We switch to forward mode as step to switch cache layer off.
Right now we have "samsung 850 pro" in cache layer (10 ssd, 2 per nodes) and they show 2MB for 4K blocks... 250 IOPS... intead of 18-20K for intel S3500 240G which we choose as replacement..
So with such good disks - cache layer - very big bottleneck for us...
2015-08-21 2:02 GMT+03:00 Samuel Just <sjust@xxxxxxxxxx>:
What's supposed to happen is that the client transparently directs all
requests to the cache pool rather than the cold pool when there is a
cache pool. If the kernel is sending requests to the cold pool,
that's probably where the bug is. Odd. It could also be a bug
specific 'forward' mode either in the client or on the osd. Why did
you have it in that mode?
-Sam
On Thu, Aug 20, 2015 at 3:58 PM, Voloshanenko Igor
<igor.voloshanenko@xxxxxxxxx> wrote:
> We used 4.x branch, as we have "very good" Samsung 850 pro in production,
> and they don;t support ncq_trim...
>
> And 4,x first branch which include exceptions for this in libsata.c.
>
> sure we can backport this 1 line to 3.x branch, but we prefer no to go
> deeper if packege for new kernel exist.
>
> 2015-08-21 1:56 GMT+03:00 Voloshanenko Igor <igor.voloshanenko@xxxxxxxxx>:
>>
>> root@test:~# uname -a
>> Linux ix-s5 4.0.4-040004-generic #201505171336 SMP Sun May 17 17:37:22 UTC
>> 2015 x86_64 x86_64 x86_64 GNU/Linux
>>
>> 2015-08-21 1:54 GMT+03:00 Samuel Just <sjust@xxxxxxxxxx>:
>>>
>>> Also, can you include the kernel version?
>>> -Sam
>>>
>>> On Thu, Aug 20, 2015 at 3:51 PM, Samuel Just <sjust@xxxxxxxxxx> wrote:
>>> > Snapshotting with cache/tiering *is* supposed to work. Can you open a
>>> > bug?
>>> > -Sam
>>> >
>>> > On Thu, Aug 20, 2015 at 3:36 PM, Andrija Panic
>>> > <andrija.panic@xxxxxxxxx> wrote:
>>> >> This was related to the caching layer, which doesnt support
>>> >> snapshooting per
>>> >> docs...for sake of closing the thread.
>>> >>
>>> >> On 17 August 2015 at 21:15, Voloshanenko Igor
>>> >> <igor.voloshanenko@xxxxxxxxx>
>>> >> wrote:
>>> >>>
>>> >>> Hi all, can you please help me with unexplained situation...
>>> >>>
>>> >>> All snapshot inside ceph broken...
>>> >>>
>>> >>> So, as example, we have VM template, as rbd inside ceph.
>>> >>> We can map it and mount to check that all ok with it
>>> >>>
>>> >>> root@test:~# rbd map
>>> >>> cold-storage/0e23c701-401d-4465-b9b4-c02939d57bb5
>>> >>> /dev/rbd0
>>> >>> root@test:~# parted /dev/rbd0 print
>>> >>> Model: Unknown (unknown)
>>> >>> Disk /dev/rbd0: 10.7GB
>>> >>> Sector size (logical/physical): 512B/512B
>>> >>> Partition Table: msdos
>>> >>>
>>> >>> Number Start End Size Type File system Flags
>>> >>> 1 1049kB 525MB 524MB primary ext4 boot
>>> >>> 2 525MB 10.7GB 10.2GB primary lvm
>>> >>>
>>> >>> Than i want to create snap, so i do:
>>> >>> root@test:~# rbd snap create
>>> >>> cold-storage/0e23c701-401d-4465-b9b4-c02939d57bb5@new_snap
>>> >>>
>>> >>> And now i want to map it:
>>> >>>
>>> >>> root@test:~# rbd map
>>> >>> cold-storage/0e23c701-401d-4465-b9b4-c02939d57bb5@new_snap
>>> >>> /dev/rbd1
>>> >>> root@test:~# parted /dev/rbd1 print
>>> >>> Warning: Unable to open /dev/rbd1 read-write (Read-only file system).
>>> >>> /dev/rbd1 has been opened read-only.
>>> >>> Warning: Unable to open /dev/rbd1 read-write (Read-only file system).
>>> >>> /dev/rbd1 has been opened read-only.
>>> >>> Error: /dev/rbd1: unrecognised disk label
>>> >>>
>>> >>> Even md5 different...
>>> >>> root@ix-s2:~# md5sum /dev/rbd0
>>> >>> 9a47797a07fee3a3d71316e22891d752 /dev/rbd0
>>> >>> root@ix-s2:~# md5sum /dev/rbd1
>>> >>> e450f50b9ffa0073fae940ee858a43ce /dev/rbd1
>>> >>>
>>> >>>
>>> >>> Ok, now i protect snap and create clone... but same thing...
>>> >>> md5 for clone same as for snap,,
>>> >>>
>>> >>> root@test:~# rbd unmap /dev/rbd1
>>> >>> root@test:~# rbd snap protect
>>> >>> cold-storage/0e23c701-401d-4465-b9b4-c02939d57bb5@new_snap
>>> >>> root@test:~# rbd clone
>>> >>> cold-storage/0e23c701-401d-4465-b9b4-c02939d57bb5@new_snap
>>> >>> cold-storage/test-image
>>> >>> root@test:~# rbd map cold-storage/test-image
>>> >>> /dev/rbd1
>>> >>> root@test:~# md5sum /dev/rbd1
>>> >>> e450f50b9ffa0073fae940ee858a43ce /dev/rbd1
>>> >>>
>>> >>> .... but it's broken...
>>> >>> root@test:~# parted /dev/rbd1 print
>>> >>> Error: /dev/rbd1: unrecognised disk label
>>> >>>
>>> >>>
>>> >>> =========
>>> >>>
>>> >>> tech details:
>>> >>>
>>> >>> root@test:~# ceph -v
>>> >>> ceph version 0.94.2 (5fb85614ca8f354284c713a2f9c610860720bbf3)
>>> >>>
>>> >>> We have 2 inconstistent pgs, but all images not placed on this pgs...
>>> >>>
>>> >>> root@test:~# ceph health detail
>>> >>> HEALTH_ERR 2 pgs inconsistent; 18 scrub errors
>>> >>> pg 2.490 is active+clean+inconsistent, acting [56,15,29]
>>> >>> pg 2.c4 is active+clean+inconsistent, acting [56,10,42]
>>> >>> 18 scrub errors
>>> >>>
>>> >>> ============
>>> >>>
>>> >>> root@test:~# ceph osd map cold-storage
>>> >>> 0e23c701-401d-4465-b9b4-c02939d57bb5
>>> >>> osdmap e16770 pool 'cold-storage' (2) object
>>> >>> '0e23c701-401d-4465-b9b4-c02939d57bb5' -> pg 2.74458f70 (2.770) -> up
>>> >>> ([37,15,14], p37) acting ([37,15,14], p37)
>>> >>> root@test:~# ceph osd map cold-storage
>>> >>> 0e23c701-401d-4465-b9b4-c02939d57bb5@snap
>>> >>> osdmap e16770 pool 'cold-storage' (2) object
>>> >>> '0e23c701-401d-4465-b9b4-c02939d57bb5@snap' -> pg 2.793cd4a3 (2.4a3)
>>> >>> -> up
>>> >>> ([12,23,17], p12) acting ([12,23,17], p12)
>>> >>> root@test:~# ceph osd map cold-storage
>>> >>> 0e23c701-401d-4465-b9b4-c02939d57bb5@test-image
>>> >>> osdmap e16770 pool 'cold-storage' (2) object
>>> >>> '0e23c701-401d-4465-b9b4-c02939d57bb5@test-image' -> pg 2.9519c2a9
>>> >>> (2.2a9)
>>> >>> -> up ([12,44,23], p12) acting ([12,44,23], p12)
>>> >>>
>>> >>>
>>> >>> Also we use cache layer, which in current moment - in forward mode...
>>> >>>
>>> >>> Can you please help me with this.. As my brain stop to understand
>>> >>> what is
>>> >>> going on...
>>> >>>
>>> >>> Thank in advance!
>>> >>>
>>> >>>
>>> >>>
>>> >>>
>>> >>>
>>> >>> _______________________________________________
>>> >>> ceph-users mailing list
>>> >>> ceph-users@xxxxxxxxxxxxxx
>>> >>> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
>>> >>>
>>> >>
>>> >>
>>> >>
>>> >> --
>>> >>
>>> >> Andrija Panić
>>> >>
>>> >> _______________________________________________
>>> >> 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
