Re: [ceph-users] Query regarding installation of Emperor

You may have set HTTP_PROXY environment and that's why you can succeed to execute the command directly. Please try to edit the /etc/wgetrc to enable proxy (this needs to be done on all the Ceph nodes): https_proxy = http://[proxy-server]:[port]/ http_proxy =

Re: [ceph-users] Networking questions

Hello Kyle, On Thu, 26 Dec 2013 16:33:01 -0800 Kyle Bader wrote: > > Do monitors have to be on the cluster network as well or is it > > sufficient for them to be on the public network as > > http://ceph.com/docs/master/rados/configuration/network-config-ref/ > > suggests? > > Monitors only need

Re: [ceph-users] Networking questions

> Do monitors have to be on the cluster network as well or is it sufficient > for them to be on the public network as > http://ceph.com/docs/master/rados/configuration/network-config-ref/ > suggests? Monitors only need to be on the public network. > Also would the OSDs re-route their traffic over

[ceph-users] Journal configuration

Hi to All, I want to know what is the best configuration for the Journal devices, for example i got: ceph-node01: OSDs: 73GB (x10): /dev/sda /dev/sdb /dev/sdc /dev/sdd /dev/sde /dev/sdf /dev/sdg /dev/sdh /dev/sdi /dev/sdj 500GB (x2): /dev/sdk /dev/sdl Journals (x2): 73GB:

Re: [ceph-users] Failure probability with largish deployments

> Yes, that also makes perfect sense, so the aforementioned 12500 objects > for a 50GB image, at a 60 TB cluster/pool with 72 disk/OSDs and 3 way > replication that makes 2400 PGs, following the recommended formula. > >> > What amount of disks (OSDs) did you punch in for the following run? >> >> Di

[ceph-users] RBD is very slow

Hello, i have 2 OSD's and 3 MON's. Every OSD is on a 2,5TB LVM/EXT4 storage. Why is the access to the rbd device so slow and what is meaning with "Stddev Bandwidth" in rados bench? See the below statistics: #Create a 1GB file on local Storage and test a local OSD for bandwidth cd /ceph/; dd if=/d

[ceph-users] RBD is very slow

Hello, i have 2 OSD's and 3 MON's. Every OSD is on a 2,5TB LVM/EXT4 storage. Why is the access to the rbd device so slow and what is meaning with "Stddev Bandwidth" in rados bench? See the below statistics: #Create a 1GB file on local Storage and test a local OSD for bandwidth cd /ceph/; dd if=/d

Re: [ceph-users] [RadosGW] FastCGI: comm with server "/var/www/s3gw.fcgi" aborted: read failed

On Thu, Dec 26, 2013 at 7:58 AM, Kuo Hugo wrote: > Hi Yehuda, > > I did hit the apache concurrent connections that limit to 150 by default in > apache at beginning. Several parameters were been bumped up. After that the > concurrent connections of apache looks sufficient for the benchmark run or >

Re: [ceph-users] [RadosGW] FastCGI: comm with server "/var/www/s3gw.fcgi" aborted: read failed

Hi Yehuda, I did hit the apache concurrent connections that limit to 150 by default in apache at beginning. Several parameters were been bumped up. After that the concurrent connections of apache looks sufficient for the benchmark run or those failed requests will not been sent to FastCGI sever fr

Re: [ceph-users] [RadosGW] FastCGI: comm with server "/var/www/s3gw.fcgi" aborted: read failed

On Thu, Dec 26, 2013 at 3:00 AM, Kuo Hugo wrote: > Hi all, > > > I think the FastCGI module is the latest one on my server. > > root@p01:/var/log# dpkg -l | grep cgi > > ii libapache2-mod-fastcgi2.4.7~0910052141-2~bpo70+1.ceph > Apache 2 FastCGI module for long-running CGI scripts > i

[ceph-users] Query regarding installation of Emperor

Hii, I am kiriti. I have tried installing ceph(Emperor) but I am facing some proxy problems like Connection timed out while installing ceph-deploy especially during the command *wget -q -O- 'https://ceph.com/git/?p=ceph.git;a=blob_plain;f=keys/release.asc

[ceph-users] Some question.... restarting whole cluster(OSD+MON)

Hi, We are planing to test and use ceph object storage(no RBD/no cephFS), but before that we have some questions: 1. How much time it will take to be ready for processing client request if we restart the the whole cluster(MON+OSDs). In this case how cluster map will create(OSD dependency). 2.

[ceph-users] ceph primetime

Hello, We are thinking of deploying a distributed filesystem for our lab. At the moment we are using Dell storage for a 20 node computational cluster. We are having many bottlenecks in storage i/o. We are interesting in Lustre, XtremeFS, and Ceph. It seems Ceph isn't ready for production accordin

Re: [ceph-users] compiling ceph

I suppose I can manage the dependencies for building ceph. Does the ceph source code also have the ceph kernel module? On Thu, Dec 26, 2013 at 2:52 AM, Loic Dachary wrote: > > > On 26/12/2013 01:29, Keith Brown wrote: > > Hello, > > > > I would like to compile ceph from scratch. I am reading

Re: [ceph-users] [RadosGW] FastCGI: comm with server "/var/www/s3gw.fcgi" aborted: read failed

Hi all, I think the FastCGI module is the latest one on my server. root@p01:/var/log# dpkg -l | grep cgi ii libapache2-mod-fastcgi2.4.7~0910052141-2~bpo70+1.ceph Apache 2 FastCGI module for long-running CGI scripts ii libfcgi0ldbl 2.4.0-8.1 Shared library of F

Re: [ceph-users] Sanity check of deploying Ceph very unconventionally (on top of RAID6, with very few nodes and OSDs)

Hello, I'm leaving the original post below for your perusal, but will top-post here with the conclusions/decisions so far. Firstly, I got a quotation from our vendors here (assuming a slightly incorrect 100:1 exchange rate of Japanese Yen to USD). A storage node as described below, with 24 3TB d

[ceph-users] Bucket seems not deleted completely in RGW

Hello, I noticed that when a bucket deleted the some objects related to that bucket in pool .rgw and .rgw.buckets (I used default configuration of RGW) weren't deleted. For example, I created a bucket named "test-bucket" and put two objects("test-object" and "test-object2") into it. After "test

[ceph-users] How to replace admin node

HI I have a setup of 3 nodes, where 1 node running as admin + mon + osd, and the other two nodes as osd. Now question is what will happen if the first node die? then how I can add another node as admin? How we can replace admin node? Br. ___ ceph-user

Re: [ceph-users] Upcoming Erasure coding

Hello, On Tue, 24 Dec 2013 16:33:49 +0100 Loic Dachary wrote: > > > On 24/12/2013 10:22, Wido den Hollander wrote: > > On 12/24/2013 09:34 AM, Christian Balzer wrote: > >> > >> Hello Loic, > >> > >> On Tue, 24 Dec 2013 08:29:38 +0100 Loic Dachary wrote: > >> > >>> > >>> > >>> On 24/12/2013 05:4