On Thu, 14 Aug 2014, Pawe? Sadowski wrote: > Hello, > > I've a cluster of 35 OSD (30 HDD, 5 SSD) with cache tiering configured. > During tests it looks like ceph is not respecting target_max_bytes > settings. Steps to reproduce: > - configure cache tiering > - set target_max_bytes to 32G (on hot pool) > - write more than 32G of data > - nothing happens Can you 'ceph pg dump pools -f json-pretty' at this point? And pick a random PG in the cache pool and capture the output of 'ceph pg <pgid> query'. Then 'ceph tell osd.* injectargs '--debug-ms 1 --debug-osd 20'. > If I set target_max_bytes again (to the same value or any other option, > for example cache_min_evict_age) ceph will start to move data from hot > to base pool. Once it starts going, capture an OSD log (/var/log/ceph/ceph-osd.NNN.log) for an OSD that is now moving data. Thanks! sage > > I'm using ceph in version 0.80.4 (with cherry-picked patch from bug > http://tracker.ceph.com/issues/8982. > > Is there away to make it work as expected? > > -- > PS > > > _______________________________________________ > ceph-users mailing list > ceph-users at lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > >
 |