Questions tagged [Ceph] (23)

1
answer

The organization of a network for Ceph cluster?

There was a need to raise up the Ceph cluster to distribute files to multiple servers (currently 3, but soon will be 5 - 10): portforwarding in Docker, storage, database, static files, possibly in a virtual machine. This: pfsense is also a router (ports 1 Gbps), unmanaged switch D-Link DS-1024 (24 x 1 Gbit/s), a small possi...
Avis asked April 3rd 20 at 16:39
1
answer

How to configure ceph OSD disks behind a hardware RAID?

Prepare the infrastructure for migration to a cluster with ceph support (proxmox 5.4, stopped until it since 6.1 came only yesterday, and with the 6.0 had lots of problems, according to his colleagues). There are nodes on which part of the sata ports are behind a hardware RAID and if inicializacija OSD screw over the RAID s...
alek.Hodkiewicz asked April 3rd 20 at 16:37
0
answer

How to write your CRUSH Maps-ceph?

performed manual and at the time of map change stuck. like all right, but throws an error.bad mapping rule 0 x 0 num_rep 3 result [0,2] bad mapping rule 0 x 1 num_rep 3 result [0,3] bad mapping rule 0 x 2 num_rep 3 result [1,2] bad mapping rule 0 x 3 num_rep 3 result [0,2] bad mapping rule 0 x 4 num_rep 3 result [2,1] bad m...
ciara16 asked March 31st 20 at 15:36
0
answer

What to choose for web cluster — CephFS vs. GlusterFS vs something else?

Good evening. There is a cluster of 5 virtual machines (DigitalOcean instances) with the following characteristics: 8 GB RAM, 4 vCPU, 50 GB SSD disk. In this cluster you plan to host a few(up to 10 pieces) online stores on Wordpress and Magento. To provide failover on three nodes deployed database cluster (MySQL Percona Xt...
darrell asked March 30th 20 at 00:04
0
answer

How to fix warning "ceph MDS cache is too large (2GB/1GB)" and is it necessary?

Has Proxmox on 3 nodes. From its interface normally raised ceph. Periodically one (arbitrary) of MDS at the interface Proxmox issued a warning from the header. In the console it looks like this:root@px0:~# ceph -s cluster: id: 36229d63-223e-4511-94f4-2069eeb50b6f health: HEALTH_WARN 1 MDSs report oversized cache service...
Emma.Hamm asked March 25th 20 at 13:33
4
answers

What to do with the hard drive, whose errors print_req_error: I/O error?

Hello. Please tell me what I can do with this disk?(except to change) Critical are these errors?[531787.817056] ata4.00: configured for UDMA/133 [531787.817074] sd 3:0:0:0: [sdd] tag#4 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE [531787.817078] sd 3:0:0:0: [sdd] tag#4 Sense Key : Medium Error [current] [531787.81...
Otha.Reinger asked March 23rd 20 at 19:59
1
answer

What RAID configuration is best for a Ceph?

Good day! We need to organize the cluster's Ceph. There are five servers, each with four HDD. Disks connected via hardware RAID. It is clear that fault tolerance is achieved programmatically using the Ceph, so to speak, for this and all afoot. The question is more about performance. What RAID configuration is better from th...
moses86 asked March 19th 20 at 08:54
1
answer

It possible to simultaneously access the same file from multiple nodes?

There are CephFS, for example there is this store any one file. This file directory mounted on different hosts, called clients. Will ceph simultaneously to write to this file from two different hosts? That is, is there some kind of lock file during recording or the order of the records?
dee_Ritch asked March 18th 20 at 20:05
1
answer

Why when you use rados gateway ceph there is a low job?

Colleagues, good afternoon, everyone. Set up the cluster: cluster: id: 68d893db-be6f-4b52-ba62-b0e99f492e65 health: HEALTH_WARN 1 pools have many more objects per pg than average services: mon: 3 daemons, quorum mn-2,mn-node3,mn-node4 mgr: mn-node2(active) standbys: mn-node3, mn-node4 osd: 15 osds: 15 up, 15 in rgw: daemon ...
Felicia_Green asked March 17th 20 at 12:59
1
answer

Whether to use a rook.io to deploy a CephFS in Kubernetes?

Hello. Whether to use a Rook to deploy a CephFS cluster on the basis of Kubernetes and use this cluster as storage? Or better hands to do?
andy73 asked March 14th 20 at 15:32