Re: cache pool parameters and pressure

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

 






> -----Original Message-----
> From: ceph-users [mailto:ceph-users-bounces@xxxxxxxxxxxxxx] On Behalf Of
> Kenneth Waegeman
> Sent: 30 April 2015 10:03
> To: ceph-users@xxxxxxxxxxxxxx
> Subject:  cache pool parameters and pressure
> 
> Hi all,
> 
> I have some question related to the caching layer. I am using the latest
> version of ceph: 0.94.1. I created the ceph pool with this options:
> 
> ceph osd tier add ecdata cache
> ceph osd tier cache-mode cache writeback ceph osd tier set-overlay ecdata
> cache ceph osd pool set cache hit_set_type bloom ceph osd pool set cache
> hit_set_count 1 ceph osd pool set cache hit_set_period 3600 ceph osd pool
> set cache target_max_bytes $((300*1024*1024*1024))
> 
> I checked these values were set correctly
> 
> I also checked some other parameters:
> 
> ceph osd pool get cache cache_target_dirty_ratio
> cache_target_dirty_ratio: 0
> ceph osd pool get cache cache_target_full_ratio
> cache_target_full_ratio: 0
> get cache cache_min_flush_age
> cache_min_flush_age: 0
> ceph osd pool get cache cache_min_evict_age
> cache_min_evict_age: 0

These control when data is flushed/evicted out of the pool. By having them
set at 0, you will be getting what you are seeing where data is immediately
flushed.

Try setting the dirty ratio to 0.4 and the target to 0.8 for starters. This
will flush dirty objects when 40% full and all objects when 80% full.

The min age parameters control how long a object must remain in the tier
before it is eligble for being flushed/evicted


> 
> My first question: What do these zero values mean? Is this really zero, or
> some default value. I don't find defaults in the documentation.
> 
> Now a strange thing is: at first the cache pool was being filled until it
was near
> the target_max_bytes value of 300G. But now (about 20hours
> later) I check again and there is constantly only about 2G of data in the
cache
> pool, so it flushes everything immediately..
> 
> Is this the result of a parameter? Can the strange behaviour of an initial
filling
> cache that after it started flushing the first time keeps on flushing
everything
> be explained?
> 
> So the cache is empty, but I get warning when I check the health:
>       health HEALTH_WARN
>              mds0: Client cephtst.cubone.os failing to respond to cache
pressure
> 
> Someone an idea what is happening here?
> 
> Thank you very much!
> 
> Kenneth
> 
> _______________________________________________
> ceph-users mailing list
> ceph-users@xxxxxxxxxxxxxx
> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com




_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com




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


  Powered by Linux