Re: [ceph-users] kraken-bluestore 11.2.0 memory leak issue

Thanks IIya Letkowski for the information we will change this value accordingly. Thanks, Muthu On 15 February 2017 at 17:03, Ilya Letkowski wrote: > Hi, Muthusamy Muthiah > > I'm not totally sure that this is a memory leak. > We had same problems with bluestore on ceph

Re: [ceph-users] bcache vs flashcache vs cache tiering

Hello Sage, On Wed, 15 Feb 2017 22:10:26 + (UTC) Sage Weil wrote: > On Wed, 15 Feb 2017, Nick Fisk wrote: > > Just an update. I spoke to Sage today and the general consensus is that > > something like bcache or dmcache is probably the long term goal, but work > > needs to be done before its

[ceph-users] How to integrate rgw with hadoop?

Hi all, I would like to integrate radosgw with hadoop. We are using 2.7.3 version of hadoop. So, s3a can be suitable plugin. But, I can't find any guide for it. Could anybody help me? Thanks. John Haan ___ ceph-users mailing list

Re: [ceph-users] Passing LUA script via python rados execute

Hi Nick, First thing to note is that in Kraken that object classes not whitelisted need to be enabled explicitly. This is in the Kraken release notes ( http://docs.ceph.com/docs/master/release-notes/): tldr: add 'osd class load list = *' and 'osd class default list = *' to ceph.conf. - The ‘osd

Re: [ceph-users] bcache vs flashcache vs cache tiering

> On Wed, 15 Feb 2017, Nick Fisk wrote: > > Just an update. I spoke to Sage today and the general consensus is > > that something like bcache or dmcache is probably the long term goal, > > but work needs to be done before its ready for prime time. The current > > tiering functionality won't be

Re: [ceph-users] bcache vs flashcache vs cache tiering

On Wed, 15 Feb 2017, Nick Fisk wrote: > Just an update. I spoke to Sage today and the general consensus is that > something like bcache or dmcache is probably the long term goal, but work > needs to be done before its ready for prime time. The current tiering > functionality won't be going away in

Re: [ceph-users] bcache vs flashcache vs cache tiering

> -Original Message- > From: ceph-users [mailto:ceph-users-boun...@lists.ceph.com] On Behalf Of > Nick Fisk > Sent: 15 February 2017 09:53 > To: 'Christian Balzer' ; 'Ceph Users' us...@lists.ceph.com> > Subject: Re: [ceph-users] bcache vs flashcache vs cache tiering > > >

[ceph-users] Passing LUA script via python rados execute

Hi Noah, I'm trying to follow your example where you can pass a LUA script as json when calling the rados execute function in Python. However I'm getting a rados permission denied error saying its failed to read the test object I have placed on the pool. I have also tried calling the cls_hello

Re: [ceph-users] kraken-bluestore 11.2.0 memory leak issue

Hi, Muthusamy Muthiah I'm not totally sure that this is a memory leak. We had same problems with bluestore on ceph v11.2.0. Reduce bluestore cache helped us to solve it and stabilize OSD memory consumption on the 3GB level. Perhaps this will help you: bluestore_cache_size = 104857600 On Wed,

Re: [ceph-users] [RFC] rbdmap unmap - unmap all, or only RBDMAPFILE listed images?

In the case where the rbdmap utility is invoked by systemd (i.e. to map RBD devices on system startup and unmap RBD devices on system shutdown) [1], I would think the current unmap-all behavior is desirable in much the same way that only filesystems listed in "/etc/fstab" are mounted at startup

[ceph-users] [RFC] rbdmap unmap - unmap all, or only RBDMAPFILE listed images?

Hi, I'm working on an rbdmap change https://github.com/ceph/ceph/pull/13361, and would appreciate some input from existing users. Currently "rbdmap map" maps any RBD images listed in the rbdmap config file (RBDMAPFILE), whereas "rbdmap unmap" unmaps all mapped RBD images, regardless of whether

Re: [ceph-users] Ceph OSDs advice

You should subtract buffers and cache from the used memory to get a more accurate representation of how much memory is actually available to processes. In this case that puts you around 22G of used - or a better term might be unavailable memory. Buffers and cache can be reallocated when needed -

Re: [ceph-users] bcache vs flashcache vs cache tiering

> -Original Message- > From: Christian Balzer [mailto:ch...@gol.com] > Sent: 15 February 2017 01:42 > To: 'Ceph Users' > Cc: Nick Fisk ; 'Gregory Farnum' > Subject: Re: [ceph-users] bcache vs flashcache vs cache tiering > >

Re: [ceph-users] bcache vs flashcache vs cache tiering

thanx Nick, Gregory and Wido, So at least, we can say the cache tiering in Jewel is stable enough I think. I like cache tiering more than the others, but yes, there is a problem about cache tiering in flushing data between different nodes, which are not a problem in local caching solution.