[ceph-users] Re: Delete objects on large bucket very slow

Anyone help me explain: How does ceph delete an object? how flow of operator detete work? - Br, Dương Tuấn Dũng 0986153686 On Fri, Sep 13, 2019 at 10:51 AM Konstantin Shalygin wrote: > On 9/13/19 10:46 AM, tuan dung wrote: > > thank for your answer. Can you explain c

[ceph-users] Re: Delete objects on large bucket very slow

On 9/13/19 10:46 AM, tuan dung wrote: thank for your answer. Can you explain clearly for me about: - How does ceph's deleting object operator work? (it mean about how delete object flow in ceph works?) - using s3cmd tool which install in client to delete object, how to improve speed of delete o

[ceph-users] Re: Delete objects on large bucket very slow

Hi Paul, thank for your answer. Can you explain clearly for me about: - How does ceph's deleting object operator work? (it mean about how delete object flow in ceph works?) - using s3cmd tool which install in client to delete object, how to improve speed of delete object operator (objects/s) with i

[ceph-users] Legacy Bluestore Stats

I created a cluster on Nautilus 14.2.0, then upgraded to 14.2.1, and finally 14.2.3 Now I am seeing this warning that I thought should only appear if the cluster was created pre-Nautilus. Legacy BlueStore stats reporting detected on XX OSD I can't seem to find any information about this on a c

[ceph-users] Re: ceph fs crashes on simple fio test

On Tue, Sep 10, 2019 at 1:11 PM Frank Schilder wrote: > Hi Robert, > > I have meta data on SSD (3xrep) and data on 8+2 EC on spinning disks, so > the speed difference is orders of magnitudes. Our usage is quite meta data > heavy, so this suits us well. In particular since EC pools are high > thro

[ceph-users] Re: 645% Clean PG's in Dashboard

Hi, On 8/30/19 10:54 AM, c.li...@falseprivacy.org wrote: > I also enabled the Dashboard and saw that the PG Status showed "645% > clean" PG's. This cluster was originally installed with version > Jewel, so may it be any legacy setting or such that causing this? No, this seems to be a genuine bug

[ceph-users] Re: Bug identified: Dashboard proxy configuration is not working as expected

Hi Thomas, On 9/12/19 10:21 AM, Thomas wrote: > I have successfully configured Ceph dashboard following the this > documentation. > > According to the documentation you can configure a URL prefix with this > command: > ceph config set mgr mgr/das

[ceph-users] Re: the different between flag system and admin when create user rgw

Admin users bypass most authorization checks, so are useful for maintenance on buckets/objects owned by other users. They also have full read/write capabilities to the /admin/ apis. System users are for requests made on behalf of radosgw itself. Multisite replication requires a system user in

[ceph-users] Re: Delete objects on large bucket very slow

--max-concurrent-ios helps with deletion speed, there's also a memory leak during deletion which is being fixed here: https://github.com/ceph/ceph/pull/30174 Paul -- Paul Emmerich Looking for help with your Ceph cluster? Contact us at https://croit.io croit GmbH Freseniusstr. 31h 81247 München

[ceph-users] How to use radosgw-min find ?

Hi! We're looking to mantain our rgw pools out of orphans objects, checking the documentation and mailist is not really clear how it works and what will do. Radosgw-admin orphands find -pool= --job-id= Loops over all objects in the cluster looking for leaked objects and add it to a shar

[ceph-users] CentOS deps for ceph-mgr-diskprediction-local

Hi there, Did anyone get the mgr diskprediction-local plugin working on CentOS ? When I enable the plugin with v14.2.3 I get: HEALTH_ERR 2 mgr modules have failed MGR_MODULE_ERROR 2 mgr modules have failed Module 'devicehealth' has failed: Failed to import _strptime because the import lockis

[ceph-users] Re: MDSs report slow metadata IOs

The current setup is only for testing functionality with ceph. My idea was to install a production setup with suitable hardware if all goes fine ... MDS is running on a node with 4 GB RAM, 1Gb/E and 4 core processor, The metadata pool is on 3 OSDs servers with 2 OSD per node: 24 GB RAM, 12 core

[ceph-users] the different between flag system and admin when create user rgw

when create user rgw using command : radosgw-admin user create --uid={username} --display-name="{display-name}" [--email={email}] There are 2 flag that I can use —system and —admin. What is this flag for ? -- Wahyu Muqsita Wardana Sys

[ceph-users] Bug identified: Dashboard proxy configuration is not working as expected

Hi, I have successfully configured Ceph dashboard following the this documentation. According to the documentation you can configure a URL prefix with this command: ceph config set mgr mgr/dashboard/url_prefix $PREFIX However when I try to access

[ceph-users] Re: ceph-volume lvm create leaves half-built OSDs lying around

On 11/09/2019 12:23, Jan Fajerski wrote: On Wed, Sep 11, 2019 at 11:17:47AM +0100, Matthew Vernon wrote: We keep finding part-made OSDs (they appear not attached to any host, and down and out; but still counting towards the number of OSDs); we never saw this with ceph-disk. On investigation, t