Hi,
On 20/11/2015 02:13, Yonghua Peng wrote:
> I have been using firefly release. is there an official documentation for
> upgrading? thanks.
Here it is : http://docs.ceph.com/docs/firefly/install/upgrading-ceph/
Enjoy !
>
>
> On 2015/11/20 6:08, Sage Weil wrote:
>> This is a bugfix release
Flushing a GPT partition table using dd does not work as the table is
duplicated at the end of the disk as well
Use the sgdisk –Z command
Paul
From: ceph-users
mailto:ceph-users-boun...@lists.ceph.com>>
on behalf of Mykola mailto:mykola.dvor...@gmail.com>>
Date: Thursday, 19 November 2015 at
Yes, It seems the problem was caused by cloning image without
deep-flatten feature.
在 2015年11月19日 23:13, Jason Dillaman 写道:
Does child image "images/0a38b10d-2184-40fc-82b8-8bbd459d62d2" have snapshots?
___
ceph-users mailing list
ceph-users@lists.
Paul, thanks for the reply, I try to run the command and then run again the
ceph-deploy command again with the same parameters, but I'm getting the
same error msg. I try with XFS and run ok, so the problem is with the btrfs
*German*
2015-11-20 6:08 GMT-03:00 HEWLETT, Paul (Paul) <
paul.hewl...@a
Hi,
I recently started a test to upgrade ceph from 0.94.5 to 9.2.0 on
Centos7. I had some issues not mentioned in the release notes. Hereby
some notes:
* Upgrading instructions are only in the release notes, not updated on
the upgrade page in the docs:
http://docs.ceph.com/docs/master/insta
Hi all, I've finished the install of a new ceph cluster with infernalis
9.2.0 release. But I'm getting the following error msg:
$ ceph -w
cluster 29xx-3xxx-xxx9-xxx7-b8xx
health HEALTH_WARN
64 pgs degraded
64 pgs stale
64 pgs stuck degraded
This usually means your crush mapping for the pool in question is
unsatisfiable. Check what the rule is doing.
-Greg
On Friday, November 20, 2015, German Anders wrote:
> Hi all, I've finished the install of a new ceph cluster with infernalis
> 9.2.0 release. But I'm getting the following error m
Hi,
We've experienced a very weird problem last week with our Ceph
cluster. We would like to ask your opinion(s) and advice
Our dedicated Ceph OSD nodes run with:
Total platform
- IO Average: 2500 wrps, ~ 600 rps
- Replica's: 3x
2 pools:
- SSD (~50x1TB)
- Spinner (~36x2TB)
- 1024 PG's per pool
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
We are seeing some of these issues as well, here is some things we have learned.
We found in our testing that enabling DISCARD as a mount option on the
SSD OSDs did not measurably affect performance (be sure to test on
your SSDs to be sure though).
Here's the actual crush map:
$ cat /home/ceph/actual_map.out
# begin crush map
tunable choose_local_tries 0
tunable choose_local_fallback_tries 0
tunable choose_total_tries 50
tunable chooseleaf_descend_once 1
tunable straw_calc_version 1
# devices
device 0 osd.0
device 1 osd.1
device 2 osd.2
de
I had a similar issue when upgrading. When I originally created by
journal partitions, I never set the partition type GUID to the Ceph
Journal GUID
(https://en.wikipedia.org/wiki/GUID_Partition_Table#Partition_type_GUIDs).
Instead it was set as "basic data partition". Pre-Infernalis this wasn't
a p
On journal device permissions see my reply in "Can't activate osd in
infernalis". Basically, if you set the partition type GUID to
45B0969E-9B03-4F30-B4C6-B4B80CEFF106 (the Ceph journal type GUID), the
existing Ceph udev rules will set permissions on the partitions
correctly at boot.
Changing the
I am developing a python application (using rbd.py) that requires querying
information about tens of thousands of rbd images. I have noticed that the
number of threads in my process grow linearly with each Image object that
is created.
After creating about 800 Image objects (that all share a sing
Hi Guys,
Now we have a very small cluster with 3 OSDs but using 40Gb NIC. We use
ceph-fuse as cephfs client and enable readahead,
but testing single reading a large file from cephfs via fio, dd or cp can only
achieve ~70+MB/s, even if fio or dd's block size is set to 1MB or 4MB.
From the cep
What's your ceph version?
Do you enable rbd cache? By default, each Image should only have one
extra thread(maybe we also should obsolete this?).
On Sat, Nov 21, 2015 at 9:26 AM, Allen Liao wrote:
> I am developing a python application (using rbd.py) that requires querying
> information about te
I'm on 0.94.5.
No, rbd cache is not enabled. Even if each Image creates only one extra
thread, if I have tens of thousands of Image objects open, there will be
tens of thousands of threads in my process. Practically speaking, am I not
allowed to cache Image objects?
On Fri, Nov 20, 2015 at 8:24
16 matches
Mail list logo