Re: [ceph-users] Use telegraf/influx to detect problems is very difficult
I just briefly peaked into source of module and I suppose it's because main design idea is just to forward existing metrics from ceph core and do not calculate anything. To me it seems most users probably use prometheus which doesn't have this kind of issue. Monitor down is also easy as pie, because it's just "num_mon - mon_quorum". But there is also metric mon_outside_quorum which I have always zero and don't really know how it works. OSD near full will be probably more tricky, you have to use "osd.stat_bytes_used / osd.stat_bytes" and compare it with your own configured value (not metric so not exported) per each OSD. Or you can just watch general cluster health metric (what you should anyway) and rise general alarm in this case. M. On 11. 12. 19 21:18, Mario Giammarco wrote: > Miroslav replied better for us why "is not so simple" to use math. > And osd down was the easiest. How can I calculate: > - monitor down > - osd near full > > ? > > I do not understand why ceph plugin cannot send to influx all the > metrics it has, especially the most useful for creating alarms. > > Il giorno mer 11 dic 2019 alle ore 04:58 Konstantin Shalygin > mailto:k0...@k0ste.ru>> ha scritto: > >> But it is very difficult/complicated to make simple queries because, for >> example I have osd up and osd total but not osd down metric. >> > To determine how much osds down you don't need special metric, > because you already > > have osd_up and osd_in metrics. Just use math. > > > > > k > > > ___ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com -- Miroslav Kalina Systems development specialist miroslav.kal...@livesport.eu +420 773 071 848 Livesport s.r.o. Aspira Business Centre Bucharova 2928/14a, 158 00 Praha 5 www.livesport.eu ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Use telegraf/influx to detect problems is very difficult
Quoting Miroslav Kalina (miroslav.kal...@livesport.eu): > Monitor down is also easy as pie, because it's just "num_mon - > mon_quorum". But there is also metric mon_outside_quorum which I have > always zero and don't really know how it works. See this issue if you want to know where it is used for: https://tracker.ceph.com/issues/35947 TL;DR: it's not what you think it is. Gr. Stefan -- | BIT BV https://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] RBD Object-Map Usuage incorrect
On Thu, Dec 12, 2019 at 9:12 AM Ashley Merrick wrote: > > Due to the recent 5.3.x kernel having support for Object-Map and other > features required in KRBD I have now enabled object-map,fast-diff on some RBD > images with CEPH (14.2.5), I have rebuilt the object map using "rbd > object-map rebuild" > > However for some RBD images, the Provisioned/Total Provisioned then listed in > the Ceph MGR for some images is the full RBD size and not the true size > reflected in a VM using df -h, I have discard enabled and have run fstrim but > I know that for example a 20TB RBD has never gone above the current 9TB shown > in df -h but in CEPH MGR shows as 20TB under Provisioned/Total Provisioned. > > Not sure if I am hitting a bug? Or if this is expected behavior? Unless you know *exactly* what the filesystem is doing in your case and see an inconsistency, this is expected. If you are interested, here is an example: $ rbd create --size 1G img $ sudo rbd map img /dev/rbd0 $ sudo mkfs.ext4 /dev/rbd0 $ sudo mount /dev/rbd0 /mnt $ df -h /mnt Filesystem Size Used Avail Use% Mounted on /dev/rbd0 976M 2.6M 907M 1% /mnt $ rbd du img NAME PROVISIONED USED img1 GiB 60 MiB $ ceph df | grep -B1 rbd POOL ID STORED OBJECTS USED %USED MAX AVAIL rbd 1 33 MiB 20 33 MiB 0 1013 GiB After I create a big file, almost the entire image is shown as used: $ dd if=/dev/zero of=/mnt/file bs=1M count=900 $ df -h /mnt Filesystem Size Used Avail Use% Mounted on /dev/rbd0 976M 903M 6.2M 100% /mnt $ rbd du img NAME PROVISIONED USED img1 GiB 956 MiB $ ceph df | grep -B1 rbd POOL ID STORED OBJECTS USED%USED MAX AVAIL rbd 1 933 MiB 248 933 MiB 0.09 1012 GiB Now if I carefully punch out most of that file, leaving one page in each megabyte, and run fstrim: $ for ((i = 0; i < 900; i++)); do fallocate -p -n -o $((i * 2**20)) -l $((2**20 - 4096)) /mnt/file; done $ sudo fstrim /mnt $ df -h /mnt Filesystem Size Used Avail Use% Mounted on /dev/rbd0 976M 6.1M 903M 1% /mnt $ rbd du img NAME PROVISIONED USED img1 GiB 956 MiB $ ceph df | grep -B1 rbd POOL ID STORED OBJECTS USED %USED MAX AVAIL rbd 1 36 MiB 248 36 MiB 0 1013 GiB You can see that df -h is back to ~6M, but "rbd du" USED remained the same. This is because "rbd du" is very coarse-grained, it works at the object level and doesn't go any deeper. If the number of objects and their sizes remain the same, "rbd du" USED remains the same. It doesn't account for sparseness which I produced above. "ceph df" goes down to the individual bluestore blobs, but only per pool. Looking at STORED, you can see that the space is back, even though the number of objects remained the same. Unfortunately, there is no (fast) way to get the same information per image. So what you see in the dashboard is basically "rbd du". It is fast to compute (especially when object map is enabled), but it shows you the picture at the object level, not at the blob level. Thanks, Ilya ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] help! pg inactive and slow requests after filestore to bluestore migration, version 12.2.12
Hello everybody, I got a tree node ceph cluster made of E3-1220v3, 24GB ram, 6 hdd osd's with 32GB Intel Optane NVMe journal, 10GB networking. I wanted to move to bluestore due to dropping support of filestore, our cluster was working fine with filestore and we could take complete nodes out for maintenance without issues. root@ceph04:~# ceph osd pool get libvirt-pool size size: 3 root@ceph04:~# ceph osd pool get libvirt-pool min_size min_size: 2 I removed all osds from one node, zapping the osd and journal devices, we recreated the osds as bluestore and used a small 5GB partition as rockdb device instead of journal for all osd's. I saw the cluster suffer with pgs inactive and slow request. I tried setting the following on all nodes, but no diffrence: ceph tell osd.* injectargs '--osd_recovery_max_active 1' ceph tell osd.* injectargs '--osd_recovery_op_priority 1' ceph tell osd.* injectargs '--osd_recovery_sleep 0.3' systemctl restart ceph-osd.target It took three days to recover and during this time clients were not responsive. How can I migrate to bluestore without inactive pgs or slow request. I got several more filestore clusters and I would like to know how to migrate without inactive pgs and slow reguests? As a side question, I optimized our cluster for filestore, the Intel Optane NVMe journals showed good fio dsync write tests, does bluestore also use dsync writes for rockdb caching or can we select NVMe devices on other specifications? My test with filestores showed that Optane NVMe SSD was faster then the Samsung NVMe SSD 970 Pro and I only need a a few GB for filestore journals, but with bluestore rockdb caching the situation is different and I can't find documentation on how to speed test NVMe devices for bluestore. Kind regards, Jelle root@ceph04:~# ceph osd tree ID CLASS WEIGHT TYPE NAME STATUS REWEIGHT PRI-AFF -1 60.04524 root default -2 20.01263 host ceph04 0 hdd 2.72899 osd.0 up 1.0 1.0 1 hdd 2.72899 osd.1 up 1.0 1.0 2 hdd 5.45799 osd.2 up 1.0 1.0 3 hdd 2.72899 osd.3 up 1.0 1.0 14 hdd 3.63869 osd.14 up 1.0 1.0 15 hdd 2.72899 osd.15 up 1.0 1.0 -3 20.01263 host ceph05 4 hdd 5.45799 osd.4 up 1.0 1.0 5 hdd 2.72899 osd.5 up 1.0 1.0 6 hdd 2.72899 osd.6 up 1.0 1.0 13 hdd 3.63869 osd.13 up 1.0 1.0 16 hdd 2.72899 osd.16 up 1.0 1.0 18 hdd 2.72899 osd.18 up 1.0 1.0 -4 20.01997 host ceph06 8 hdd 5.45999 osd.8 up 1.0 1.0 9 hdd 2.73000 osd.9 up 1.0 1.0 10 hdd 2.73000 osd.10 up 1.0 1.0 11 hdd 2.73000 osd.11 up 1.0 1.0 12 hdd 3.64000 osd.12 up 1.0 1.0 17 hdd 2.73000 osd.17 up 1.0 1.0 root@ceph04:~# ceph status cluster: id: 85873cda-4865-4147-819d-8deda5345db5 health: HEALTH_WARN 18962/11801097 objects misplaced (0.161%) 1/3933699 objects unfound (0.000%) Reduced data availability: 42 pgs inactive Degraded data redundancy: 3645135/11801097 objects degraded (30.888%), 959 pgs degraded, 960 pgs undersized 110 slow requests are blocked > 32 sec. Implicated osds 3,10,11 services: mon: 3 daemons, quorum ceph04,ceph05,ceph06 mgr: ceph04(active), standbys: ceph06, ceph05 osd: 18 osds: 18 up, 18 in; 964 remapped pgs data: pools: 1 pools, 1024 pgs objects: 3.93M objects, 15.0TiB usage: 31.2TiB used, 28.8TiB / 60.0TiB avail pgs: 4.102% pgs not active 3645135/11801097 objects degraded (30.888%) 18962/11801097 objects misplaced (0.161%) 1/3933699 objects unfound (0.000%) 913 active+undersized+degraded+remapped+backfill_wait 60 active+clean 41 activating+undersized+degraded+remapped 4 active+remapped+backfill_wait 4 active+undersized+degraded+remapped+backfilling 1 undersized+degraded+remapped+backfilling+peered 1 active+recovery_wait+undersized+remapped io: recovery: 197MiB/s, 49objects/s root@ceph04:~# ceph health detail HEALTH_WARN 18962/11801097 objects misplaced (0.161%); 1/3933699 objects unfound (0.000%); Reduced data availability: 42 pgs inactive; Degraded data redundancy: 3643636/11801097 objects degraded (30.875%), 959 pgs degraded, 960 pgs undersized; 110 slow requests are blocked > 32 sec. Implicated osds 3,10,11 OBJECT_MISPLACED 18962/11801097 objects misplaced (0.161%) OBJECT_UNFOUND 1/3933699 objects unfound (0.000%) pg 3.361 has 1 unfound objects PG_AVAILABILITY Reduced data availability: 42 pgs i
Re: [ceph-users] help! pg inactive and slow requests after filestore to bluestore migration, version 12.2.12
Jelle, Try putting just the WAL on the Optane NVMe. I'm guessing your DB is too big to fit within 5GB. We used a 5GB journal on our nodes as well, but when we switched to BlueStore (using ceph-volume lvm batch) it created 37GiB logical volumes (200GB SSD / 5 or 400GB SSD / 10) for our DBs. Also, injecting those settings into the cluster will only work until the OSD is restarted. You'll need to add them to ceph.conf to be persistent. Bryan > On Dec 12, 2019, at 3:40 PM, Jelle de Jong wrote: > > Notice: This email is from an external sender. > > > > Hello everybody, > > I got a tree node ceph cluster made of E3-1220v3, 24GB ram, 6 hdd osd's > with 32GB Intel Optane NVMe journal, 10GB networking. > > I wanted to move to bluestore due to dropping support of filestore, our > cluster was working fine with filestore and we could take complete nodes > out for maintenance without issues. > > root@ceph04:~# ceph osd pool get libvirt-pool size > size: 3 > root@ceph04:~# ceph osd pool get libvirt-pool min_size > min_size: 2 > > I removed all osds from one node, zapping the osd and journal devices, > we recreated the osds as bluestore and used a small 5GB partition as > rockdb device instead of journal for all osd's. > > I saw the cluster suffer with pgs inactive and slow request. > > I tried setting the following on all nodes, but no diffrence: > ceph tell osd.* injectargs '--osd_recovery_max_active 1' > ceph tell osd.* injectargs '--osd_recovery_op_priority 1' > ceph tell osd.* injectargs '--osd_recovery_sleep 0.3' > systemctl restart ceph-osd.target > > It took three days to recover and during this time clients were not > responsive. > > How can I migrate to bluestore without inactive pgs or slow request. I > got several more filestore clusters and I would like to know how to > migrate without inactive pgs and slow reguests? > > As a side question, I optimized our cluster for filestore, the Intel > Optane NVMe journals showed good fio dsync write tests, does bluestore > also use dsync writes for rockdb caching or can we select NVMe devices > on other specifications? My test with filestores showed that Optane NVMe > SSD was faster then the Samsung NVMe SSD 970 Pro and I only need a a few > GB for filestore journals, but with bluestore rockdb caching the > situation is different and I can't find documentation on how to speed > test NVMe devices for bluestore. > > Kind regards, > > Jelle > > root@ceph04:~# ceph osd tree > ID CLASS WEIGHT TYPE NAME STATUS REWEIGHT PRI-AFF > -1 60.04524 root default > -2 20.01263 host ceph04 > 0 hdd 2.72899 osd.0 up 1.0 1.0 > 1 hdd 2.72899 osd.1 up 1.0 1.0 > 2 hdd 5.45799 osd.2 up 1.0 1.0 > 3 hdd 2.72899 osd.3 up 1.0 1.0 > 14 hdd 3.63869 osd.14 up 1.0 1.0 > 15 hdd 2.72899 osd.15 up 1.0 1.0 > -3 20.01263 host ceph05 > 4 hdd 5.45799 osd.4 up 1.0 1.0 > 5 hdd 2.72899 osd.5 up 1.0 1.0 > 6 hdd 2.72899 osd.6 up 1.0 1.0 > 13 hdd 3.63869 osd.13 up 1.0 1.0 > 16 hdd 2.72899 osd.16 up 1.0 1.0 > 18 hdd 2.72899 osd.18 up 1.0 1.0 > -4 20.01997 host ceph06 > 8 hdd 5.45999 osd.8 up 1.0 1.0 > 9 hdd 2.73000 osd.9 up 1.0 1.0 > 10 hdd 2.73000 osd.10 up 1.0 1.0 > 11 hdd 2.73000 osd.11 up 1.0 1.0 > 12 hdd 3.64000 osd.12 up 1.0 1.0 > 17 hdd 2.73000 osd.17 up 1.0 1.0 > > > root@ceph04:~# ceph status > cluster: >id: 85873cda-4865-4147-819d-8deda5345db5 >health: HEALTH_WARN >18962/11801097 objects misplaced (0.161%) >1/3933699 objects unfound (0.000%) >Reduced data availability: 42 pgs inactive >Degraded data redundancy: 3645135/11801097 objects degraded > (30.888%), 959 pgs degraded, 960 pgs undersized >110 slow requests are blocked > 32 sec. Implicated osds 3,10,11 > > services: >mon: 3 daemons, quorum ceph04,ceph05,ceph06 >mgr: ceph04(active), standbys: ceph06, ceph05 >osd: 18 osds: 18 up, 18 in; 964 remapped pgs > > data: >pools: 1 pools, 1024 pgs >objects: 3.93M objects, 15.0TiB >usage: 31.2TiB used, 28.8TiB / 60.0TiB avail >pgs: 4.102% pgs not active > 3645135/11801097 objects degraded (30.888%) > 18962/11801097 objects misplaced (0.161%) > 1/3933699 objects unfound (0.000%) > 913 active+undersized+degraded+remapped+backfill_wait > 60 active+clean > 41 activating+undersized+degraded+remapped > 4 active+remapped+backfill_wait > 4 active+undersized+degraded+remapped+backfillin
[ceph-users] PG Balancer Upmap mode not working
@Wido Den Hollander Regarding the amonut of PGs, and I quote from the docs: "If you have more than 50 OSDs, we recommend approximately 50-100placement groups per OSD to balance out resource usage, datadurability and distribution." (https://docs.ceph.com/docs/master/rados/operations/placement-groups/) ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com