Re: [ceph-users] CephFS MDS server stuck in "resolve" state

On Wed, Jul 4, 2018 at 7:02 PM Dennis Kramer (DBS) wrote: > > Hi, > > I have managed to get cephfs mds online again...for a while. > > These topics covers more or less my symptoms and helped me get it up > and running again: > - https://www.spinics.net/lists/ceph-users/msg45696.h > tml > -

[ceph-users] Deep scrub interval not working

Hi, I want to try to distribute the deep scrub by running the deep scrub manually. I have set osd deep scrub interval to 30 days and osd scrub max interval to 14 days. However, I still getting deep-scrub even though all the last deep scrub did not pass the 30 days period. How do I reset the

Re: [ceph-users] Slow requests

On Wed, Jul 4, 2018 at 6:26 PM, Benjamin Naber wrote: > Hi @all, > > im currently in testing for setup an production environment based on the > following OSD Nodes: > > CEPH Version: luminous 12.2.5 > > 5x OSD Nodes with following specs: > > - 8 Core Intel Xeon 2,0 GHZ > > - 96GB Ram > > - 10x

Re: [ceph-users] "ceph pg scrub" does not start

Hi Sean, Many thanks for the suggestion, but unfortunately deep-scrub also appears to be ignored: # ceph pg deep-scrub 4.ff instructing pg 4.ffs0 on osd.318 to deep-scrub 'tail -f ceph-osd.318.log' shows no new entries. To get more info, I set debug level 10 on the osd, and issued another

Re: [ceph-users] Ceph behavior on (lots of) small objects (RGW, RADOS + erasure coding)?

Hi! * Gregory Farnum [2018-06-28 19:31:09 -0700]: > That’s close but not *quite* right. It’s not that Ceph will explicitly > “fall back” to replication. In most (though perhaps not all) erasure codes, > what you’ll see is full sized parity blocks, a full store of the data (in > the default

Re: [ceph-users] Slow requests

hi Caspar, ty for the reply. ive updatet all SSDs to actual firmware. Still having the same error. the strange thing is that this issue switches from node to node and from osd to osd. HEALTH_WARN 4 slow requests are blocked > 32 sec REQUEST_SLOW 4 slow requests are blocked > 32 sec     1 ops

Re: [ceph-users] CephFS MDS server stuck in "resolve" state

Hi, I have managed to get cephfs mds online again...for a while. These topics covers more or less my symptoms and helped me get it up and running again: - https://www.spinics.net/lists/ceph-users/msg45696.h tml - http://lists.ceph.com/pipermail/ceph-users-ceph.com/2017-December/ 023133.html

[ceph-users] WAL/DB partition on system SSD

Hello, I wonder how good or bad is to reserve space for WAL/DB on the same SSD as OS or is it better to separate them? And what are recommended size for WAL/DB partition? I am building Luminous cluster on Centos7 with 2 OSD per host, each OSD would be 2 * 8tb disks through LVM. Anton.

Re: [ceph-users] Slow requests

Hi Ben, At first glance i would say the CPU's are a bit weak for this setup. Recommended is to have at least 1 core per OSD. Since you have 8 cores and 10 OSD's there isn't much left for other processes. Furthermore, did you upgrade the firmware of those DC S4500's to the latest firmware?

Re: [ceph-users] RADOSGW err=Input/output error

Hi Drew, Try to increase debugging with debug ms = 1 debug rgw = 20 Regards Kev - Original Message - From: "Drew Weaver" To: "ceph-users" Sent: Tuesday, July 3, 2018 1:39:55 PM Subject: [ceph-users] RADOSGW err=Input/output error An application is having general failures writing

[ceph-users] Slow requests

Hi @all, im currently in testing for setup an production environment based on the following OSD Nodes: CEPH Version: luminous 12.2.5 5x OSD Nodes with following specs: - 8 Core Intel Xeon 2,0 GHZ - 96GB Ram - 10x 1,92 TB Intel DC S4500 connectet via SATA - 4x 10 Gbit NIC 2 bonded via LACP