> 在 2021年5月4日，22:31，mabi <mabi@xxxxxxxxxxxxx> 写道： > > Hello, > > I have a small Octopus cluster (3 mon/mgr nodes, 3 osd nodes) installed with cephadm and hence running inside podman containers on Ubuntu 20.04. > > I want to use CephFS so I created a fs volume and saw that two MDS containers have been automatically deployed on two of my OSD nodes. Now I saw in the documentation (https://apac01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdocs.ceph.com%2Fen%2Flatest%2Fcephfs%2F&data=04%7C01%7C%7C1d93c6c843dd43ebc36908d90f095e0d%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637557355176012495%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=1s8fpu7RHINF4HlascQyVVCDdSGSLBz2sFCPsypIRuU%3D&reserved=0) that the MDS first writes to a journal before writing into the meta pool so I was wondering where exactly is this MDS journal located? Hi Mabi, MDS also write its journal to the meta pool. And eventually located on the OSDs. > Because my MDS service is running inside a container, does this mean the journal gets written inside the container itself? No. > I am asking because my containers are running from the OS hard disk which is not as performant as the OSD hard disk. So if this is the case it would be better for performance to have a dedicated hard disk for MDS? is this correct? No. > And if yes how do I specify to cephadm that my MDS container should be writing its journal at another location? > > Finally I read (https://apac01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdocs.ceph.com%2Fen%2Flatest%2Fstart%2Fhardware-recommendations%2F&data=04%7C01%7C%7C1d93c6c843dd43ebc36908d90f095e0d%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637557355176012495%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=ygSTssVsSUSnCTH4muX7wlMY4TiyZ7LMXhVZDkYouwM%3D&reserved=0) that the Ceph MDS service only needs 1 MB storage for its journal, is that all? No. The journal can get quite big. This 1MB storage is only for something like keyring, config files. Regards, Weiwen Hu > Best regards, > Mabi > _______________________________________________ > ceph-users mailing list -- ceph-users@xxxxxxx > To unsubscribe send an email to ceph-users-leave@xxxxxxx _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |