jun 16, 2015, at 7:18 PM, Josh Durgin jdur...@redhat.com wrote:
On 06/12/2015 08:56 AM, Douglas Fuller wrote:
@@ -3132,6 +3132,26 @@ static void rbd_watch_cb(u64 ver, u64 notify_id, u8
opcode, s32 return_code,
On Jun 16, 2015, at 7:18 PM, Josh Durgin jdur...@redhat.com wrote:
On 06/12/2015 08:56 AM, Douglas Fuller wrote:
@@ -3132,6 +3132,26 @@ static void rbd_watch_cb(u64 ver, u64 notify_id, u8
opcode, s32 return_code,
rbd_warn(rbd_dev, notify_ack ret %d, ret);
}
+static
From: Mike Christie micha...@cs.wisc.edu
This syncs the ceph_osd_op struct with the current version of ceph
where the watch struct has been updated to support more ops and
the notify-ack support has been broken out of the watch struct.
Ceph commits
1a82cc3926fc7bc4cfbdd2fd4dfee8660d5107a1
Send CEPH_OSD_WATCH_OP_PING every osd_keepalive_timeout for each watch
event registered. When errors are detected, look up the watch event and
send it CEPH_WATCH_EVENT_DISCONNECTED.
Signed-off-by: Douglas Fuller dful...@redhat.com
---
include/linux/ceph/osd_client.h | 1 +
net/ceph/osd_client.c
Add support in notify events for receiving data from notify_ack. Notify
events are optional; data is discarded if no event is found.
Signed-off-by: Douglas Fuller dful...@redhat.com
---
include/linux/ceph/osd_client.h | 3 +-
net/ceph/osd_client.c | 135
osd_client, rbd: add support for version 2 of watch/notify
Support watch/notify 2 in osd_client and update rbd for compatibility.
Map ceph_osd_event to expected watch/notify messages and store watch
and watch error callbacks. Implement CEPH_OSD_WATCH_OP_PING and
CEPH_OSD_WATCH_OP_RECONNECT.
8AM PST as usual! Discussion topics for this week include:
- Alexandre's tcmalloc / memory pressure QEMU performance tests.
- Continuation of SimpleMessenger fastpath discussion?
Please feel free to add your own!
Here's the links:
Etherpad URL:
http://pad.ceph.com/p/performance_weekly
To
Change unused ceph_osd_event structure to refer to pending watch/notify2
messages. Watch events include the separate watch and watch error callbacks
used for watch/notify2. Update rbd to use separate watch and watch error
callbacks via the new watch event.
Signed-off-by: Douglas Fuller
From: Mike Christie micha...@cs.wisc.edu
This patch adds support for proto version 1 of watch-notify,
so drivers like rbd can be sent a buffer with information like
the notify operation being performed.
Signed-off-by: Mike Christie micha...@cs.wisc.edu
Reviewed-by: Josh Durgin jdur...@redhat.com
From: Mike Christie micha...@cs.wisc.edu
This adds support watch-notify header 2 and 3 support, so we can
get a return_code from those operations.
Signed-off-by: Mike Christie micha...@cs.wisc.edu
[djf: fixed decoding bits]
Signed-off-by: Douglas Fuller dful...@redhat.com
Reviewed-by: Josh
On Wed, 17 Jun 2015, Nathan Cutler wrote:
We've since merged something
that stripes over several small xattrs so that we can keep things inline,
but it hasn't been backported to hammer yet. See
c6cdb4081e366f471b372102905a1192910ab2da.
Hi Sage:
You wrote yet - should we earmark it
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
Well, I think this has gone well past my ability to implement. Should
this be turned into a BP and see if someone is able to work on it?
-
Robert LeBlanc
GPG Fingerprint 79A2 9CA4 6CC4 45DD A904 C70E E654 3BB2 FA62 B9F1
On Tue,
On Jun 17, 2015, at 12:11 PM, Mike Christie mchri...@redhat.com wrote:
On 06/17/2015 09:25 AM, Douglas Fuller wrote:
@@ -2486,6 +2579,7 @@ static void __do_event(struct ceph_osd_client *osdc,
u8 opcode,
case CEPH_WATCH_EVENT_NOTIFY_COMPLETE:
if (event) {
On 06/17/2015 09:25 AM, Douglas Fuller wrote:
@@ -2486,6 +2579,7 @@ static void __do_event(struct ceph_osd_client *osdc, u8
opcode,
case CEPH_WATCH_EVENT_NOTIFY_COMPLETE:
if (event) {
event-notify.notify_data = data;
+
24 matches
Mail list logo