BlueFS spillover detected, why, what?
Hi Recently our ceph cluster (nautilus) is experiencing bluefs spillovers, just 2 osd's and I disabled the warning for these osds. (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false) I'm wondering what causes this and how this can be prevented. As I understand it the rocksdb for the OSD needs to store more than fits on the NVME logical volume (123G for 12T OSD). A way to fix it could be to increase the logical volume on the nvme (if there was space on the nvme, which there isn't at the moment). This is the current size of the cluster and how much is free: [root@cephmon1 ~]# ceph df RAW STORAGE: CLASS SIZE AVAIL USED RAW USED %RAW USED hdd 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 TOTAL 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 POOLS: POOL ID STORED OBJECTS USED %USED MAX AVAIL cephfs_data 1 572 MiB 121.26M 2.4 GiB 0 167 TiB cephfs_metadata 2 56 GiB 5.15M 57 GiB 0 167 TiB cephfs_data_3copy 8 201 GiB 51.68k 602 GiB 0.09 222 TiB cephfs_data_ec83 13 643 TiB 279.75M 953 TiB 58.86 485 TiB rbd 14 21 GiB 5.66k 64 GiB 0 222 TiB .rgw.root 15 1.2 KiB 4 1 MiB 0 167 TiB default.rgw.control 16 0 B 8 0 B 0 167 TiB default.rgw.meta 17 765 B 4 1 MiB 0 167 TiB default.rgw.log 18 0 B 207 0 B 0 167 TiB cephfs_data_ec57 20 433 MiB 230 1.2 GiB 0 278 TiB The amount used can still grow a bit before we need to add nodes, but apparently we are running into the limits of our rocskdb partitions. Did we choose a parameter (e.g. minimal object size) too small, so we have too much objects on these spillover OSDs? Or is it that too many small files are stored on the cephfs filesystems? When we expand the cluster, we can choose larger nvme devices to allow larger rocksdb partitions, but is that the right way to deal with this, or should we adjust some parameters on the cluster that will reduce the rocksdb size? Cheers /Simon
Hi Simon As far as I know, RocksDB only uses "leveled" space on the NVME partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every DB space above such a limit will automatically end up on slow devices. In your setup where you have 123GB per OSD that means you only use 30GB of fast device. The DB which spills over this limit will be offloaded to the HDD and accordingly, it slows down requests and compactions. You can proof what your OSD currently consumes with: ceph daemon osd.X perf dump Informative values are `db_total_bytes`, `db_used_bytes` and `slow_used_bytes`. This changes regularly because of the ongoing compactions but Prometheus mgr module exports these values such that you can track it. Small files generally leads to bigger RocksDB, especially when you use EC, but this depends on the actual amount and file sizes. I hope this helps. Regards, Michael On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl> wrote: Hi Recently our ceph cluster (nautilus) is experiencing bluefs spillovers, just 2 osd's and I disabled the warning for these osds. (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false) I'm wondering what causes this and how this can be prevented. As I understand it the rocksdb for the OSD needs to store more than fits on the NVME logical volume (123G for 12T OSD). A way to fix it could be to increase the logical volume on the nvme (if there was space on the nvme, which there isn't at the moment). This is the current size of the cluster and how much is free: [root@cephmon1 ~]# ceph df RAW STORAGE: CLASS SIZE AVAIL USED RAW USED %RAW USED hdd 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 TOTAL 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 POOLS: POOL ID STORED OBJECTS USED %USED MAX AVAIL cephfs_data 1 572 MiB 121.26M 2.4 GiB 0 167 TiB cephfs_metadata 2 56 GiB 5.15M 57 GiB 0 167 TiB cephfs_data_3copy 8 201 GiB 51.68k 602 GiB 0.09 222 TiB cephfs_data_ec83 13 643 TiB 279.75M 953 TiB 58.86 485 TiB rbd 14 21 GiB 5.66k 64 GiB 0 222 TiB .rgw.root 15 1.2 KiB 4 1 MiB 0 167 TiB default.rgw.control 16 0 B 8 0 B 0 167 TiB default.rgw.meta 17 765 B 4 1 MiB 0 167 TiB default.rgw.log 18 0 B 207 0 B 0 167 TiB cephfs_data_ec57 20 433 MiB 230 1.2 GiB 0 278 TiB The amount used can still grow a bit before we need to add nodes, but apparently we are running into the limits of our rocskdb partitions. Did we choose a parameter (e.g. minimal object size) too small, so we have too much objects on these spillover OSDs? Or is it that too many small files are stored on the cephfs filesystems? When we expand the cluster, we can choose larger nvme devices to allow larger rocksdb partitions, but is that the right way to deal with this, or should we adjust some parameters on the cluster that will reduce the rocksdb size? Cheers /Simon _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
Hi Michael, thanks for the explanation! So if I understand correctly, we waste 93 GB per OSD on unused NVME space, because only 30GB is actually used...? And to improve the space for rocksdb, we need to plan for 300GB per rocksdb partition in order to benefit from this advantage.... Reducing the number of small files is something we always ask of our users, but reality is what it is ;-) I'll have to look into how I can get an informative view on these metrics... It's pretty overwhelming the amount of information coming out of the ceph cluster, even when you look only superficially... Cheers, /Simon On 20/08/2020 10:16, Michael Bisig wrote:
Hi Simon
As far as I know, RocksDB only uses "leveled" space on the NVME partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every DB space above such a limit will automatically end up on slow devices. In your setup where you have 123GB per OSD that means you only use 30GB of fast device. The DB which spills over this limit will be offloaded to the HDD and accordingly, it slows down requests and compactions.
You can proof what your OSD currently consumes with: ceph daemon osd.X perf dump
Informative values are `db_total_bytes`, `db_used_bytes` and `slow_used_bytes`. This changes regularly because of the ongoing compactions but Prometheus mgr module exports these values such that you can track it.
Small files generally leads to bigger RocksDB, especially when you use EC, but this depends on the actual amount and file sizes.
I hope this helps. Regards, Michael
On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl> wrote:
Hi
Recently our ceph cluster (nautilus) is experiencing bluefs spillovers, just 2 osd's and I disabled the warning for these osds. (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false)
I'm wondering what causes this and how this can be prevented.
As I understand it the rocksdb for the OSD needs to store more than fits on the NVME logical volume (123G for 12T OSD). A way to fix it could be to increase the logical volume on the nvme (if there was space on the nvme, which there isn't at the moment).
This is the current size of the cluster and how much is free:
[root@cephmon1 ~]# ceph df RAW STORAGE: CLASS SIZE AVAIL USED RAW USED %RAW USED hdd 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 TOTAL 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63
POOLS: POOL ID STORED OBJECTS USED %USED MAX AVAIL cephfs_data 1 572 MiB 121.26M 2.4 GiB 0 167 TiB cephfs_metadata 2 56 GiB 5.15M 57 GiB 0 167 TiB cephfs_data_3copy 8 201 GiB 51.68k 602 GiB 0.09 222 TiB cephfs_data_ec83 13 643 TiB 279.75M 953 TiB 58.86 485 TiB rbd 14 21 GiB 5.66k 64 GiB 0 222 TiB .rgw.root 15 1.2 KiB 4 1 MiB 0 167 TiB default.rgw.control 16 0 B 8 0 B 0 167 TiB default.rgw.meta 17 765 B 4 1 MiB 0 167 TiB default.rgw.log 18 0 B 207 0 B 0 167 TiB cephfs_data_ec57 20 433 MiB 230 1.2 GiB 0 278 TiB
The amount used can still grow a bit before we need to add nodes, but apparently we are running into the limits of our rocskdb partitions.
Did we choose a parameter (e.g. minimal object size) too small, so we have too much objects on these spillover OSDs? Or is it that too many small files are stored on the cephfs filesystems?
When we expand the cluster, we can choose larger nvme devices to allow larger rocksdb partitions, but is that the right way to deal with this, or should we adjust some parameters on the cluster that will reduce the rocksdb size?
Cheers
/Simon _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
Hi Simon Unfortunately, the other NVME space is wasted or at least, this is the information we gathered during our research. This fact is due to the RocksDB level management which is explained here (https://github.com/facebook/rocksdb/wiki/Leveled-Compaction). I don't think it's a hard limit but it will be something above these values. Also consult this thread (http://lists.ceph.com/pipermail/ceph-users-ceph.com/2019-February/033286.htm...). It's probably better to go a bit over these limits to be on the safe side. Exactly, reality is always different. We also struggle with small files which lead to further problems. Accordingly, the right initial setting is pretty important and depends on your individual usecase. Regards, Michael On 20.08.20, 10:40, "Simon Oosthoek" <s.oosthoek@science.ru.nl> wrote: Hi Michael, thanks for the explanation! So if I understand correctly, we waste 93 GB per OSD on unused NVME space, because only 30GB is actually used...? And to improve the space for rocksdb, we need to plan for 300GB per rocksdb partition in order to benefit from this advantage.... Reducing the number of small files is something we always ask of our users, but reality is what it is ;-) I'll have to look into how I can get an informative view on these metrics... It's pretty overwhelming the amount of information coming out of the ceph cluster, even when you look only superficially... Cheers, /Simon On 20/08/2020 10:16, Michael Bisig wrote: > Hi Simon > > As far as I know, RocksDB only uses "leveled" space on the NVME partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every DB space above such a limit will automatically end up on slow devices. > In your setup where you have 123GB per OSD that means you only use 30GB of fast device. The DB which spills over this limit will be offloaded to the HDD and accordingly, it slows down requests and compactions. > > You can proof what your OSD currently consumes with: > ceph daemon osd.X perf dump > > Informative values are `db_total_bytes`, `db_used_bytes` and `slow_used_bytes`. This changes regularly because of the ongoing compactions but Prometheus mgr module exports these values such that you can track it. > > Small files generally leads to bigger RocksDB, especially when you use EC, but this depends on the actual amount and file sizes. > > I hope this helps. > Regards, > Michael > > On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl> wrote: > > Hi > > Recently our ceph cluster (nautilus) is experiencing bluefs spillovers, > just 2 osd's and I disabled the warning for these osds. > (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false) > > I'm wondering what causes this and how this can be prevented. > > As I understand it the rocksdb for the OSD needs to store more than fits > on the NVME logical volume (123G for 12T OSD). A way to fix it could be > to increase the logical volume on the nvme (if there was space on the > nvme, which there isn't at the moment). > > This is the current size of the cluster and how much is free: > > [root@cephmon1 ~]# ceph df > RAW STORAGE: > CLASS SIZE AVAIL USED RAW USED %RAW USED > hdd 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 > TOTAL 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 > > POOLS: > POOL ID STORED OBJECTS USED > %USED MAX AVAIL > cephfs_data 1 572 MiB 121.26M 2.4 GiB > 0 167 TiB > cephfs_metadata 2 56 GiB 5.15M 57 GiB > 0 167 TiB > cephfs_data_3copy 8 201 GiB 51.68k 602 GiB > 0.09 222 TiB > cephfs_data_ec83 13 643 TiB 279.75M 953 TiB > 58.86 485 TiB > rbd 14 21 GiB 5.66k 64 GiB > 0 222 TiB > .rgw.root 15 1.2 KiB 4 1 MiB > 0 167 TiB > default.rgw.control 16 0 B 8 0 B > 0 167 TiB > default.rgw.meta 17 765 B 4 1 MiB > 0 167 TiB > default.rgw.log 18 0 B 207 0 B > 0 167 TiB > cephfs_data_ec57 20 433 MiB 230 1.2 GiB > 0 278 TiB > > The amount used can still grow a bit before we need to add nodes, but > apparently we are running into the limits of our rocskdb partitions. > > Did we choose a parameter (e.g. minimal object size) too small, so we > have too much objects on these spillover OSDs? Or is it that too many > small files are stored on the cephfs filesystems? > > When we expand the cluster, we can choose larger nvme devices to allow > larger rocksdb partitions, but is that the right way to deal with this, > or should we adjust some parameters on the cluster that will reduce the > rocksdb size? > > Cheers > > /Simon > _______________________________________________ > ceph-users mailing list -- ceph-users@ceph.io > To unsubscribe send an email to ceph-users-leave@ceph.io > _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
Hi Michael, thanks for the pointers! This is our first production ceph cluster and we have to learn as we go... Small files is always a problem for all (networked) filesystems, usually it just trashes performance, but in this case it has another unfortunate side effect with the rocksdb :-( Cheers /Simon On 20/08/2020 11:06, Michael Bisig wrote:
Hi Simon
Unfortunately, the other NVME space is wasted or at least, this is the information we gathered during our research. This fact is due to the RocksDB level management which is explained here (https://github.com/facebook/rocksdb/wiki/Leveled-Compaction). I don't think it's a hard limit but it will be something above these values. Also consult this thread (http://lists.ceph.com/pipermail/ceph-users-ceph.com/2019-February/033286.htm...). It's probably better to go a bit over these limits to be on the safe side.
Exactly, reality is always different. We also struggle with small files which lead to further problems. Accordingly, the right initial setting is pretty important and depends on your individual usecase.
Regards, Michael
On 20.08.20, 10:40, "Simon Oosthoek" <s.oosthoek@science.ru.nl> wrote:
Hi Michael,
thanks for the explanation! So if I understand correctly, we waste 93 GB per OSD on unused NVME space, because only 30GB is actually used...?
And to improve the space for rocksdb, we need to plan for 300GB per rocksdb partition in order to benefit from this advantage....
Reducing the number of small files is something we always ask of our users, but reality is what it is ;-)
I'll have to look into how I can get an informative view on these metrics... It's pretty overwhelming the amount of information coming out of the ceph cluster, even when you look only superficially...
Cheers,
/Simon
On 20/08/2020 10:16, Michael Bisig wrote: > Hi Simon > > As far as I know, RocksDB only uses "leveled" space on the NVME partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every DB space above such a limit will automatically end up on slow devices. > In your setup where you have 123GB per OSD that means you only use 30GB of fast device. The DB which spills over this limit will be offloaded to the HDD and accordingly, it slows down requests and compactions. > > You can proof what your OSD currently consumes with: > ceph daemon osd.X perf dump > > Informative values are `db_total_bytes`, `db_used_bytes` and `slow_used_bytes`. This changes regularly because of the ongoing compactions but Prometheus mgr module exports these values such that you can track it. > > Small files generally leads to bigger RocksDB, especially when you use EC, but this depends on the actual amount and file sizes. > > I hope this helps. > Regards, > Michael > > On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl> wrote: > > Hi > > Recently our ceph cluster (nautilus) is experiencing bluefs spillovers, > just 2 osd's and I disabled the warning for these osds. > (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false) > > I'm wondering what causes this and how this can be prevented. > > As I understand it the rocksdb for the OSD needs to store more than fits > on the NVME logical volume (123G for 12T OSD). A way to fix it could be > to increase the logical volume on the nvme (if there was space on the > nvme, which there isn't at the moment). > > This is the current size of the cluster and how much is free: > > [root@cephmon1 ~]# ceph df > RAW STORAGE: > CLASS SIZE AVAIL USED RAW USED %RAW USED > hdd 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 > TOTAL 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 > > POOLS: > POOL ID STORED OBJECTS USED > %USED MAX AVAIL > cephfs_data 1 572 MiB 121.26M 2.4 GiB > 0 167 TiB > cephfs_metadata 2 56 GiB 5.15M 57 GiB > 0 167 TiB > cephfs_data_3copy 8 201 GiB 51.68k 602 GiB > 0.09 222 TiB > cephfs_data_ec83 13 643 TiB 279.75M 953 TiB > 58.86 485 TiB > rbd 14 21 GiB 5.66k 64 GiB > 0 222 TiB > .rgw.root 15 1.2 KiB 4 1 MiB > 0 167 TiB > default.rgw.control 16 0 B 8 0 B > 0 167 TiB > default.rgw.meta 17 765 B 4 1 MiB > 0 167 TiB > default.rgw.log 18 0 B 207 0 B > 0 167 TiB > cephfs_data_ec57 20 433 MiB 230 1.2 GiB > 0 278 TiB > > The amount used can still grow a bit before we need to add nodes, but > apparently we are running into the limits of our rocskdb partitions. > > Did we choose a parameter (e.g. minimal object size) too small, so we > have too much objects on these spillover OSDs? Or is it that too many > small files are stored on the cephfs filesystems? > > When we expand the cluster, we can choose larger nvme devices to allow > larger rocksdb partitions, but is that the right way to deal with this, > or should we adjust some parameters on the cluster that will reduce the > rocksdb size? > > Cheers > > /Simon > _______________________________________________ > ceph-users mailing list -- ceph-users@ceph.io > To unsubscribe send an email to ceph-users-leave@ceph.io > _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
Hi Simon, starting Nautlus v14.2.10 Bluestore is able to use 'wasted' space at DB volume. see this PR: https://github.com/ceph/ceph/pull/29687 Nice overview on the overall BlueFS/RocksDB design can be find here: https://cf2.cloudferro.com:8080/swift/v1/AUTH_5e376cddf8a94f9294259b5f48d7b2... Which also includes some overview (as well as additional concerns) for changes brought by the above-mentioned PR. Thanks, Igor On 8/20/2020 11:39 AM, Simon Oosthoek wrote:
Hi Michael,
thanks for the explanation! So if I understand correctly, we waste 93 GB per OSD on unused NVME space, because only 30GB is actually used...?
And to improve the space for rocksdb, we need to plan for 300GB per rocksdb partition in order to benefit from this advantage....
Reducing the number of small files is something we always ask of our users, but reality is what it is ;-)
I'll have to look into how I can get an informative view on these metrics... It's pretty overwhelming the amount of information coming out of the ceph cluster, even when you look only superficially...
Cheers,
/Simon
On 20/08/2020 10:16, Michael Bisig wrote:
Hi Simon
As far as I know, RocksDB only uses "leveled" space on the NVME partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every DB space above such a limit will automatically end up on slow devices. In your setup where you have 123GB per OSD that means you only use 30GB of fast device. The DB which spills over this limit will be offloaded to the HDD and accordingly, it slows down requests and compactions.
You can proof what your OSD currently consumes with: ceph daemon osd.X perf dump
Informative values are `db_total_bytes`, `db_used_bytes` and `slow_used_bytes`. This changes regularly because of the ongoing compactions but Prometheus mgr module exports these values such that you can track it.
Small files generally leads to bigger RocksDB, especially when you use EC, but this depends on the actual amount and file sizes.
I hope this helps. Regards, Michael
On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl> wrote:
Hi
Recently our ceph cluster (nautilus) is experiencing bluefs spillovers, just 2 osd's and I disabled the warning for these osds. (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false)
I'm wondering what causes this and how this can be prevented.
As I understand it the rocksdb for the OSD needs to store more than fits on the NVME logical volume (123G for 12T OSD). A way to fix it could be to increase the logical volume on the nvme (if there was space on the nvme, which there isn't at the moment).
This is the current size of the cluster and how much is free:
[root@cephmon1 ~]# ceph df RAW STORAGE: CLASS SIZE AVAIL USED RAW USED %RAW USED hdd 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 TOTAL 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63
POOLS: POOL ID STORED OBJECTS USED %USED MAX AVAIL cephfs_data 1 572 MiB 121.26M 2.4 GiB 0 167 TiB cephfs_metadata 2 56 GiB 5.15M 57 GiB 0 167 TiB cephfs_data_3copy 8 201 GiB 51.68k 602 GiB 0.09 222 TiB cephfs_data_ec83 13 643 TiB 279.75M 953 TiB 58.86 485 TiB rbd 14 21 GiB 5.66k 64 GiB 0 222 TiB .rgw.root 15 1.2 KiB 4 1 MiB 0 167 TiB default.rgw.control 16 0 B 8 0 B 0 167 TiB default.rgw.meta 17 765 B 4 1 MiB 0 167 TiB default.rgw.log 18 0 B 207 0 B 0 167 TiB cephfs_data_ec57 20 433 MiB 230 1.2 GiB 0 278 TiB
The amount used can still grow a bit before we need to add nodes, but apparently we are running into the limits of our rocskdb partitions.
Did we choose a parameter (e.g. minimal object size) too small, so we have too much objects on these spillover OSDs? Or is it that too many small files are stored on the cephfs filesystems?
When we expand the cluster, we can choose larger nvme devices to allow larger rocksdb partitions, but is that the right way to deal with this, or should we adjust some parameters on the cluster that will reduce the rocksdb size?
Cheers
/Simon _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
_______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
Hi Igor. Could you please tell why this config is in LEVEL_DEV ( https://github.com/ceph/ceph/pull/29687/files#diff-3d7a065928b2852c228ffe669... As it is documented in Ceph we can't use LEVEL_DEV in production environments! Thanks On Thu, Aug 20, 2020 at 1:58 PM Igor Fedotov <ifedotov@suse.de> wrote:
Hi Simon,
starting Nautlus v14.2.10 Bluestore is able to use 'wasted' space at DB volume.
see this PR: https://github.com/ceph/ceph/pull/29687
Nice overview on the overall BlueFS/RocksDB design can be find here:
https://cf2.cloudferro.com:8080/swift/v1/AUTH_5e376cddf8a94f9294259b5f48d7b2...
Which also includes some overview (as well as additional concerns) for changes brought by the above-mentioned PR.
Thanks,
Igor
On 8/20/2020 11:39 AM, Simon Oosthoek wrote:
Hi Michael,
thanks for the explanation! So if I understand correctly, we waste 93 GB per OSD on unused NVME space, because only 30GB is actually used...?
And to improve the space for rocksdb, we need to plan for 300GB per rocksdb partition in order to benefit from this advantage....
Reducing the number of small files is something we always ask of our users, but reality is what it is ;-)
I'll have to look into how I can get an informative view on these metrics... It's pretty overwhelming the amount of information coming out of the ceph cluster, even when you look only superficially...
Cheers,
/Simon
On 20/08/2020 10:16, Michael Bisig wrote:
Hi Simon
As far as I know, RocksDB only uses "leveled" space on the NVME partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every DB space above such a limit will automatically end up on slow devices. In your setup where you have 123GB per OSD that means you only use 30GB of fast device. The DB which spills over this limit will be offloaded to the HDD and accordingly, it slows down requests and compactions.
You can proof what your OSD currently consumes with: ceph daemon osd.X perf dump
Informative values are `db_total_bytes`, `db_used_bytes` and `slow_used_bytes`. This changes regularly because of the ongoing compactions but Prometheus mgr module exports these values such that you can track it.
Small files generally leads to bigger RocksDB, especially when you use EC, but this depends on the actual amount and file sizes.
I hope this helps. Regards, Michael
On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl> wrote:
Hi
Recently our ceph cluster (nautilus) is experiencing bluefs spillovers, just 2 osd's and I disabled the warning for these osds. (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false)
I'm wondering what causes this and how this can be prevented.
As I understand it the rocksdb for the OSD needs to store more than fits on the NVME logical volume (123G for 12T OSD). A way to fix it could be to increase the logical volume on the nvme (if there was space on the nvme, which there isn't at the moment).
This is the current size of the cluster and how much is free:
[root@cephmon1 ~]# ceph df RAW STORAGE: CLASS SIZE AVAIL USED RAW USED %RAW USED hdd 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 TOTAL 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63
POOLS: POOL ID STORED OBJECTS USED %USED MAX AVAIL cephfs_data 1 572 MiB 121.26M 2.4 GiB 0 167 TiB cephfs_metadata 2 56 GiB 5.15M 57 GiB 0 167 TiB cephfs_data_3copy 8 201 GiB 51.68k 602 GiB 0.09 222 TiB cephfs_data_ec83 13 643 TiB 279.75M 953 TiB 58.86 485 TiB rbd 14 21 GiB 5.66k 64 GiB 0 222 TiB .rgw.root 15 1.2 KiB 4 1 MiB 0 167 TiB default.rgw.control 16 0 B 8 0 B 0 167 TiB default.rgw.meta 17 765 B 4 1 MiB 0 167 TiB default.rgw.log 18 0 B 207 0 B 0 167 TiB cephfs_data_ec57 20 433 MiB 230 1.2 GiB 0 278 TiB
The amount used can still grow a bit before we need to add nodes, but apparently we are running into the limits of our rocskdb partitions.
Did we choose a parameter (e.g. minimal object size) too small, so we have too much objects on these spillover OSDs? Or is it that too many small files are stored on the cephfs filesystems?
When we expand the cluster, we can choose larger nvme devices to allow larger rocksdb partitions, but is that the right way to deal with this, or should we adjust some parameters on the cluster that will reduce the rocksdb size?
Cheers
/Simon _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
_______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
Hi Seena, this parameter isn't intended to be adjusted in production environments - it's supposed that default behavior covers all regular customers' needs. The issue though is that default setting is invalid. It should be 'use_some_extra'. Gonna fix that shortly... Thanks, Igor On 8/20/2020 1:44 PM, Seena Fallah wrote:
Hi Igor.
Could you please tell why this config is in LEVEL_DEV (https://github.com/ceph/ceph/pull/29687/files#diff-3d7a065928b2852c228ffe669... As it is documented in Ceph we can't use LEVEL_DEV in production environments!
Thanks
On Thu, Aug 20, 2020 at 1:58 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
Hi Simon,
starting Nautlus v14.2.10 Bluestore is able to use 'wasted' space at DB volume.
see this PR: https://github.com/ceph/ceph/pull/29687
Nice overview on the overall BlueFS/RocksDB design can be find here:
https://cf2.cloudferro.com:8080/swift/v1/AUTH_5e376cddf8a94f9294259b5f48d7b2...
Which also includes some overview (as well as additional concerns) for changes brought by the above-mentioned PR.
Thanks,
Igor
On 8/20/2020 11:39 AM, Simon Oosthoek wrote: > Hi Michael, > > thanks for the explanation! So if I understand correctly, we waste 93 > GB per OSD on unused NVME space, because only 30GB is actually used...? > > And to improve the space for rocksdb, we need to plan for 300GB per > rocksdb partition in order to benefit from this advantage.... > > Reducing the number of small files is something we always ask of our > users, but reality is what it is ;-) > > I'll have to look into how I can get an informative view on these > metrics... It's pretty overwhelming the amount of information coming > out of the ceph cluster, even when you look only superficially... > > Cheers, > > /Simon > > On 20/08/2020 10:16, Michael Bisig wrote: >> Hi Simon >> >> As far as I know, RocksDB only uses "leveled" space on the NVME >> partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every >> DB space above such a limit will automatically end up on slow devices. >> In your setup where you have 123GB per OSD that means you only use >> 30GB of fast device. The DB which spills over this limit will be >> offloaded to the HDD and accordingly, it slows down requests and >> compactions. >> >> You can proof what your OSD currently consumes with: >> ceph daemon osd.X perf dump >> >> Informative values are `db_total_bytes`, `db_used_bytes` and >> `slow_used_bytes`. This changes regularly because of the ongoing >> compactions but Prometheus mgr module exports these values such that >> you can track it. >> >> Small files generally leads to bigger RocksDB, especially when you >> use EC, but this depends on the actual amount and file sizes. >> >> I hope this helps. >> Regards, >> Michael >> >> On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl <mailto:s.oosthoek@science.ru.nl>> wrote: >> >> Hi >> >> Recently our ceph cluster (nautilus) is experiencing bluefs >> spillovers, >> just 2 osd's and I disabled the warning for these osds. >> (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false) >> >> I'm wondering what causes this and how this can be prevented. >> >> As I understand it the rocksdb for the OSD needs to store more >> than fits >> on the NVME logical volume (123G for 12T OSD). A way to fix it >> could be >> to increase the logical volume on the nvme (if there was space >> on the >> nvme, which there isn't at the moment). >> >> This is the current size of the cluster and how much is free: >> >> [root@cephmon1 ~]# ceph df >> RAW STORAGE: >> CLASS SIZE AVAIL USED RAW USED >> %RAW USED >> hdd 1.8 PiB 842 TiB 974 TiB 974 >> TiB 53.63 >> TOTAL 1.8 PiB 842 TiB 974 TiB 974 >> TiB 53.63 >> >> POOLS: >> POOL ID STORED OBJECTS USED >> %USED MAX AVAIL >> cephfs_data 1 572 MiB 121.26M 2.4 GiB >> 0 167 TiB >> cephfs_metadata 2 56 GiB 5.15M 57 GiB >> 0 167 TiB >> cephfs_data_3copy 8 201 GiB 51.68k 602 GiB >> 0.09 222 TiB >> cephfs_data_ec83 13 643 TiB 279.75M 953 TiB >> 58.86 485 TiB >> rbd 14 21 GiB 5.66k 64 GiB >> 0 222 TiB >> .rgw.root 15 1.2 KiB 4 1 MiB >> 0 167 TiB >> default.rgw.control 16 0 B 8 0 B >> 0 167 TiB >> default.rgw.meta 17 765 B 4 1 MiB >> 0 167 TiB >> default.rgw.log 18 0 B 207 0 B >> 0 167 TiB >> cephfs_data_ec57 20 433 MiB 230 1.2 GiB >> 0 278 TiB >> >> The amount used can still grow a bit before we need to add >> nodes, but >> apparently we are running into the limits of our rocskdb >> partitions. >> >> Did we choose a parameter (e.g. minimal object size) too small, >> so we >> have too much objects on these spillover OSDs? Or is it that too >> many >> small files are stored on the cephfs filesystems? >> >> When we expand the cluster, we can choose larger nvme devices to >> allow >> larger rocksdb partitions, but is that the right way to deal >> with this, >> or should we adjust some parameters on the cluster that will >> reduce the >> rocksdb size? >> >> Cheers >> >> /Simon >> _______________________________________________ >> ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> >> To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io> >> > _______________________________________________ > ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> > To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io> _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io>
Greate, thanks. Is it safe to change it manually in ceph.conf until next nautilus release or should I wait for the next nautilus release for this change? I mean does qa run on this value for this config that we could trust and change it or should we wait until the next nautilus release that qa ran on this value? On Thu, Aug 20, 2020 at 5:25 PM Igor Fedotov <ifedotov@suse.de> wrote:
Hi Seena,
this parameter isn't intended to be adjusted in production environments - it's supposed that default behavior covers all regular customers' needs.
The issue though is that default setting is invalid. It should be 'use_some_extra'. Gonna fix that shortly...
Thanks,
Igor
On 8/20/2020 1:44 PM, Seena Fallah wrote:
Hi Igor.
Could you please tell why this config is in LEVEL_DEV ( https://github.com/ceph/ceph/pull/29687/files#diff-3d7a065928b2852c228ffe669... As it is documented in Ceph we can't use LEVEL_DEV in production environments!
Thanks
On Thu, Aug 20, 2020 at 1:58 PM Igor Fedotov <ifedotov@suse.de> wrote:
Hi Simon,
starting Nautlus v14.2.10 Bluestore is able to use 'wasted' space at DB volume.
see this PR: https://github.com/ceph/ceph/pull/29687
Nice overview on the overall BlueFS/RocksDB design can be find here:
https://cf2.cloudferro.com:8080/swift/v1/AUTH_5e376cddf8a94f9294259b5f48d7b2...
Which also includes some overview (as well as additional concerns) for changes brought by the above-mentioned PR.
Thanks,
Igor
On 8/20/2020 11:39 AM, Simon Oosthoek wrote:
Hi Michael,
thanks for the explanation! So if I understand correctly, we waste 93 GB per OSD on unused NVME space, because only 30GB is actually used...?
And to improve the space for rocksdb, we need to plan for 300GB per rocksdb partition in order to benefit from this advantage....
Reducing the number of small files is something we always ask of our users, but reality is what it is ;-)
I'll have to look into how I can get an informative view on these metrics... It's pretty overwhelming the amount of information coming out of the ceph cluster, even when you look only superficially...
Cheers,
/Simon
On 20/08/2020 10:16, Michael Bisig wrote:
Hi Simon
As far as I know, RocksDB only uses "leveled" space on the NVME partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every DB space above such a limit will automatically end up on slow devices. In your setup where you have 123GB per OSD that means you only use 30GB of fast device. The DB which spills over this limit will be offloaded to the HDD and accordingly, it slows down requests and compactions.
You can proof what your OSD currently consumes with: ceph daemon osd.X perf dump
Informative values are `db_total_bytes`, `db_used_bytes` and `slow_used_bytes`. This changes regularly because of the ongoing compactions but Prometheus mgr module exports these values such that you can track it.
Small files generally leads to bigger RocksDB, especially when you use EC, but this depends on the actual amount and file sizes.
I hope this helps. Regards, Michael
On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl> wrote:
Hi
Recently our ceph cluster (nautilus) is experiencing bluefs spillovers, just 2 osd's and I disabled the warning for these osds. (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false)
I'm wondering what causes this and how this can be prevented.
As I understand it the rocksdb for the OSD needs to store more than fits on the NVME logical volume (123G for 12T OSD). A way to fix it could be to increase the logical volume on the nvme (if there was space on the nvme, which there isn't at the moment).
This is the current size of the cluster and how much is free:
[root@cephmon1 ~]# ceph df RAW STORAGE: CLASS SIZE AVAIL USED RAW USED %RAW USED hdd 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 TOTAL 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63
POOLS: POOL ID STORED OBJECTS USED %USED MAX AVAIL cephfs_data 1 572 MiB 121.26M 2.4 GiB 0 167 TiB cephfs_metadata 2 56 GiB 5.15M 57 GiB 0 167 TiB cephfs_data_3copy 8 201 GiB 51.68k 602 GiB 0.09 222 TiB cephfs_data_ec83 13 643 TiB 279.75M 953 TiB 58.86 485 TiB rbd 14 21 GiB 5.66k 64 GiB 0 222 TiB .rgw.root 15 1.2 KiB 4 1 MiB 0 167 TiB default.rgw.control 16 0 B 8 0 B 0 167 TiB default.rgw.meta 17 765 B 4 1 MiB 0 167 TiB default.rgw.log 18 0 B 207 0 B 0 167 TiB cephfs_data_ec57 20 433 MiB 230 1.2 GiB 0 278 TiB
The amount used can still grow a bit before we need to add nodes, but apparently we are running into the limits of our rocskdb partitions.
Did we choose a parameter (e.g. minimal object size) too small, so we have too much objects on these spillover OSDs? Or is it that too many small files are stored on the cephfs filesystems?
When we expand the cluster, we can choose larger nvme devices to allow larger rocksdb partitions, but is that the right way to deal with this, or should we adjust some parameters on the cluster that will reduce the rocksdb size?
Cheers
/Simon _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
_______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
From technical/developer's point of view I don't see any issues with tuning this option. But since now I wouldn't recommend to enable it in production as it partially bypassed our regular development cycle. Being enabled in master for a while by default allows more develpers to use/try the feature before release. This can be considered as an additional implicit QA process. But as we just discovered this hasn't happened. Hence you can definitely try it but this exposes your cluster(s) to some risk as for any new (and incompletely tested) feature.... Thanks, Igor On 8/20/2020 4:06 PM, Seena Fallah wrote:
Greate, thanks.
Is it safe to change it manually in ceph.conf until next nautilus release or should I wait for the next nautilus release for this change? I mean does qa run on this value for this config that we could trust and change it or should we wait until the next nautilus release that qa ran on this value?
On Thu, Aug 20, 2020 at 5:25 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
Hi Seena,
this parameter isn't intended to be adjusted in production environments - it's supposed that default behavior covers all regular customers' needs.
The issue though is that default setting is invalid. It should be 'use_some_extra'. Gonna fix that shortly...
Thanks,
Igor
On 8/20/2020 1:44 PM, Seena Fallah wrote:
Hi Igor.
Could you please tell why this config is in LEVEL_DEV (https://github.com/ceph/ceph/pull/29687/files#diff-3d7a065928b2852c228ffe669... As it is documented in Ceph we can't use LEVEL_DEV in production environments!
Thanks
On Thu, Aug 20, 2020 at 1:58 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
Hi Simon,
starting Nautlus v14.2.10 Bluestore is able to use 'wasted' space at DB volume.
see this PR: https://github.com/ceph/ceph/pull/29687
Nice overview on the overall BlueFS/RocksDB design can be find here:
https://cf2.cloudferro.com:8080/swift/v1/AUTH_5e376cddf8a94f9294259b5f48d7b2...
Which also includes some overview (as well as additional concerns) for changes brought by the above-mentioned PR.
Thanks,
Igor
On 8/20/2020 11:39 AM, Simon Oosthoek wrote: > Hi Michael, > > thanks for the explanation! So if I understand correctly, we waste 93 > GB per OSD on unused NVME space, because only 30GB is actually used...? > > And to improve the space for rocksdb, we need to plan for 300GB per > rocksdb partition in order to benefit from this advantage.... > > Reducing the number of small files is something we always ask of our > users, but reality is what it is ;-) > > I'll have to look into how I can get an informative view on these > metrics... It's pretty overwhelming the amount of information coming > out of the ceph cluster, even when you look only superficially... > > Cheers, > > /Simon > > On 20/08/2020 10:16, Michael Bisig wrote: >> Hi Simon >> >> As far as I know, RocksDB only uses "leveled" space on the NVME >> partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every >> DB space above such a limit will automatically end up on slow devices. >> In your setup where you have 123GB per OSD that means you only use >> 30GB of fast device. The DB which spills over this limit will be >> offloaded to the HDD and accordingly, it slows down requests and >> compactions. >> >> You can proof what your OSD currently consumes with: >> ceph daemon osd.X perf dump >> >> Informative values are `db_total_bytes`, `db_used_bytes` and >> `slow_used_bytes`. This changes regularly because of the ongoing >> compactions but Prometheus mgr module exports these values such that >> you can track it. >> >> Small files generally leads to bigger RocksDB, especially when you >> use EC, but this depends on the actual amount and file sizes. >> >> I hope this helps. >> Regards, >> Michael >> >> On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl <mailto:s.oosthoek@science.ru.nl>> wrote: >> >> Hi >> >> Recently our ceph cluster (nautilus) is experiencing bluefs >> spillovers, >> just 2 osd's and I disabled the warning for these osds. >> (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false) >> >> I'm wondering what causes this and how this can be prevented. >> >> As I understand it the rocksdb for the OSD needs to store more >> than fits >> on the NVME logical volume (123G for 12T OSD). A way to fix it >> could be >> to increase the logical volume on the nvme (if there was space >> on the >> nvme, which there isn't at the moment). >> >> This is the current size of the cluster and how much is free: >> >> [root@cephmon1 ~]# ceph df >> RAW STORAGE: >> CLASS SIZE AVAIL USED RAW USED >> %RAW USED >> hdd 1.8 PiB 842 TiB 974 TiB 974 >> TiB 53.63 >> TOTAL 1.8 PiB 842 TiB 974 TiB 974 >> TiB 53.63 >> >> POOLS: >> POOL ID STORED OBJECTS USED >> %USED MAX AVAIL >> cephfs_data 1 572 MiB 121.26M 2.4 GiB >> 0 167 TiB >> cephfs_metadata 2 56 GiB 5.15M 57 GiB >> 0 167 TiB >> cephfs_data_3copy 8 201 GiB 51.68k 602 GiB >> 0.09 222 TiB >> cephfs_data_ec83 13 643 TiB 279.75M 953 TiB >> 58.86 485 TiB >> rbd 14 21 GiB 5.66k 64 GiB >> 0 222 TiB >> .rgw.root 15 1.2 KiB 4 1 MiB >> 0 167 TiB >> default.rgw.control 16 0 B 8 0 B >> 0 167 TiB >> default.rgw.meta 17 765 B 4 1 MiB >> 0 167 TiB >> default.rgw.log 18 0 B 207 0 B >> 0 167 TiB >> cephfs_data_ec57 20 433 MiB 230 1.2 GiB >> 0 278 TiB >> >> The amount used can still grow a bit before we need to add >> nodes, but >> apparently we are running into the limits of our rocskdb >> partitions. >> >> Did we choose a parameter (e.g. minimal object size) too small, >> so we >> have too much objects on these spillover OSDs? Or is it that too >> many >> small files are stored on the cephfs filesystems? >> >> When we expand the cluster, we can choose larger nvme devices to >> allow >> larger rocksdb partitions, but is that the right way to deal >> with this, >> or should we adjust some parameters on the cluster that will >> reduce the >> rocksdb size? >> >> Cheers >> >> /Simon >> _______________________________________________ >> ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> >> To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io> >> > _______________________________________________ > ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> > To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io> _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io>
So you won't backport it to nautilus until it gets default to master for a while? On Thu, Aug 20, 2020 at 6:00 PM Igor Fedotov <ifedotov@suse.de> wrote:
From technical/developer's point of view I don't see any issues with tuning this option. But since now I wouldn't recommend to enable it in production as it partially bypassed our regular development cycle. Being enabled in master for a while by default allows more develpers to use/try the feature before release. This can be considered as an additional implicit QA process. But as we just discovered this hasn't happened.
Hence you can definitely try it but this exposes your cluster(s) to some risk as for any new (and incompletely tested) feature....
Thanks,
Igor
On 8/20/2020 4:06 PM, Seena Fallah wrote:
Greate, thanks.
Is it safe to change it manually in ceph.conf until next nautilus release or should I wait for the next nautilus release for this change? I mean does qa run on this value for this config that we could trust and change it or should we wait until the next nautilus release that qa ran on this value?
On Thu, Aug 20, 2020 at 5:25 PM Igor Fedotov <ifedotov@suse.de> wrote:
Hi Seena,
this parameter isn't intended to be adjusted in production environments - it's supposed that default behavior covers all regular customers' needs.
The issue though is that default setting is invalid. It should be 'use_some_extra'. Gonna fix that shortly...
Thanks,
Igor
On 8/20/2020 1:44 PM, Seena Fallah wrote:
Hi Igor.
Could you please tell why this config is in LEVEL_DEV ( https://github.com/ceph/ceph/pull/29687/files#diff-3d7a065928b2852c228ffe669... As it is documented in Ceph we can't use LEVEL_DEV in production environments!
Thanks
On Thu, Aug 20, 2020 at 1:58 PM Igor Fedotov <ifedotov@suse.de> wrote:
Hi Simon,
starting Nautlus v14.2.10 Bluestore is able to use 'wasted' space at DB volume.
see this PR: https://github.com/ceph/ceph/pull/29687
Nice overview on the overall BlueFS/RocksDB design can be find here:
https://cf2.cloudferro.com:8080/swift/v1/AUTH_5e376cddf8a94f9294259b5f48d7b2...
Which also includes some overview (as well as additional concerns) for changes brought by the above-mentioned PR.
Thanks,
Igor
On 8/20/2020 11:39 AM, Simon Oosthoek wrote:
Hi Michael,
thanks for the explanation! So if I understand correctly, we waste 93 GB per OSD on unused NVME space, because only 30GB is actually used...?
And to improve the space for rocksdb, we need to plan for 300GB per rocksdb partition in order to benefit from this advantage....
Reducing the number of small files is something we always ask of our users, but reality is what it is ;-)
I'll have to look into how I can get an informative view on these metrics... It's pretty overwhelming the amount of information coming out of the ceph cluster, even when you look only superficially...
Cheers,
/Simon
On 20/08/2020 10:16, Michael Bisig wrote:
Hi Simon
As far as I know, RocksDB only uses "leveled" space on the NVME partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every DB space above such a limit will automatically end up on slow devices. In your setup where you have 123GB per OSD that means you only use 30GB of fast device. The DB which spills over this limit will be offloaded to the HDD and accordingly, it slows down requests and compactions.
You can proof what your OSD currently consumes with: ceph daemon osd.X perf dump
Informative values are `db_total_bytes`, `db_used_bytes` and `slow_used_bytes`. This changes regularly because of the ongoing compactions but Prometheus mgr module exports these values such that you can track it.
Small files generally leads to bigger RocksDB, especially when you use EC, but this depends on the actual amount and file sizes.
I hope this helps. Regards, Michael
On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl> wrote:
Hi
Recently our ceph cluster (nautilus) is experiencing bluefs spillovers, just 2 osd's and I disabled the warning for these osds. (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false)
I'm wondering what causes this and how this can be prevented.
As I understand it the rocksdb for the OSD needs to store more than fits on the NVME logical volume (123G for 12T OSD). A way to fix it could be to increase the logical volume on the nvme (if there was space on the nvme, which there isn't at the moment).
This is the current size of the cluster and how much is free:
[root@cephmon1 ~]# ceph df RAW STORAGE: CLASS SIZE AVAIL USED RAW USED %RAW USED hdd 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 TOTAL 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63
POOLS: POOL ID STORED OBJECTS USED %USED MAX AVAIL cephfs_data 1 572 MiB 121.26M 2.4 GiB 0 167 TiB cephfs_metadata 2 56 GiB 5.15M 57 GiB 0 167 TiB cephfs_data_3copy 8 201 GiB 51.68k 602 GiB 0.09 222 TiB cephfs_data_ec83 13 643 TiB 279.75M 953 TiB 58.86 485 TiB rbd 14 21 GiB 5.66k 64 GiB 0 222 TiB .rgw.root 15 1.2 KiB 4 1 MiB 0 167 TiB default.rgw.control 16 0 B 8 0 B 0 167 TiB default.rgw.meta 17 765 B 4 1 MiB 0 167 TiB default.rgw.log 18 0 B 207 0 B 0 167 TiB cephfs_data_ec57 20 433 MiB 230 1.2 GiB 0 278 TiB
The amount used can still grow a bit before we need to add nodes, but apparently we are running into the limits of our rocskdb partitions.
Did we choose a parameter (e.g. minimal object size) too small, so we have too much objects on these spillover OSDs? Or is it that too many small files are stored on the cephfs filesystems?
When we expand the cluster, we can choose larger nvme devices to allow larger rocksdb partitions, but is that the right way to deal with this, or should we adjust some parameters on the cluster that will reduce the rocksdb size?
Cheers
/Simon _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
_______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
Correct. On 8/20/2020 5:15 PM, Seena Fallah wrote:
So you won't backport it to nautilus until it gets default to master for a while?
On Thu, Aug 20, 2020 at 6:00 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
From technical/developer's point of view I don't see any issues with tuning this option. But since now I wouldn't recommend to enable it in production as it partially bypassed our regular development cycle. Being enabled in master for a while by default allows more develpers to use/try the feature before release. This can be considered as an additional implicit QA process. But as we just discovered this hasn't happened.
Hence you can definitely try it but this exposes your cluster(s) to some risk as for any new (and incompletely tested) feature....
Thanks,
Igor
On 8/20/2020 4:06 PM, Seena Fallah wrote:
Greate, thanks.
Is it safe to change it manually in ceph.conf until next nautilus release or should I wait for the next nautilus release for this change? I mean does qa run on this value for this config that we could trust and change it or should we wait until the next nautilus release that qa ran on this value?
On Thu, Aug 20, 2020 at 5:25 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
Hi Seena,
this parameter isn't intended to be adjusted in production environments - it's supposed that default behavior covers all regular customers' needs.
The issue though is that default setting is invalid. It should be 'use_some_extra'. Gonna fix that shortly...
Thanks,
Igor
On 8/20/2020 1:44 PM, Seena Fallah wrote:
Hi Igor.
Could you please tell why this config is in LEVEL_DEV (https://github.com/ceph/ceph/pull/29687/files#diff-3d7a065928b2852c228ffe669... As it is documented in Ceph we can't use LEVEL_DEV in production environments!
Thanks
On Thu, Aug 20, 2020 at 1:58 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
Hi Simon,
starting Nautlus v14.2.10 Bluestore is able to use 'wasted' space at DB volume.
see this PR: https://github.com/ceph/ceph/pull/29687
Nice overview on the overall BlueFS/RocksDB design can be find here:
https://cf2.cloudferro.com:8080/swift/v1/AUTH_5e376cddf8a94f9294259b5f48d7b2...
Which also includes some overview (as well as additional concerns) for changes brought by the above-mentioned PR.
Thanks,
Igor
On 8/20/2020 11:39 AM, Simon Oosthoek wrote: > Hi Michael, > > thanks for the explanation! So if I understand correctly, we waste 93 > GB per OSD on unused NVME space, because only 30GB is actually used...? > > And to improve the space for rocksdb, we need to plan for 300GB per > rocksdb partition in order to benefit from this advantage.... > > Reducing the number of small files is something we always ask of our > users, but reality is what it is ;-) > > I'll have to look into how I can get an informative view on these > metrics... It's pretty overwhelming the amount of information coming > out of the ceph cluster, even when you look only superficially... > > Cheers, > > /Simon > > On 20/08/2020 10:16, Michael Bisig wrote: >> Hi Simon >> >> As far as I know, RocksDB only uses "leveled" space on the NVME >> partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every >> DB space above such a limit will automatically end up on slow devices. >> In your setup where you have 123GB per OSD that means you only use >> 30GB of fast device. The DB which spills over this limit will be >> offloaded to the HDD and accordingly, it slows down requests and >> compactions. >> >> You can proof what your OSD currently consumes with: >> ceph daemon osd.X perf dump >> >> Informative values are `db_total_bytes`, `db_used_bytes` and >> `slow_used_bytes`. This changes regularly because of the ongoing >> compactions but Prometheus mgr module exports these values such that >> you can track it. >> >> Small files generally leads to bigger RocksDB, especially when you >> use EC, but this depends on the actual amount and file sizes. >> >> I hope this helps. >> Regards, >> Michael >> >> On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl <mailto:s.oosthoek@science.ru.nl>> wrote: >> >> Hi >> >> Recently our ceph cluster (nautilus) is experiencing bluefs >> spillovers, >> just 2 osd's and I disabled the warning for these osds. >> (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false) >> >> I'm wondering what causes this and how this can be prevented. >> >> As I understand it the rocksdb for the OSD needs to store more >> than fits >> on the NVME logical volume (123G for 12T OSD). A way to fix it >> could be >> to increase the logical volume on the nvme (if there was space >> on the >> nvme, which there isn't at the moment). >> >> This is the current size of the cluster and how much is free: >> >> [root@cephmon1 ~]# ceph df >> RAW STORAGE: >> CLASS SIZE AVAIL USED RAW USED >> %RAW USED >> hdd 1.8 PiB 842 TiB 974 TiB 974 >> TiB 53.63 >> TOTAL 1.8 PiB 842 TiB 974 TiB 974 >> TiB 53.63 >> >> POOLS: >> POOL ID STORED OBJECTS USED >> %USED MAX AVAIL >> cephfs_data 1 572 MiB 121.26M 2.4 GiB >> 0 167 TiB >> cephfs_metadata 2 56 GiB 5.15M 57 GiB >> 0 167 TiB >> cephfs_data_3copy 8 201 GiB 51.68k 602 GiB >> 0.09 222 TiB >> cephfs_data_ec83 13 643 TiB 279.75M 953 TiB >> 58.86 485 TiB >> rbd 14 21 GiB 5.66k 64 GiB >> 0 222 TiB >> .rgw.root 15 1.2 KiB 4 1 MiB >> 0 167 TiB >> default.rgw.control 16 0 B 8 0 B >> 0 167 TiB >> default.rgw.meta 17 765 B 4 1 MiB >> 0 167 TiB >> default.rgw.log 18 0 B 207 0 B >> 0 167 TiB >> cephfs_data_ec57 20 433 MiB 230 1.2 GiB >> 0 278 TiB >> >> The amount used can still grow a bit before we need to add >> nodes, but >> apparently we are running into the limits of our rocskdb >> partitions. >> >> Did we choose a parameter (e.g. minimal object size) too small, >> so we >> have too much objects on these spillover OSDs? Or is it that too >> many >> small files are stored on the cephfs filesystems? >> >> When we expand the cluster, we can choose larger nvme devices to >> allow >> larger rocksdb partitions, but is that the right way to deal >> with this, >> or should we adjust some parameters on the cluster that will >> reduce the >> rocksdb size? >> >> Cheers >> >> /Simon >> _______________________________________________ >> ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> >> To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io> >> > _______________________________________________ > ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> > To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io> _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io>
So what do you suggest for a short term solution? (I think you won't backport it to nautilus at least about 6 month) Changing db size is too expensive because I should buy new NVME devices with double size and also redeploy all my OSDs. Manual compaction will still have an impact on performance and doing it for a month doesn't look very good! On Thu, Aug 20, 2020 at 6:52 PM Igor Fedotov <ifedotov@suse.de> wrote:
Correct. On 8/20/2020 5:15 PM, Seena Fallah wrote:
So you won't backport it to nautilus until it gets default to master for a while?
On Thu, Aug 20, 2020 at 6:00 PM Igor Fedotov <ifedotov@suse.de> wrote:
From technical/developer's point of view I don't see any issues with tuning this option. But since now I wouldn't recommend to enable it in production as it partially bypassed our regular development cycle. Being enabled in master for a while by default allows more develpers to use/try the feature before release. This can be considered as an additional implicit QA process. But as we just discovered this hasn't happened.
Hence you can definitely try it but this exposes your cluster(s) to some risk as for any new (and incompletely tested) feature....
Thanks,
Igor
On 8/20/2020 4:06 PM, Seena Fallah wrote:
Greate, thanks.
Is it safe to change it manually in ceph.conf until next nautilus release or should I wait for the next nautilus release for this change? I mean does qa run on this value for this config that we could trust and change it or should we wait until the next nautilus release that qa ran on this value?
On Thu, Aug 20, 2020 at 5:25 PM Igor Fedotov <ifedotov@suse.de> wrote:
Hi Seena,
this parameter isn't intended to be adjusted in production environments - it's supposed that default behavior covers all regular customers' needs.
The issue though is that default setting is invalid. It should be 'use_some_extra'. Gonna fix that shortly...
Thanks,
Igor
On 8/20/2020 1:44 PM, Seena Fallah wrote:
Hi Igor.
Could you please tell why this config is in LEVEL_DEV ( https://github.com/ceph/ceph/pull/29687/files#diff-3d7a065928b2852c228ffe669... As it is documented in Ceph we can't use LEVEL_DEV in production environments!
Thanks
On Thu, Aug 20, 2020 at 1:58 PM Igor Fedotov <ifedotov@suse.de> wrote:
Hi Simon,
starting Nautlus v14.2.10 Bluestore is able to use 'wasted' space at DB volume.
see this PR: https://github.com/ceph/ceph/pull/29687
Nice overview on the overall BlueFS/RocksDB design can be find here:
https://cf2.cloudferro.com:8080/swift/v1/AUTH_5e376cddf8a94f9294259b5f48d7b2...
Which also includes some overview (as well as additional concerns) for changes brought by the above-mentioned PR.
Thanks,
Igor
On 8/20/2020 11:39 AM, Simon Oosthoek wrote:
Hi Michael,
thanks for the explanation! So if I understand correctly, we waste 93 GB per OSD on unused NVME space, because only 30GB is actually used...?
And to improve the space for rocksdb, we need to plan for 300GB per rocksdb partition in order to benefit from this advantage....
Reducing the number of small files is something we always ask of our users, but reality is what it is ;-)
I'll have to look into how I can get an informative view on these metrics... It's pretty overwhelming the amount of information coming out of the ceph cluster, even when you look only superficially...
Cheers,
/Simon
On 20/08/2020 10:16, Michael Bisig wrote:
Hi Simon
As far as I know, RocksDB only uses "leveled" space on the NVME partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every DB space above such a limit will automatically end up on slow devices. In your setup where you have 123GB per OSD that means you only use 30GB of fast device. The DB which spills over this limit will be offloaded to the HDD and accordingly, it slows down requests and compactions.
You can proof what your OSD currently consumes with: ceph daemon osd.X perf dump
Informative values are `db_total_bytes`, `db_used_bytes` and `slow_used_bytes`. This changes regularly because of the ongoing compactions but Prometheus mgr module exports these values such that you can track it.
Small files generally leads to bigger RocksDB, especially when you use EC, but this depends on the actual amount and file sizes.
I hope this helps. Regards, Michael
On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl> wrote:
Hi
Recently our ceph cluster (nautilus) is experiencing bluefs spillovers, just 2 osd's and I disabled the warning for these osds. (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false)
I'm wondering what causes this and how this can be prevented.
As I understand it the rocksdb for the OSD needs to store more than fits on the NVME logical volume (123G for 12T OSD). A way to fix it could be to increase the logical volume on the nvme (if there was space on the nvme, which there isn't at the moment).
This is the current size of the cluster and how much is free:
[root@cephmon1 ~]# ceph df RAW STORAGE: CLASS SIZE AVAIL USED RAW USED %RAW USED hdd 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63 TOTAL 1.8 PiB 842 TiB 974 TiB 974 TiB 53.63
POOLS: POOL ID STORED OBJECTS USED %USED MAX AVAIL cephfs_data 1 572 MiB 121.26M 2.4 GiB 0 167 TiB cephfs_metadata 2 56 GiB 5.15M 57 GiB 0 167 TiB cephfs_data_3copy 8 201 GiB 51.68k 602 GiB 0.09 222 TiB cephfs_data_ec83 13 643 TiB 279.75M 953 TiB 58.86 485 TiB rbd 14 21 GiB 5.66k 64 GiB 0 222 TiB .rgw.root 15 1.2 KiB 4 1 MiB 0 167 TiB default.rgw.control 16 0 B 8 0 B 0 167 TiB default.rgw.meta 17 765 B 4 1 MiB 0 167 TiB default.rgw.log 18 0 B 207 0 B 0 167 TiB cephfs_data_ec57 20 433 MiB 230 1.2 GiB 0 278 TiB
The amount used can still grow a bit before we need to add nodes, but apparently we are running into the limits of our rocskdb partitions.
Did we choose a parameter (e.g. minimal object size) too small, so we have too much objects on these spillover OSDs? Or is it that too many small files are stored on the cephfs filesystems?
When we expand the cluster, we can choose larger nvme devices to allow larger rocksdb partitions, but is that the right way to deal with this, or should we adjust some parameters on the cluster that will reduce the rocksdb size?
Cheers
/Simon _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
_______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
Honestly I don't have any perfect solution for now. If this is urgent you probably better to proceed with enabling the new DB space management feature. But please do that eventually, modify 1-2 OSDs at the first stage and test them for some period (may be a week or two). Thanks, Igor On 8/20/2020 5:36 PM, Seena Fallah wrote:
So what do you suggest for a short term solution? (I think you won't backport it to nautilus at least about 6 month)
Changing db size is too expensive because I should buy new NVME devices with double size and also redeploy all my OSDs. Manual compaction will still have an impact on performance and doing it for a month doesn't look very good!
On Thu, Aug 20, 2020 at 6:52 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
Correct.
On 8/20/2020 5:15 PM, Seena Fallah wrote:
So you won't backport it to nautilus until it gets default to master for a while?
On Thu, Aug 20, 2020 at 6:00 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
From technical/developer's point of view I don't see any issues with tuning this option. But since now I wouldn't recommend to enable it in production as it partially bypassed our regular development cycle. Being enabled in master for a while by default allows more develpers to use/try the feature before release. This can be considered as an additional implicit QA process. But as we just discovered this hasn't happened.
Hence you can definitely try it but this exposes your cluster(s) to some risk as for any new (and incompletely tested) feature....
Thanks,
Igor
On 8/20/2020 4:06 PM, Seena Fallah wrote:
Greate, thanks.
Is it safe to change it manually in ceph.conf until next nautilus release or should I wait for the next nautilus release for this change? I mean does qa run on this value for this config that we could trust and change it or should we wait until the next nautilus release that qa ran on this value?
On Thu, Aug 20, 2020 at 5:25 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
Hi Seena,
this parameter isn't intended to be adjusted in production environments - it's supposed that default behavior covers all regular customers' needs.
The issue though is that default setting is invalid. It should be 'use_some_extra'. Gonna fix that shortly...
Thanks,
Igor
On 8/20/2020 1:44 PM, Seena Fallah wrote:
Hi Igor.
Could you please tell why this config is in LEVEL_DEV (https://github.com/ceph/ceph/pull/29687/files#diff-3d7a065928b2852c228ffe669... As it is documented in Ceph we can't use LEVEL_DEV in production environments!
Thanks
On Thu, Aug 20, 2020 at 1:58 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
Hi Simon,
starting Nautlus v14.2.10 Bluestore is able to use 'wasted' space at DB volume.
see this PR: https://github.com/ceph/ceph/pull/29687
Nice overview on the overall BlueFS/RocksDB design can be find here:
https://cf2.cloudferro.com:8080/swift/v1/AUTH_5e376cddf8a94f9294259b5f48d7b2...
Which also includes some overview (as well as additional concerns) for changes brought by the above-mentioned PR.
Thanks,
Igor
On 8/20/2020 11:39 AM, Simon Oosthoek wrote: > Hi Michael, > > thanks for the explanation! So if I understand correctly, we waste 93 > GB per OSD on unused NVME space, because only 30GB is actually used...? > > And to improve the space for rocksdb, we need to plan for 300GB per > rocksdb partition in order to benefit from this advantage.... > > Reducing the number of small files is something we always ask of our > users, but reality is what it is ;-) > > I'll have to look into how I can get an informative view on these > metrics... It's pretty overwhelming the amount of information coming > out of the ceph cluster, even when you look only superficially... > > Cheers, > > /Simon > > On 20/08/2020 10:16, Michael Bisig wrote: >> Hi Simon >> >> As far as I know, RocksDB only uses "leveled" space on the NVME >> partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every >> DB space above such a limit will automatically end up on slow devices. >> In your setup where you have 123GB per OSD that means you only use >> 30GB of fast device. The DB which spills over this limit will be >> offloaded to the HDD and accordingly, it slows down requests and >> compactions. >> >> You can proof what your OSD currently consumes with: >> ceph daemon osd.X perf dump >> >> Informative values are `db_total_bytes`, `db_used_bytes` and >> `slow_used_bytes`. This changes regularly because of the ongoing >> compactions but Prometheus mgr module exports these values such that >> you can track it. >> >> Small files generally leads to bigger RocksDB, especially when you >> use EC, but this depends on the actual amount and file sizes. >> >> I hope this helps. >> Regards, >> Michael >> >> On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl <mailto:s.oosthoek@science.ru.nl>> wrote: >> >> Hi >> >> Recently our ceph cluster (nautilus) is experiencing bluefs >> spillovers, >> just 2 osd's and I disabled the warning for these osds. >> (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false) >> >> I'm wondering what causes this and how this can be prevented. >> >> As I understand it the rocksdb for the OSD needs to store more >> than fits >> on the NVME logical volume (123G for 12T OSD). A way to fix it >> could be >> to increase the logical volume on the nvme (if there was space >> on the >> nvme, which there isn't at the moment). >> >> This is the current size of the cluster and how much is free: >> >> [root@cephmon1 ~]# ceph df >> RAW STORAGE: >> CLASS SIZE AVAIL USED RAW USED >> %RAW USED >> hdd 1.8 PiB 842 TiB 974 TiB 974 >> TiB 53.63 >> TOTAL 1.8 PiB 842 TiB 974 TiB 974 >> TiB 53.63 >> >> POOLS: >> POOL ID STORED OBJECTS USED >> %USED MAX AVAIL >> cephfs_data 1 572 MiB 121.26M 2.4 GiB >> 0 167 TiB >> cephfs_metadata 2 56 GiB 5.15M 57 GiB >> 0 167 TiB >> cephfs_data_3copy 8 201 GiB 51.68k 602 GiB >> 0.09 222 TiB >> cephfs_data_ec83 13 643 TiB 279.75M 953 TiB >> 58.86 485 TiB >> rbd 14 21 GiB 5.66k 64 GiB >> 0 222 TiB >> .rgw.root 15 1.2 KiB 4 1 MiB >> 0 167 TiB >> default.rgw.control 16 0 B 8 0 B >> 0 167 TiB >> default.rgw.meta 17 765 B 4 1 MiB >> 0 167 TiB >> default.rgw.log 18 0 B 207 0 B >> 0 167 TiB >> cephfs_data_ec57 20 433 MiB 230 1.2 GiB >> 0 278 TiB >> >> The amount used can still grow a bit before we need to add >> nodes, but >> apparently we are running into the limits of our rocskdb >> partitions. >> >> Did we choose a parameter (e.g. minimal object size) too small, >> so we >> have too much objects on these spillover OSDs? Or is it that too >> many >> small files are stored on the cephfs filesystems? >> >> When we expand the cluster, we can choose larger nvme devices to >> allow >> larger rocksdb partitions, but is that the right way to deal >> with this, >> or should we adjust some parameters on the cluster that will >> reduce the >> rocksdb size? >> >> Cheers >> >> /Simon >> _______________________________________________ >> ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> >> To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io> >> > _______________________________________________ > ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> > To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io> _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io>
Ok thanks. And also as you mentioned in the doc you shared from cloudferro, It's not good to change `write_buffer_size` for bluestore rocksdb to fit our db? On Fri, Aug 21, 2020 at 1:46 AM Igor Fedotov <ifedotov@suse.de> wrote:
Honestly I don't have any perfect solution for now.
If this is urgent you probably better to proceed with enabling the new DB space management feature.
But please do that eventually, modify 1-2 OSDs at the first stage and test them for some period (may be a week or two).
Thanks,
Igor
On 8/20/2020 5:36 PM, Seena Fallah wrote:
So what do you suggest for a short term solution? (I think you won't backport it to nautilus at least about 6 month)
Changing db size is too expensive because I should buy new NVME devices with double size and also redeploy all my OSDs. Manual compaction will still have an impact on performance and doing it for a month doesn't look very good!
On Thu, Aug 20, 2020 at 6:52 PM Igor Fedotov <ifedotov@suse.de> wrote:
Correct. On 8/20/2020 5:15 PM, Seena Fallah wrote:
So you won't backport it to nautilus until it gets default to master for a while?
On Thu, Aug 20, 2020 at 6:00 PM Igor Fedotov <ifedotov@suse.de> wrote:
From technical/developer's point of view I don't see any issues with tuning this option. But since now I wouldn't recommend to enable it in production as it partially bypassed our regular development cycle. Being enabled in master for a while by default allows more develpers to use/try the feature before release. This can be considered as an additional implicit QA process. But as we just discovered this hasn't happened.
Hence you can definitely try it but this exposes your cluster(s) to some risk as for any new (and incompletely tested) feature....
Thanks,
Igor
On 8/20/2020 4:06 PM, Seena Fallah wrote:
Greate, thanks.
Is it safe to change it manually in ceph.conf until next nautilus release or should I wait for the next nautilus release for this change? I mean does qa run on this value for this config that we could trust and change it or should we wait until the next nautilus release that qa ran on this value?
On Thu, Aug 20, 2020 at 5:25 PM Igor Fedotov <ifedotov@suse.de> wrote:
Hi Seena,
this parameter isn't intended to be adjusted in production environments - it's supposed that default behavior covers all regular customers' needs.
The issue though is that default setting is invalid. It should be 'use_some_extra'. Gonna fix that shortly...
Thanks,
Igor
On 8/20/2020 1:44 PM, Seena Fallah wrote:
Hi Igor.
Could you please tell why this config is in LEVEL_DEV ( https://github.com/ceph/ceph/pull/29687/files#diff-3d7a065928b2852c228ffe669... As it is documented in Ceph we can't use LEVEL_DEV in production environments!
Thanks
On Thu, Aug 20, 2020 at 1:58 PM Igor Fedotov <ifedotov@suse.de> wrote:
Hi Simon,
starting Nautlus v14.2.10 Bluestore is able to use 'wasted' space at DB volume.
see this PR: https://github.com/ceph/ceph/pull/29687
Nice overview on the overall BlueFS/RocksDB design can be find here:
https://cf2.cloudferro.com:8080/swift/v1/AUTH_5e376cddf8a94f9294259b5f48d7b2...
Which also includes some overview (as well as additional concerns) for changes brought by the above-mentioned PR.
Thanks,
Igor
Hi Michael,
thanks for the explanation! So if I understand correctly, we waste 93 GB per OSD on unused NVME space, because only 30GB is actually used...?
And to improve the space for rocksdb, we need to plan for 300GB per rocksdb partition in order to benefit from this advantage....
Reducing the number of small files is something we always ask of our users, but reality is what it is ;-)
I'll have to look into how I can get an informative view on these metrics... It's pretty overwhelming the amount of information coming out of the ceph cluster, even when you look only superficially...
Cheers,
/Simon
On 20/08/2020 10:16, Michael Bisig wrote: > Hi Simon > > As far as I know, RocksDB only uses "leveled" space on the NVME > partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every > DB space above such a limit will automatically end up on slow devices. > In your setup where you have 123GB per OSD that means you only use > 30GB of fast device. The DB which spills over this limit will be > offloaded to the HDD and accordingly, it slows down requests and > compactions. > > You can proof what your OSD currently consumes with: > ceph daemon osd.X perf dump > > Informative values are `db_total_bytes`, `db_used_bytes` and > `slow_used_bytes`. This changes regularly because of the ongoing > compactions but Prometheus mgr module exports these values such
On 8/20/2020 11:39 AM, Simon Oosthoek wrote: that
> you can track it. > > Small files generally leads to bigger RocksDB, especially when you > use EC, but this depends on the actual amount and file sizes. > > I hope this helps. > Regards, > Michael > > On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl> wrote: > > Hi > > Recently our ceph cluster (nautilus) is experiencing bluefs > spillovers, > just 2 osd's and I disabled the warning for these osds. > (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false) > > I'm wondering what causes this and how this can be prevented. > > As I understand it the rocksdb for the OSD needs to store more > than fits > on the NVME logical volume (123G for 12T OSD). A way to fix it > could be > to increase the logical volume on the nvme (if there was space > on the > nvme, which there isn't at the moment). > > This is the current size of the cluster and how much is free: > > [root@cephmon1 ~]# ceph df > RAW STORAGE: > CLASS SIZE AVAIL USED RAW USED > %RAW USED > hdd 1.8 PiB 842 TiB 974 TiB 974 > TiB 53.63 > TOTAL 1.8 PiB 842 TiB 974 TiB 974 > TiB 53.63 > > POOLS: > POOL ID STORED OBJECTS USED > %USED MAX AVAIL > cephfs_data 1 572 MiB 121.26M 2.4 GiB > 0 167 TiB > cephfs_metadata 2 56 GiB 5.15M 57 GiB > 0 167 TiB > cephfs_data_3copy 8 201 GiB 51.68k 602 GiB > 0.09 222 TiB > cephfs_data_ec83 13 643 TiB 279.75M 953 TiB > 58.86 485 TiB > rbd 14 21 GiB 5.66k 64 GiB > 0 222 TiB > .rgw.root 15 1.2 KiB 4 1 MiB > 0 167 TiB > default.rgw.control 16 0 B 8 0 B > 0 167 TiB > default.rgw.meta 17 765 B 4 1 MiB > 0 167 TiB > default.rgw.log 18 0 B 207 0 B > 0 167 TiB > cephfs_data_ec57 20 433 MiB 230 1.2 GiB > 0 278 TiB > > The amount used can still grow a bit before we need to add > nodes, but > apparently we are running into the limits of our rocskdb > partitions. > > Did we choose a parameter (e.g. minimal object size) too small, > so we > have too much objects on these spillover OSDs? Or is it that too > many > small files are stored on the cephfs filesystems? > > When we expand the cluster, we can choose larger nvme devices to > allow > larger rocksdb partitions, but is that the right way to deal > with this, > or should we adjust some parameters on the cluster that will > reduce the > rocksdb size? > > Cheers > > /Simon > _______________________________________________ > ceph-users mailing list -- ceph-users@ceph.io > To unsubscribe send an email to ceph-users-leave@ceph.io > _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
Can't say nothing about "write_buffer_size" tuning.. Never tried that. But I presume that these are *"max_bytes_for_level_base*" and *"**max_bytes_for_level_multiplier*" params which rather should be tuned to modify RocksDB level granularity. But I have no ideas how safe this is in a production environment. Thanks, Igor On 8/21/2020 12:51 AM, Seena Fallah wrote:
Ok thanks. And also as you mentioned in the doc you shared from cloudferro, It's not good to change `write_buffer_size` for bluestore rocksdb to fit our db?
On Fri, Aug 21, 2020 at 1:46 AM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
Honestly I don't have any perfect solution for now.
If this is urgent you probably better to proceed with enabling the new DB space management feature.
But please do that eventually, modify 1-2 OSDs at the first stage and test them for some period (may be a week or two).
Thanks,
Igor
On 8/20/2020 5:36 PM, Seena Fallah wrote:
So what do you suggest for a short term solution? (I think you won't backport it to nautilus at least about 6 month)
Changing db size is too expensive because I should buy new NVME devices with double size and also redeploy all my OSDs. Manual compaction will still have an impact on performance and doing it for a month doesn't look very good!
On Thu, Aug 20, 2020 at 6:52 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
Correct.
On 8/20/2020 5:15 PM, Seena Fallah wrote:
So you won't backport it to nautilus until it gets default to master for a while?
On Thu, Aug 20, 2020 at 6:00 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
From technical/developer's point of view I don't see any issues with tuning this option. But since now I wouldn't recommend to enable it in production as it partially bypassed our regular development cycle. Being enabled in master for a while by default allows more develpers to use/try the feature before release. This can be considered as an additional implicit QA process. But as we just discovered this hasn't happened.
Hence you can definitely try it but this exposes your cluster(s) to some risk as for any new (and incompletely tested) feature....
Thanks,
Igor
On 8/20/2020 4:06 PM, Seena Fallah wrote:
Greate, thanks.
Is it safe to change it manually in ceph.conf until next nautilus release or should I wait for the next nautilus release for this change? I mean does qa run on this value for this config that we could trust and change it or should we wait until the next nautilus release that qa ran on this value?
On Thu, Aug 20, 2020 at 5:25 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
Hi Seena,
this parameter isn't intended to be adjusted in production environments - it's supposed that default behavior covers all regular customers' needs.
The issue though is that default setting is invalid. It should be 'use_some_extra'. Gonna fix that shortly...
Thanks,
Igor
On 8/20/2020 1:44 PM, Seena Fallah wrote:
Hi Igor.
Could you please tell why this config is in LEVEL_DEV (https://github.com/ceph/ceph/pull/29687/files#diff-3d7a065928b2852c228ffe669... As it is documented in Ceph we can't use LEVEL_DEV in production environments!
Thanks
On Thu, Aug 20, 2020 at 1:58 PM Igor Fedotov <ifedotov@suse.de <mailto:ifedotov@suse.de>> wrote:
Hi Simon,
starting Nautlus v14.2.10 Bluestore is able to use 'wasted' space at DB volume.
see this PR: https://github.com/ceph/ceph/pull/29687
Nice overview on the overall BlueFS/RocksDB design can be find here:
https://cf2.cloudferro.com:8080/swift/v1/AUTH_5e376cddf8a94f9294259b5f48d7b2...
Which also includes some overview (as well as additional concerns) for changes brought by the above-mentioned PR.
Thanks,
Igor
On 8/20/2020 11:39 AM, Simon Oosthoek wrote: > Hi Michael, > > thanks for the explanation! So if I understand correctly, we waste 93 > GB per OSD on unused NVME space, because only 30GB is actually used...? > > And to improve the space for rocksdb, we need to plan for 300GB per > rocksdb partition in order to benefit from this advantage.... > > Reducing the number of small files is something we always ask of our > users, but reality is what it is ;-) > > I'll have to look into how I can get an informative view on these > metrics... It's pretty overwhelming the amount of information coming > out of the ceph cluster, even when you look only superficially... > > Cheers, > > /Simon > > On 20/08/2020 10:16, Michael Bisig wrote: >> Hi Simon >> >> As far as I know, RocksDB only uses "leveled" space on the NVME >> partition. The values are set to be 300MB, 3GB, 30GB and 300GB. Every >> DB space above such a limit will automatically end up on slow devices. >> In your setup where you have 123GB per OSD that means you only use >> 30GB of fast device. The DB which spills over this limit will be >> offloaded to the HDD and accordingly, it slows down requests and >> compactions. >> >> You can proof what your OSD currently consumes with: >> ceph daemon osd.X perf dump >> >> Informative values are `db_total_bytes`, `db_used_bytes` and >> `slow_used_bytes`. This changes regularly because of the ongoing >> compactions but Prometheus mgr module exports these values such that >> you can track it. >> >> Small files generally leads to bigger RocksDB, especially when you >> use EC, but this depends on the actual amount and file sizes. >> >> I hope this helps. >> Regards, >> Michael >> >> On 20.08.20, 09:10, "Simon Oosthoek" <s.oosthoek@science.ru.nl <mailto:s.oosthoek@science.ru.nl>> wrote: >> >> Hi >> >> Recently our ceph cluster (nautilus) is experiencing bluefs >> spillovers, >> just 2 osd's and I disabled the warning for these osds. >> (ceph config set osd.125 bluestore_warn_on_bluefs_spillover false) >> >> I'm wondering what causes this and how this can be prevented. >> >> As I understand it the rocksdb for the OSD needs to store more >> than fits >> on the NVME logical volume (123G for 12T OSD). A way to fix it >> could be >> to increase the logical volume on the nvme (if there was space >> on the >> nvme, which there isn't at the moment). >> >> This is the current size of the cluster and how much is free: >> >> [root@cephmon1 ~]# ceph df >> RAW STORAGE: >> CLASS SIZE AVAIL USED RAW USED >> %RAW USED >> hdd 1.8 PiB 842 TiB 974 TiB 974 >> TiB 53.63 >> TOTAL 1.8 PiB 842 TiB 974 TiB 974 >> TiB 53.63 >> >> POOLS: >> POOL ID STORED OBJECTS USED >> %USED MAX AVAIL >> cephfs_data 1 572 MiB 121.26M 2.4 GiB >> 0 167 TiB >> cephfs_metadata 2 56 GiB 5.15M 57 GiB >> 0 167 TiB >> cephfs_data_3copy 8 201 GiB 51.68k 602 GiB >> 0.09 222 TiB >> cephfs_data_ec83 13 643 TiB 279.75M 953 TiB >> 58.86 485 TiB >> rbd 14 21 GiB 5.66k 64 GiB >> 0 222 TiB >> .rgw.root 15 1.2 KiB 4 1 MiB >> 0 167 TiB >> default.rgw.control 16 0 B 8 0 B >> 0 167 TiB >> default.rgw.meta 17 765 B 4 1 MiB >> 0 167 TiB >> default.rgw.log 18 0 B 207 0 B >> 0 167 TiB >> cephfs_data_ec57 20 433 MiB 230 1.2 GiB >> 0 278 TiB >> >> The amount used can still grow a bit before we need to add >> nodes, but >> apparently we are running into the limits of our rocskdb >> partitions. >> >> Did we choose a parameter (e.g. minimal object size) too small, >> so we >> have too much objects on these spillover OSDs? Or is it that too >> many >> small files are stored on the cephfs filesystems? >> >> When we expand the cluster, we can choose larger nvme devices to >> allow >> larger rocksdb partitions, but is that the right way to deal >> with this, >> or should we adjust some parameters on the cluster that will >> reduce the >> rocksdb size? >> >> Cheers >> >> /Simon >> _______________________________________________ >> ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> >> To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io> >> > _______________________________________________ > ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> > To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io> _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io <mailto:ceph-users@ceph.io> To unsubscribe send an email to ceph-users-leave@ceph.io <mailto:ceph-users-leave@ceph.io>
participants (4)
-
Igor Fedotov
-
Michael Bisig
-
Seena Fallah
-
Simon Oosthoek