We're newbie to Ceph. Besides using incremental snapshots with RDB to
backup data on one Ceph cluster to another running Ceph cluster, or
using backup tools like backy2, will there be any recommended way to
backup Ceph data ? Someone here suggested taking snapshot of RDB
daily and keeps 30
Hi, I read through the various documentation and had a few questions:
- From what I understand cephFS clients reach the OSDs directly, does the
cluster network need to be opened up as a public network?
All RADOS clients connect to osd's and mon's directly.
- Is it still necessary to have a
Has anyone experienced below?
2 of OSD server was down, after bring up 2 of servers, I brought 52 OSD's
in with just weight of 0.05, but it causing huge backfilling load, I saw so
many blocked requests and a number of pg stuck inactive. some of servers
was impact. so I stopped backfilling by mark
Without knowing more about the underlying hardware, you likely are reaching
some type of IO resource constraint. Are your journals colocated or
non-colocated? How fast is your backend OSD storage device?
You may also want to look at setting the norebalance flag.
Good luck!
> On Sep 20,
Hi,
>> Will the RAID 6 be mirrored to another storage in remote site for DR purpose?
>
>Not yet. Our goal is to have the backup ceph to which we will replicate spread
>across three different buildings, with 3 replicas.
May I ask if the backup ceph is a single ceph cluster span across 3
Setup a python virtual environment and install the required notario package
version. You'll want to also install ansible into that virtual environment
along with netaddr.
On Sep 20, 2018, at 18:04, solarflow99
mailto:solarflo...@gmail.com>> wrote:
oh, was that all it was... git clone
Hi there,
Has anyone experienced below?
2 of OSD server was down, after bring up 2 of servers, I brought 52 OSD's
in with just weight of 0.05, but it causing huge backfilling load, I saw so
many blocked requests and a number of pg stuck inactive. some of servers
was impact. so I stopped
Hello,
on a Luminous cluster, I have a PG incomplete and I can't find how to
fix that.
It's an EC pool (4+2) :
pg 37.9c is incomplete, acting [32,50,59,1,0,75] (reducing pool
bkp-sb-raid6 min_size from 4 may help; search ceph.com/docs for
'incomplete')
Of course, we can't reduce min_size
oh, was that all it was... git clone https://github.com/ceph/ceph-ansible/
I installed the notario package from EPEL,
python2-notario-0.0.11-2.el7.noarch and thats the newest they have
On Thu, Sep 20, 2018 at 3:57 PM Alfredo Deza wrote:
> Not sure how you installed ceph-ansible, the
Not sure how you installed ceph-ansible, the requirements mention a
version of a dependency (the notario module) which needs to be 0.0.13
or newer, and you seem to be using an older one.
On Thu, Sep 20, 2018 at 6:53 PM solarflow99 wrote:
>
> Hi, tying to get this to do a simple deployment, and
Hi, tying to get this to do a simple deployment, and i'm getting a strange
error, has anyone seen this? I'm using Centos 7, rel 5 ansible 2.5.3
python version = 2.7.5
I've tried with mimic luninous and even jewel, no luck at all.
TASK [ceph-validate : validate provided configuration]
I've done some more experiments playing with client config parameters,
and it seems like the the client_oc_size parameter is very correlated to
how big ceph-fuse grows. With its default value of 200MB, ceph-fuse
gets to about 22GB of RSS, with our previous client_oc_size value of
2GB, the
Hi Nitin,
I'm still receiving slides from the speakers but I think I will start posting
them tomorrow. I will reply back when this is done. Thanks!
--
Mike Perez (thingee)
On 16:44 Sep 20, Kamble, Nitin A wrote:
> Hi Mike,
> Are the slides of presentations available anywhere? If not, can
did you read my answer?
On Thu, Sep 20, 2018 at 8:21 PM Mike Cave wrote:
>
> I'll bump this one more time in case someone who knows why this is happening
> didn't see the thread yesterday.
>
> Cheers,
> Mike
>
> -Original Message-
> From: ceph-users on behalf of Cave Mike
>
> Date:
On Thu, Sep 20, 2018 at 3:46 AM Marc Roos wrote:
>
>
> When running ./do_cmake.sh, I get
>
> fatal: destination path '/Users/mac/ceph/src/zstd' already exists and is
> not an empty directory.
> fatal: clone of 'https://github.com/facebook/zstd' into submodule path
> '/Users/mac/ceph/src/zstd'
I'll bump this one more time in case someone who knows why this is happening
didn't see the thread yesterday.
Cheers,
Mike
-Original Message-
From: ceph-users on behalf of Cave Mike
Date: Wednesday, September 19, 2018 at 9:25 AM
Cc: ceph-users
Subject: Re: [ceph-users] total_used
Hi Mike,
Are the slides of presentations available anywhere? If not, can those be
shared?
Thanks,
Nitin
On 9/11/18, 3:47 PM, "ceph-users on behalf of Mike Perez"
wrote:
[External Email]
Hey all,
Just a reminder that Ceph Day at
On Thu, 20 Sep 2018, KEVIN MICHAEL HRPCEK wrote:
> Top results when both were taken with ms_dispatch at 100%. The mon one
> changes alot so I've included 3 snapshots of those. I'll update
> mon_osd_cache_size.
>
> After disabling auth_cluster_required and a cluster reboot I am having
> less
Top results when both were taken with ms_dispatch at 100%. The mon one changes
alot so I've included 3 snapshots of those. I'll update mon_osd_cache_size.
After disabling auth_cluster_required and a cluster reboot I am having less
problems keeping OSDs in the cluster since they seem to not be
Hi all,
after increasing mon_max_pg_per_osd number ceph starts rebalancing as usual.
However, the slow requests warnings are still there, even after setting
primary-affinity to 0 beforehand.
By the other hand, if I destroy the osd, ceph will start rebalancing unless
noout flag is set, am I
The mons have a 300gb raid 1 on 10k sas. The /var lv is 44% full with the
/var/lib/ceph/mon directory at 6.7gb. When ms_dispatch is running 100% it is
all user time with iostat showing 0-2% utilization of the drive. I'm
considering taking one the mon's raid 1 drives and dropping them into a
You can prevent creation of the PGs on the old filestore OSDs (which
seems to be the culprit here) during replacement by replacing the
disks the hard way:
* ceph osd destroy osd.X
* re-create with bluestore under the same id (ceph volume ... --osd-id X)
it will then just backfill onto the same
Thank you very much Paul.
Kevin
Am Do., 20. Sep. 2018 um 15:19 Uhr schrieb Paul Emmerich <
paul.emmer...@croit.io>:
> Hi,
>
> device classes are internally represented as completely independent
> trees/roots; showing them in one tree is just syntactic sugar.
>
> For example, if you have a
Can you try 'perf top -g -p ' to see where all of the encoding
activity is coming from? I see two possibilities (the mon attempts to
cache encoded maps, and the MOSDMap message itself will also reencode
if/when that fails).
Also: mon_osd_cache_size = 10 by default... try making that 500 or
Hi,
device classes are internally represented as completely independent
trees/roots; showing them in one tree is just syntactic sugar.
For example, if you have a hierarchy like root --> host1, host2, host3
--> nvme/ssd/sata OSDs, then you'll actually have 3 trees:
root~ssd -> host1~ssd,
To answer my own question:
ceph osd crush tree --show-shadow
Sorry for the noise...
Am Do., 20. Sep. 2018 um 14:54 Uhr schrieb Kevin Olbrich :
> Hi!
>
> Currently I have a cluster with four hosts and 4x HDDs + 4 SSDs per host.
> I also have replication rules to distinguish between HDD and SSD
Out of curiosity, what disks do you have your mons on and how does the disk
usage, both utilization% and full%, look while this is going on?
On Wed, Sep 19, 2018, 1:57 PM Kevin Hrpcek
wrote:
> Majority of the clients are luminous with a few kraken stragglers. I
> looked at ceph features and
Hi!
Currently I have a cluster with four hosts and 4x HDDs + 4 SSDs per host.
I also have replication rules to distinguish between HDD and SSD (and
failure-domain set to rack) which are mapped to pools.
What happens if I add a heterogeneous host with 1x SSD and 1x NVMe (where
NVMe will be a new
Hi,
We have a bucket that we are trying to empty. Versioning and lifecycle was
enabled. We deleted all the objects in the bucket. But this left a whole
bunch of Delete Markers.
aws s3api delete-object --bucket B --key K --version-id V is not deleting the
delete markers.
Any ideas? We
Hi,
to reduce impact on clients during migration I would set the OSD's
primary-affinity to 0 beforehand. This should prevent the slow
requests, at least this setting has helped us a lot with problematic
OSDs.
Regards
Eugen
Zitat von Jaime Ibar :
Hi all,
we recently upgrade from
Hello,
2018-09-20 09:32:58.851160 mon.dri-ceph01 [WRN] Health check update:
249 PGs pending on creation (PENDING_CREATING_PGS)
This error might indicate that you are hitting a PG limit per osd.
Here some information on it
https://ceph.com/community/new-luminous-pg-overdose-protection/ . You
Hi all,
we recently upgrade from Jewel 10.2.10 to Luminous 12.2.7, now we're
trying to migrate the
osd's to Bluestore following this document[0], however when I mark the
osd as out,
I'm getting warnings similar to these ones
2018-09-20 09:32:46.079630 mon.dri-ceph01 [WRN] Health check
On Thu, Sep 20, 2018 at 9:42 AM fatkun chan wrote:
>
> OSD : filestore FileSystem: ext4 system: centos7.2
>
> I use https://rook.io/ deploy ceph on kubernetes
>
>
> ceph config
> --
> mon keyvaluedb = rocksdb
> mon_allow_pool_delete= true
>
When running ./do_cmake.sh, I get
fatal: destination path '/Users/mac/ceph/src/zstd' already exists and is
not an empty directory.
fatal: clone of 'https://github.com/facebook/zstd' into submodule path
'/Users/mac/ceph/src/zstd' failed
Failed to clone 'src/zstd'. Retry scheduled
fatal:
Has anyone been able to build according to this manual? Because here it
fails.
http://docs.ceph.com/docs/mimic/dev/macos/
I have prepared macos as it is described, took 2h to build this llvm, is
that really necessary?
I do the
git clone --single-branch -b mimic
In case you or anyone else reading is interested, I tried using the latest fuse
client instead of the kernel client and my problem seems to be gone.
I think our kernel is recent enough that it should include the bug fix you
mentioned? So maybe some else going on there...
Regards,
Tom
From:
OSD : filestore FileSystem: ext4 system: centos7.2
I use https://rook.io/ deploy ceph on kubernetes
ceph config
--
mon keyvaluedb = rocksdb
mon_allow_pool_delete= true
mon_max_pg_per_osd = 1000
debug default= 0
debug
I was thinking of iscsi gateways colocated on the osd nodes and trying to
distribute the luns as evenly as possible, would that setup work ? Also
regarding the configuration of the iscsi target is it stored inside ceph
cluster ?
Le jeu. 20 sept. 2018 à 08:23, Jan Fajerski a écrit :
> Hi,
> if
12.2.8 improves the deep scrub code to automatically repair these
inconsistencies. Once the entire cluster has been upgraded and then fully deep
scrubbed, and all such inconsistencies are resolved; it will be safe to disable
the `osd distrust data digest = true` workaround option.
Just for
Hi,
if you want to isolate your HV from ceph's public network a gateway would do
that (like iscsi gateway). Note however that this will also add an extra network
hop and a potential bottleneck since all client traffic has to pass through the
gateway node(s).
HTH,
Jan
On Wed, Sep 19, 2018 at
1 it is for the perfomance nginx is more faster than civetweb base on the
cosbench test.
2 I want to use some extra function with nginx. such as rtmp streaming data and
add watermask for picture and so
on. with nginx it has more free and opensource powerful plug-in units.
At
41 matches
Mail list logo