Hi!
the last days, I updated my jessie evaluation cluster to jewel and now
osds are not started automatically after reboot because they are not
mounted. This is the output of ceph-disk list after boot:
/dev/sdh :
/dev/sdh1 ceph data, prepared, cluster ceph, osd.47, journal /dev/sde1
/dev/sdi :
Add-on-Question:
What is the intended purpose of the ceph-disk@.service? I can run
systemctl start ceph-disk@/dev/sdr1
but I can't 'enable' it like the the ceph-osd@.service so why is it there?
Best regards
Karsten
2016-04-27 9:33 GMT+02:00 Karsten Heymann :
> Hi!
>
> the last days, I updated
all informations i have so far are from the FOSDEM
https://fosdem.org/2016/schedule/event/virt_iaas_ceph_rados_gateway_overview/attachments/audio/1079/export/events/attachments/virt_iaas_ceph_rados_gateway_overview/audio/1079/Fosdem_RGW.pdf
Cheers,
Ansgar
2016-04-27 2:28 GMT+02:00 :
> Hello:
>
Hi Ansgar,
Thanks for your information.
I have tried 's3fs-fuse' to mount RADOSGW buckets on Ubuntu client node. It
works.
But I am not sure this is the technique that access RADOSGW buckets via NFS
on Jewel.
Best Regards,
WD
-Original Message-
From: ceph-users [mailto:ceph-users-
On Tue, Apr 26, 2016 at 10:17:31PM -0700, Ben Hines wrote:
[...]
> --> 10.30.1.6:6800/10350 -- osd_op(client.44852756.0:79
> default.42048218. [getxattrs,stat,read 0~524288] 12.aa730416
> ack+read+known_if_redirected e100207) v6 -- ?+0 0x7f49c41880b0 con
> 0x7f49c4145eb0
> 0> 2016-04-26 22:07:
Anyone can give me some advice?
-- Forwarded message --
From:
Date: 2016-04-26 18:50 GMT+08:00
Subject: google perftools on ceph-osd
To: Stefan Priebe - Profihost AG
hi Stefan:
When We are using ceph, I found osd process use much more CPU,
especially when small rand write
Hi Karsten,
I've had the same experience updating our test cluster (Debian 8) from
Infernalis to Jewel.
I've update udev/systemd to the one in testing (so, from 215 to 229),
and it worked much better at reboot.
So... Are the udev rules written for the udev version in RedHat (219) or
greate
Here's one of my nodes that can't unmap a device:
[root@nodebr6 ~]# rbd unmap /dev/rbd0
2016-04-27 11:36:56.975668 7fcd61ae67c0 -1 did not load config file, using
default settings.
Option -q no longer supported.
rbd: run_cmd(udevadm): exited with status 1
rbd: sysfs write failed
rbd: unmap failed:
Hi Loris,
thank you for your feedback. As I plan to go productive with the
cluster later this year I'm really hesitant to update udev and systemd
to a version newer than jessie, especially as there is no official
backport for those packages yet. I really would expect ceph to work
out of the box wi
Le 27/04/2016 13:51, Karsten Heymann a écrit :
Hi Loris,
thank you for your feedback. As I plan to go productive with the
cluster later this year I'm really hesitant to update udev and systemd
to a version newer than jessie, especially as there is no official
backport for those packages yet. I
Hi cephers,
I've been following the upgrade intrucctions...but..I sure I did
something wrong.
I just upgrade using ceph-deploy on one monitor (after ofcourse down de
mon service).
Then the chow to var/lib/ceph and /var/log/ceph for ceph user
[root@cephmon03 ~]# systemctl start ceph.targe
Nick, all,
fantastic, that did it!
I installed kernel 4.5.2 on the client, now the single threaded read
performance using krbd mount is up to about 370 MB/s with standard 256
readahead size, and touching 400 MB/s with larger readahead sizes.
All single threaded.
Multi-threaded krbd read on t
Hi,
one workaround I found was to add
[Install]
WantedBy=ceph-osd.target
to /lib/systemd/system/ceph-disk@.service and then manually enable my disks with
# systemctl enable ceph-disk\@dev-sdi1
# systemctl start ceph-disk\@dev-sdi1
That way they at least are started at boot time.
Best regards,
Hi Iban,
the current jewel packages seem to be missing some important systemd
files. Try to copy
https://github.com/ceph/ceph/blob/master/systemd/ceph-mon.target to
/lib/systemd/system and enable it:
systemctl enable ceph-mon.target
I also would disable the legacy init script with
systemctl mas
Le 27/04/2016 14:45, Karsten Heymann a écrit :
Hi,
one workaround I found was to add
[Install]
WantedBy=ceph-osd.target
to /lib/systemd/system/ceph-disk@.service and then manually enable my disks with
# systemctl enable ceph-disk\@dev-sdi1
# systemctl start ceph-disk\@dev-sdi1
That way the
2016-04-27 15:18 GMT+02:00 Loris Cuoghi :
> Le 27/04/2016 14:45, Karsten Heymann a écrit :
>> one workaround I found was to add
>>
>> [Install]
>> WantedBy=ceph-osd.target
>>
>> to /lib/systemd/system/ceph-disk@.service and then manually enable my
>> disks with
>>
>> # systemctl enable ceph-disk\@d
Hi WD,
No, it's not the same. The new mechanism uses an nfs-ganesha server to export
the RGW namespace. Some upstream documentation will be forthcoming...
Regards,
Matt
- Original Message -
> From: "WD Hwang"
> To: "a jazdzewski"
> Cc: ceph-users@lists.ceph.com
> Sent: Wednesday, A
Hi Karsten,
I have checked taht files arethe same that git ones.
-rw-r--r-- 1 root root 810 Apr 20 18:45 /lib/systemd/system/ceph-mon@
.service
-rw-r--r-- 1 root root 162 Apr 20 18:45
/lib/systemd/system/ceph-mon.target
[root@cephmon03 ~]# cat /lib/systemd/system/ceph-mon.target
[Unit]
Des
Hello All,
I am currently trying to get an accurate count of bytes used for an rbd
image. I've tried trimming the filesystem which relieves about 1.7gb
however there is still a huge disparity of size reported in the filesystem
vs what 'rbd diff' shows;
$ rbd map backup/cd4e5d37-3023-4640-be5a-557
On Wed, Apr 27, 2016 at 2:07 PM, Tyler Wilson wrote:
> $ rbd diff backup/cd4e5d37-3023-4640-be5a-5577d3f9307e | awk '{ SUM += $2 }
> END { print SUM/1024/1024 " MB" }'
> 49345.4 MB
Is this a cloned image? That awk trick doesn't account for discarded
regions (i.e. when column three says "zero" in
Hello Jason,
Thanks for the quick reply, this was copied from an VM instance snapshot to
my backup pool (rbd snap create, rbd cp (to backup pool), rbd snap rm).
I've tried piping through grep per your recommendation and it still reports
the same usage
$ rbd diff backup/cd4e5d37-3023-4640-be5a-557
What kernel versions are required to be able to use CephFS thru mount -t ceph?
-- Tom Deneau
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
On Wed, Apr 27, 2016 at 2:55 PM, Deneau, Tom wrote:
> What kernel versions are required to be able to use CephFS thru mount -t ceph?
The CephFS kernel client has been in for ages (2.6.34, I think?), but
you want the absolute latest you can make happen if you're going to
try it out.
The actual mou
I was using SLES 12, SP1 which has 3.12.49
It did have a /usr/sbin/mount.ceph command but using it gave
modprobe: FATAL: Module ceph not found.
failed to load ceph kernel module (1)
-- Tom
> -Original Message-
> From: Gregory Farnum [mailto:gfar...@redhat.com]
> Sent: Wednesday, A
On Wed, Apr 27, 2016 at 3:17 PM, Deneau, Tom wrote:
> I was using SLES 12, SP1 which has 3.12.49
>
> It did have a /usr/sbin/mount.ceph command but using it gave
> modprobe: FATAL: Module ceph not found.
> failed to load ceph kernel module (1)
So that's about what is in your distro kernel, no
The image size (50G) minus the fstrim size (1.7G) approximately equals
the actual usage (48.19G). Therefore, I guess the question is why
doesn't fstrim think it can discard more space?
On a semi-related note, we should probably improve the rbd copy
sparsify logic. Right now it requires the full
Hi,
I have a small dev/test ceph cluster that sat neglected for quite some
time. It was on the firefly release until very recently. I successfully
upgraded from firefly to hammer without issue as an intermediate step to
get to the latest jewel release.
This cluster has 3 ubuntu 14.04 hosts with k
Hi Tom,
On Wed, 27 Apr 2016 20:17:51 +, Deneau, Tom wrote:
> I was using SLES 12, SP1 which has 3.12.49
>
> It did have a /usr/sbin/mount.ceph command but using it gave
> modprobe: FATAL: Module ceph not found.
> failed to load ceph kernel module (1)
The SLES 12 SP1 kernel doesn't cur
Hi,
My current setup is running on 12 OSD's split between 3 hosts. We're using
this for VM's (Proxmox) and nothing else.
According to:
http://docs.ceph.com/docs/master/rados/operations/placement-groups/ - my pg_num
should be set to 4096
If I use the calculator, and put in Size 3, OSD 12, and
Hello Jason,
Yes, I believe that is my question. Is there any way I can either reclaim
the space for this disk?
On Wed, Apr 27, 2016 at 1:25 PM, Jason Dillaman wrote:
> The image size (50G) minus the fstrim size (1.7G) approximately equals
> the actual usage (48.19G). Therefore, I guess the qu
Hello,
On Wed, 27 Apr 2016 22:55:35 + Carlos M. Perez wrote:
> Hi,
>
> My current setup is running on 12 OSD's split between 3 hosts. We're
> using this for VM's (Proxmox) and nothing else.
>
I assume evenly split (4 OSDs per host)?
> According to:
> http://docs.ceph.com/docs/master/rado
Hi List,
Ceph 10.2.0 errors out during compilation when compiling without radowgw
support.
./configure --prefix=/usr --build=i686-pc-linux-gnu
--host=i686-pc-linux-gnu --mandir=/usr/share/man
--infodir=/usr/share/info --datadir=/usr/share --sysconfdir=/etc
--localstatedir=/var/lib --disabl
- Original Message -
> From: "Karol Mroz"
> To: "Ben Hines"
> Cc: "ceph-users"
> Sent: Wednesday, 27 April, 2016 7:06:56 PM
> Subject: Re: [ceph-users] radosgw crash - Infernalis
>
> On Tue, Apr 26, 2016 at 10:17:31PM -0700, Ben Hines wrote:
> [...]
> > --> 10.30.1.6:6800/10350 -- osd_o
hi sage:
I find the slides of VAULT of 2016 on this page(
http://events.linuxfoundation.org/events/vault/program/slides), it seems
not the whole accoding to the schedule info, and I didn't find yours. Can
you share your slides or any things usefull on VAULT about BlueStore.
regards~
zhiyo
Yes, CentOS 7.2. Happened twice in a row, both times shortly after a
restart, so i expect i'll be able to reproduce it. However, i've now tried
a bunch of times and it's not happening again.
In any case i have glibc + ceph-debuginfo installed so we can get more info
if it does happen.
thanks!
On
Got it again - however, the stack is exactly the same, no symbols -
debuginfo didn't resolve. Do i need to do something to enable that?
The server in 'debug ms=10' this time, so there is a bit more spew:
-14> 2016-04-27 21:59:58.811919 7f9e817fa700 1 -- 10.30.1.8:0/3291985349
--> 10.30.2.13:6
- Original Message -
> From: "Ben Hines"
> To: "Brad Hubbard"
> Cc: "Karol Mroz" , "ceph-users"
> Sent: Thursday, 28 April, 2016 3:09:16 PM
> Subject: Re: [ceph-users] radosgw crash - Infernalis
> Got it again - however, the stack is exactly the same, no symbols - debuginfo
> didn't r
Aha, i see how to use the debuginfo - trying it by running through gdb.
On Wed, Apr 27, 2016 at 10:09 PM, Ben Hines wrote:
> Got it again - however, the stack is exactly the same, no symbols -
> debuginfo didn't resolve. Do i need to do something to enable that?
>
> The server in 'debug ms=10'
Hi,
they are missing target files in debian packages
http://tracker.ceph.com/issues/15573
https://github.com/ceph/ceph/pull/8700
I have also done some other trackers about packaging bug
jewel: debian package: wrong /etc/default/ceph/ceph location
http://tracker.ceph.com/issues/15587
debian/ubu
Hi,
we recently stumbled over a problem with the kernel based CephFS driver
(Ubuntu Trusty with 4.4.0-18 kernel from xenial lts backport package).
Our MDS failed for some unknown reason, and the standby MDS became active.
After rejoining the MDS cluster, the former standby MDS stuck at the
c
Just got into a discussion today where I may have a chance to do work
with a db guy who wants hadoop and I want to steer him to it on cephfs.
While I'd really like to run gentoo with either infernalis or jewel
(when it becomes stable in portage), odds are more likely that I will be
required to
41 matches
Mail list logo