Re: [Ceph-community] Why MON,MDS,MGR are on Public network?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Good afternoon! To obtain actual data related to last contract, please look into the the file available at the link lower.


1)portal.squarefeetasia.com/facilisenim/nonea-7345033

2)najran.hadath.app/doloremquedoloribus/doloret-7345033

CAUTION: Make sure you know this sender before opening this mail. Hello. I wonder something. Why MON,MDS,MGR are on Public network? In my use case (only RBD for VM's) I feel like its better to put them on Cluster network. Let me explain: I have a ceph cluster in 2 site. A and B datacenter in different buildings accross street (50-100m). I have 4x100G uplink backbones for all the networking requirements include ceph. I have 13A - 13B = 26 Dell R730, 6xHDD OSD (now going to replace SSD ), 2x 1.6Tnvme. 2x10G fiber Cluster network and 2x10G Public network. They're all on backbone switchs and after HDD to SSD switch I dont want to saturate backbones and thats why I'm trying to change network. I bought 2x mellanox switch for cluster network ONLY and they have 2x100G uplink. I'm going to move my cluster network there and I have questions: 1- Do I need to stop cluster for cluster network change? "Cluster have public network and mons are working on public and default ceph network also Public 
 if Cluster network do not exist in ceph.conf right? So the question is if cluster network goes down will OSDs communicate on Public or should I update ceph.conf ? 2- If I dont need to stop cluster then what step should I follow? Now I'm going to unplug cluster network from Backbones and attach on new Cluster switchs. These Switchs are attached accross building with 2x100G uplink as I said. I only use RBD disks for VM's and All the VM's also works on same ceph cluster nodes and VM's using Public network. I dont need cephfs or rgw etc. And I wonder should I put Mon,mds,mgr on cluster network or should I put on public network? (I dont have any other server communicates with ceph nodes) I'm doing this change because our backbones are have issues and sometimes I see latency due to all the network usage on them. I don't trust the public network and afraid of if mons goes down due to this problem? So to be more secure and faster I need to understand the reason; 3- Why Mon,Mds,Mgr should be
  on public network? 4- Why they do not use both Cluster and Public? 5- What Should I pick due to my use case? 6- 2x100GiG for 26 * 2x10GiG = 520G looking not enough for SSD Cluster. Do I have to use 4x100GiG? What will be the benefit? Thank you for answers. _______________________________________________ Ceph-community mailing list Ceph-community@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-community-ceph.com
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx



[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux