[Starlingx-discuss] ceph objects

Chen, Haochuan Z haochuan.z.chen at intel.com
Mon Nov 25 01:23:20 UTC 2019


Hi Von

This is incorrect ceph conf setting for openstack.

http://tkannadhasan.blogspot.com/2018/04/ceph-pool-rbd-has-many-more-objects-per.html
https://bugs.launchpad.net/charm-ceph-mon/+bug/1804846

1, you can google to find ceph command to update this config item, "mon pg warn max object skew"
2, Or submit a launchpad issue, maybe we should update ceph config item,  to change puppet setting in starlingx for ceph.

BR!

Martin, Chen
SSP, Software Engineer
021-61164330

-----Original Message-----
From: starlingx-discuss-request at lists.starlingx.io <starlingx-discuss-request at lists.starlingx.io> 
Sent: Friday, November 22, 2019 8:30 PM
To: starlingx-discuss at lists.starlingx.io
Subject: Starlingx-discuss Digest, Vol 18, Issue 139

Send Starlingx-discuss mailing list submissions to
	starlingx-discuss at lists.starlingx.io

To subscribe or unsubscribe via the World Wide Web, visit
	http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss
or, via email, send a message with subject or body 'help' to
	starlingx-discuss-request at lists.starlingx.io

You can reach the person managing the list at
	starlingx-discuss-owner at lists.starlingx.io

When replying, please edit your Subject line so it is more specific than "Re: Contents of Starlingx-discuss digest..."


Today's Topics:

   1. StarlingX VM size getting very large (Saichandu Behara)
   2. ceph objects (von Hoesslin, Volker)


----------------------------------------------------------------------

Message: 1
Date: Fri, 22 Nov 2019 12:14:42 +0530
From: Saichandu Behara <saichandu.behara at calsoftinc.com>
To: "starlingx-discuss at lists.starlingx.io"
	<starlingx-discuss at lists.starlingx.io>
Cc: Pavan Gupta <pavan.gupta at calsoftinc.com>
Subject: [Starlingx-discuss] StarlingX VM size getting very large
Message-ID: <8c346c40-a490-0fa9-c5e8-5fd2cb0f1f89 at calsoftinc.com>
Content-Type: text/plain; charset=utf-8; format=flowed

Hello all,

We are installed Openstack on StarlingX ( It's containing EdgeX, k8s and Openstack ). After that, The VM size incresed to 145 GB. We converted to
Qcow2 images and the size is 45 GB and after that, compression also, Its getting around 13 GB. So, upload and transferring is bit of tough. So, Is there any way to reduce the VM size / performing some cleanup operations. So that, The VM size is decreased something.

If any one aware on this issue, Please let us know. It's a great help.


--
Thanks & Regards
Sai Chandu Behara




------------------------------

Message: 2
Date: Fri, 22 Nov 2019 09:16:43 +0000
From: "von Hoesslin, Volker" <Volker.Hoesslin at swsn.de>
To: "starlingx-discuss at lists.starlingx.io"
	<starlingx-discuss at lists.starlingx.io>
Subject: [Starlingx-discuss] ceph objects
Message-ID: <DB39DAB5F762704A93D316F9CE7D021E42929509 at emilio.sis.net>
Content-Type: text/plain; charset="iso-8859-1"

hi,

my ceph have an health warning:

[sysadmin at controller-0 ~(keystone_admin)]$ ceph -s
  cluster:
    id:     2d6cb518-420e-4a97-919a-f42b808f2049
    health: HEALTH_WARN
            2 pools have many more objects per pg than average

  services:
    mon: 3 daemons, quorum controller-0,controller-1,storage-0
    mgr: controller-0(active), standbys: controller-1
    osd: 30 osds: 30 up, 30 in

  data:
    pools:   5 pools, 600 pgs
    objects: 97.12 k objects, 596 GiB
    usage:   1.7 TiB used, 107 TiB / 109 TiB avail
    pgs:     600 active+clean

  io:
    client:   25 MiB/s rd, 36 MiB/s wr, 169 op/s rd, 104 op/s wr


after checkout "ceph df", i can see an huge number of objects:


[sysadmin at controller-0 ~(keystone_admin)]$ ceph df
GLOBAL:
    SIZE        AVAIL       RAW USED     %RAW USED
    109 TiB     107 TiB      1.7 TiB          1.58
POOLS:
    NAME               ID     USED        %USED     MAX AVAIL     OBJECTS
    kube-rbd           1      1.5 GiB         0        33 TiB         451
    images             2      437 GiB      1.28        33 TiB       55950
    cinder.backups     3          0 B         0        33 TiB           0
    cinder-volumes     4      151 GiB      0.45        33 TiB       39780
    ephemeral          5          0 B         0        33 TiB           0


is this a normal value for round about 12 images and 8 values ? maybe there is some thrash inside this pools, how can i run an cleanup?

greez & thx,
volker.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.starlingx.io/pipermail/starlingx-discuss/attachments/20191122/c18d6c6c/attachment-0001.html>

------------------------------

Subject: Digest Footer

_______________________________________________
Starlingx-discuss mailing list
Starlingx-discuss at lists.starlingx.io
http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss


------------------------------

End of Starlingx-discuss Digest, Vol 18, Issue 139
**************************************************


More information about the Starlingx-discuss mailing list