ceph docker

1 Describe my deployment environment first:2 osd,1 monitor,1 Console Management Server, 1 client, are 24 cores, 64G memory, 1.6T SSD flash card, gigabit network card; The Ceph version currently installed is 0.94.7.2 Current statusI use the DD command to write 5G data, using Iostat to observe%util immediately 100% Ah, while the await indicators are more than 4,000, and at this time the network bandwidth is only used around 10M.650) this.width=650; "Src

, type_zmfuz3pozw5nagvpdgk =) Add a new file: echo "Deb http://download.proxmox.com/debian/pve stretch PVE-no-subces">/etc/APT/sources. list. d/pve-install-repo.list! [] (Http://i2.51cto.com/images/blog/201811/02/d7f8fc1ea50ea3a7c2f87b8cb537ef17.jpg? X-OSS-process = image/watermark, size_16, expires, color_ffffff, t_100, g_se, x_10, y_10, shadow_90, type_zmfuz3pozw5nagvpdgk =) download key: wget http://download.proxmox.com/debian/proxmox-ve-release-5.x.gpg-O/etc/APT/trusted. GPG. d/proxmox-ve-re

Machine Room operation error caused the rack or host power down is the occasional thing, then how in this case, let the Ceph service with the OS startup and fast start? Here is a simple method:Execute the following command on the OSD host:sudo ln-s/usr/lib/systemd/system/[email protected]/etc/systemd/system/multi-user.target.wants/[email protected]sudo systemctl enable [email protected]sudo systemctl is-enabled [email protected]Execute the following c

=${ceph_replicas:-1}remote_ceph=falseremote_ceph_admin_key_ Path=/etc/ceph/ceph.client.admin.keyringenabled_services+=,g-api,g-regenabled_services+=,cinder,c-api,c-vol, C-sch,c-bakcinder_driver=cephcinder_enabled_backends=cephenabled_services+=,n-api,n-crt,n-cpu,n-cond,n-sch, n-netdefault_instance_type=m1.microenable_servicehorizondisable_servicen-netenable_service q-svcenable_serviceq-agtenable_serviceq-dhcpenable_serviceq-l3enable_service Q-metaenab

Calculation formula for PG: Calculation formula for the entire cluster PG The calculation formula for PG in each pool: Total PGs = (TOTAL_NUMBER_OF_OSD *)/Max_replication_count/pool_count The result of the settlement is taken up to the value of the n-th square of the 2. For example, the total number of OSD is 160, the number of copies of the number of 3,pool is also 3, then the result is calculated according to the above formula is 1777.7. Take 2 of the n-th is 2048, then the number

A 192.168.199.128 (Ceph back-end)------(Ceph client) B 192.168.199.132 A on configuration: Install TGT service, apt-get install TGT-RBD IQN command format: iqn. 2. Command line creation: Tgtadm--lld iSCSI--mode Target--op new--tid 1--targetname iqn.2016-08.com. Example:iscsi Configuration on B: Install iSCSI client, Apt-get install OPEN-ISCSI discover target iscsiadm-m discovery-t sendtargets-p 192.168

Environment DescriptionOpenstack Icehouse Ceph 0.87 Configuration Steps Configuring Ceph Add the following to the ceph.conf configuration file RBD cache = True RBD cache Writethrough until Flush = True RBD Cache also has some configuration properties that can be changed as needed #rbd Cache size #rbd cache max Dirty #rbd cache target Dirty Configure OpenStack Add the following to the [Defa

[Ceph_deploy] [ERROR] Runtimeerror:failed to execute Command:ceph-disk-activate–mark-init sysvinit–mount/dev/sdb1To be honest, the question is two.I put the OSD in a separate partition SDB executes the commands that are:Ceph-deploy OSD Prepare NETWORK:/DEV/SDB1Ceph-deploy OSD Activate NETWORK:/DEV/SDB1The above is wrong.The right thing isCeph-deploy Osd–zap-disk Create Network:sdbCeph-deploy OSD Prepare NETWORK:/DEV/SDB1:/DEV/SDB2Ceph-deploy OSD Activate NETWORK:/DEV/SDB1:/DEV/SDB2The problem is

Have you modified/sys/block/sdk/queue/read_ahead_kb read_ahead?Modified to 8192, the default value is 128;is the SATA disk OSD changed to 8192? How much of a performance boost?In addition, this value should be modified at any time, right? Can it be dynamically adjusted after deployment?What effect will it have on the OSD?can improve read performanceCan be modified in real timeWhat does this parameter mean?Pre-read bytes?This parameter is useful for sequential reads, meaning how much content to r

return {code:403, Message:forbidden} When using the admin Ops API to get user list Get/admin/user by official documents Here are two questions: the first user list request is as follows: Get/admin/metadata/usersecond, you need to configure the metadata read permissions for the administrative user: Radosgw-admin caps Add--uid=admin--caps= "Metadata=read"Bottom line: #You must use Get/admin/metadata/user to fetch the user list (with metadata capabity).Cep