Re: Scrub stuck and 'pg has invalid (post-split) stat'
Please don't drop the list from your response. The first question coming to mind is, why do you have a cache-tier if all your pools are on nvme decices anyway? I don't see any benefit here. Did you try the suggested workaround and disable the cache-tier? Zitat von Cedric <yipikai7@gmail.com>:
Thanks Eugen, see attached infos.
Some more details:
- commands that actually hangs: ceph balancer status ; rbd -p vms ls ; rados -p vms_cache cache-flush-evict-all - all scrub running on vms_caches pgs are stall / start in a loop without actually doing anything - all io are 0 both from ceph status or iostat on nodes
On Tue, Feb 20, 2024 at 10:00 AM Eugen Block <eblock@nde.ag> wrote:
Hi,
some more details would be helpful, for example what's the pool size of the cache pool? Did you issue a PG split before or during the upgrade? This thread [1] deals with the same problem, the described workaround was to set hit_set_count to 0 and disable the cache layer until that is resolved. Afterwards you could enable the cache layer again. But keep in mind that the code for cache tier is entirely removed in Reef (IIRC).
Regards, Eugen
[1] https://ceph-users.ceph.narkive.com/zChyOq5D/ceph-strange-issue-after-adding...
Zitat von Cedric <yipikai7@gmail.com>:
Hello,
Following an upgrade from Nautilus (14.2.22) to Pacific (16.2.13), we encounter an issue with a cache pool becoming completely stuck, relevant messages below:
pg xx.x has invalid (post-split) stats; must scrub before tier agent can activate
In OSD logs, scrubs are starting in a loop without succeeding for all pg of this pool.
What we already tried without luck so far:
- shutdown / restart OSD - rebalance pg between OSD - raise the memory on OSD - repeer PG
Any idea what is causing this? any help will be greatly appreciated
Thanks
Cédric _______________________________________________ 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, A bit of history might help to understand why we have the cache tier. We run openstack on top ceph since many years now (started with mimic, then an upgrade to nautilus (years 2 ago) and today and upgrade to pacific). At the beginning of the setup, we used to have a mix of hdd+ssd devices in HCI mode for openstack nova. After the upgrade to nautilus, we made a hardware refresh with brand new NVME devices. And transitionned from mixed devices to nvme. But we were never able to evict all the data from the vms_cache pools (even with being aggressive with the eviction; the last resort would have been to stop all the virtual instances, and that was not an option for our customers), so we decided to move on and set cache-mode proxy and serve data with only nvme since then. And it's been like this for 1 years and a half. But today, after the upgrade, the situation is that we cannot query any stats (with ceph pg x.x query), rados query hangs, scrub hangs even though all PGs are "active+clean". and there is no client activity reported by the cluster. Recovery, and rebalance. Also some other commands hangs, ie: "ceph balancer status". -------------- bash-4.2$ ceph -s cluster: id: <fsid> health: HEALTH_WARN mon is allowing insecure global_id reclaim noscrub,nodeep-scrub,nosnaptrim flag(s) set 18432 slow ops, oldest one blocked for 7626 sec, daemons [osd.0,osd.1,osd.10,osd.11,osd.112,osd.113,osd.118,osd.119,osd.120,osd.122]... have slow ops. services: mon: 3 daemons, quorum mon1,mon2,mon3(age 36m) mgr: bm9612541(active, since 39m) osd: 72 osds: 72 up (since 97m), 72 in (since 9h) flags noscrub,nodeep-scrub,nosnaptrim data: pools: 8 pools, 2409 pgs objects: 14.64M objects, 92 TiB usage: 276 TiB used, 143 TiB / 419 TiB avail pgs: 2409 active+clean
Hi, thanks for the context. Was there any progress over the weekend? The hanging commands seem to be MGR related, and there's only one in your cluster according to your output. Can you deploy a second one manually, then adopt it with cephadm? Can you add 'ceph versions' as well? Zitat von florian.leduc@socgen.com:
Hi, A bit of history might help to understand why we have the cache tier.
We run openstack on top ceph since many years now (started with mimic, then an upgrade to nautilus (years 2 ago) and today and upgrade to pacific). At the beginning of the setup, we used to have a mix of hdd+ssd devices in HCI mode for openstack nova. After the upgrade to nautilus, we made a hardware refresh with brand new NVME devices. And transitionned from mixed devices to nvme. But we were never able to evict all the data from the vms_cache pools (even with being aggressive with the eviction; the last resort would have been to stop all the virtual instances, and that was not an option for our customers), so we decided to move on and set cache-mode proxy and serve data with only nvme since then. And it's been like this for 1 years and a half.
But today, after the upgrade, the situation is that we cannot query any stats (with ceph pg x.x query), rados query hangs, scrub hangs even though all PGs are "active+clean". and there is no client activity reported by the cluster. Recovery, and rebalance. Also some other commands hangs, ie: "ceph balancer status".
-------------- bash-4.2$ ceph -s cluster: id: <fsid> health: HEALTH_WARN mon is allowing insecure global_id reclaim noscrub,nodeep-scrub,nosnaptrim flag(s) set 18432 slow ops, oldest one blocked for 7626 sec, daemons [osd.0,osd.1,osd.10,osd.11,osd.112,osd.113,osd.118,osd.119,osd.120,osd.122]... have slow ops.
services: mon: 3 daemons, quorum mon1,mon2,mon3(age 36m) mgr: bm9612541(active, since 39m) osd: 72 osds: 72 up (since 97m), 72 in (since 9h) flags noscrub,nodeep-scrub,nosnaptrim
data: pools: 8 pools, 2409 pgs objects: 14.64M objects, 92 TiB usage: 276 TiB used, 143 TiB / 419 TiB avail pgs: 2409 active+clean _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
Hello, Sorry for the late reply, so yes we finally find a solution, which was to split apart the cache_pool on dedicated OSD. It had the effect to clear off slow ops and allow the cluster to serves clients again, after 5 days of lock down, hopefully the majority of VM resume well, thanks to the virtio driver that does not seems to have any timeout. It seems that at least one of the main culprit was to store both cold and hot data pool on same OSD (which in the end totally make sens), maybe some others actions engaged also had an effect, we are still trying to trouble shoot the root of slow ops, weirdly it was the 5th cluster upgraded and all as almost the same configuration, but this one handles 5x time more workload. In the hope it could help. Cédric
On 26 Feb 2024, at 10:57, Eugen Block <eblock@nde.ag> wrote:
Hi,
thanks for the context. Was there any progress over the weekend? The hanging commands seem to be MGR related, and there's only one in your cluster according to your output. Can you deploy a second one manually, then adopt it with cephadm? Can you add 'ceph versions' as well?
Zitat von florian.leduc@socgen.com:
Hi, A bit of history might help to understand why we have the cache tier.
We run openstack on top ceph since many years now (started with mimic, then an upgrade to nautilus (years 2 ago) and today and upgrade to pacific). At the beginning of the setup, we used to have a mix of hdd+ssd devices in HCI mode for openstack nova. After the upgrade to nautilus, we made a hardware refresh with brand new NVME devices. And transitionned from mixed devices to nvme. But we were never able to evict all the data from the vms_cache pools (even with being aggressive with the eviction; the last resort would have been to stop all the virtual instances, and that was not an option for our customers), so we decided to move on and set cache-mode proxy and serve data with only nvme since then. And it's been like this for 1 years and a half.
But today, after the upgrade, the situation is that we cannot query any stats (with ceph pg x.x query), rados query hangs, scrub hangs even though all PGs are "active+clean". and there is no client activity reported by the cluster. Recovery, and rebalance. Also some other commands hangs, ie: "ceph balancer status".
-------------- bash-4.2$ ceph -s cluster: id: <fsid> health: HEALTH_WARN mon is allowing insecure global_id reclaim noscrub,nodeep-scrub,nosnaptrim flag(s) set 18432 slow ops, oldest one blocked for 7626 sec, daemons [osd.0,osd.1,osd.10,osd.11,osd.112,osd.113,osd.118,osd.119,osd.120,osd.122]... have slow ops.
services: mon: 3 daemons, quorum mon1,mon2,mon3(age 36m) mgr: bm9612541(active, since 39m) osd: 72 osds: 72 up (since 97m), 72 in (since 9h) flags noscrub,nodeep-scrub,nosnaptrim
data: pools: 8 pools, 2409 pgs objects: 14.64M objects, 92 TiB usage: 276 TiB used, 143 TiB / 419 TiB avail pgs: 2409 active+clean _______________________________________________ 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, great that you found a solution. Maybe that also helps to get rid of the cache-tier entirely? Zitat von Cedric <yipikai7@gmail.com>:
Hello,
Sorry for the late reply, so yes we finally find a solution, which was to split apart the cache_pool on dedicated OSD. It had the effect to clear off slow ops and allow the cluster to serves clients again, after 5 days of lock down, hopefully the majority of VM resume well, thanks to the virtio driver that does not seems to have any timeout.
It seems that at least one of the main culprit was to store both cold and hot data pool on same OSD (which in the end totally make sens), maybe some others actions engaged also had an effect, we are still trying to trouble shoot the root of slow ops, weirdly it was the 5th cluster upgraded and all as almost the same configuration, but this one handles 5x time more workload.
In the hope it could help.
Cédric
On 26 Feb 2024, at 10:57, Eugen Block <eblock@nde.ag> wrote:
Hi,
thanks for the context. Was there any progress over the weekend? The hanging commands seem to be MGR related, and there's only one in your cluster according to your output. Can you deploy a second one manually, then adopt it with cephadm? Can you add 'ceph versions' as well?
Zitat von florian.leduc@socgen.com:
Hi, A bit of history might help to understand why we have the cache tier.
We run openstack on top ceph since many years now (started with mimic, then an upgrade to nautilus (years 2 ago) and today and upgrade to pacific). At the beginning of the setup, we used to have a mix of hdd+ssd devices in HCI mode for openstack nova. After the upgrade to nautilus, we made a hardware refresh with brand new NVME devices. And transitionned from mixed devices to nvme. But we were never able to evict all the data from the vms_cache pools (even with being aggressive with the eviction; the last resort would have been to stop all the virtual instances, and that was not an option for our customers), so we decided to move on and set cache-mode proxy and serve data with only nvme since then. And it's been like this for 1 years and a half.
But today, after the upgrade, the situation is that we cannot query any stats (with ceph pg x.x query), rados query hangs, scrub hangs even though all PGs are "active+clean". and there is no client activity reported by the cluster. Recovery, and rebalance. Also some other commands hangs, ie: "ceph balancer status".
-------------- bash-4.2$ ceph -s cluster: id: <fsid> health: HEALTH_WARN mon is allowing insecure global_id reclaim noscrub,nodeep-scrub,nosnaptrim flag(s) set 18432 slow ops, oldest one blocked for 7626 sec, daemons [osd.0,osd.1,osd.10,osd.11,osd.112,osd.113,osd.118,osd.119,osd.120,osd.122]... have slow ops.
services: mon: 3 daemons, quorum mon1,mon2,mon3(age 36m) mgr: bm9612541(active, since 39m) osd: 72 osds: 72 up (since 97m), 72 in (since 9h) flags noscrub,nodeep-scrub,nosnaptrim
data: pools: 8 pools, 2409 pgs objects: 14.64M objects, 92 TiB usage: 276 TiB used, 143 TiB / 419 TiB avail pgs: 2409 active+clean _______________________________________________ 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
Not really, as unfortunately the cache eviction fails for some rbd objects that still hace some "lock", right now we need to understand why the eviction fails on these objects, and find a solution to have the cache eviction fully working. I will provide more information later on. If you have any pointers, well they will be greatly appreciated. Cheers On Wed, Feb 28, 2024 at 9:50 PM Eugen Block <eblock@nde.ag> wrote:
Hi,
great that you found a solution. Maybe that also helps to get rid of the cache-tier entirely?
Zitat von Cedric <yipikai7@gmail.com>:
Hello,
Sorry for the late reply, so yes we finally find a solution, which was to split apart the cache_pool on dedicated OSD. It had the effect to clear off slow ops and allow the cluster to serves clients again, after 5 days of lock down, hopefully the majority of VM resume well, thanks to the virtio driver that does not seems to have any timeout.
It seems that at least one of the main culprit was to store both cold and hot data pool on same OSD (which in the end totally make sens), maybe some others actions engaged also had an effect, we are still trying to trouble shoot the root of slow ops, weirdly it was the 5th cluster upgraded and all as almost the same configuration, but this one handles 5x time more workload.
In the hope it could help.
Cédric
On 26 Feb 2024, at 10:57, Eugen Block <eblock@nde.ag> wrote:
Hi,
thanks for the context. Was there any progress over the weekend? The hanging commands seem to be MGR related, and there's only one in your cluster according to your output. Can you deploy a second one manually, then adopt it with cephadm? Can you add 'ceph versions' as well?
Zitat von florian.leduc@socgen.com:
Hi, A bit of history might help to understand why we have the cache tier.
We run openstack on top ceph since many years now (started with mimic, then an upgrade to nautilus (years 2 ago) and today and upgrade to pacific). At the beginning of the setup, we used to have a mix of hdd+ssd devices in HCI mode for openstack nova. After the upgrade to nautilus, we made a hardware refresh with brand new NVME devices. And transitionned from mixed devices to nvme. But we were never able to evict all the data from the vms_cache pools (even with being aggressive with the eviction; the last resort would have been to stop all the virtual instances, and that was not an option for our customers), so we decided to move on and set cache-mode proxy and serve data with only nvme since then. And it's been like this for 1 years and a half.
But today, after the upgrade, the situation is that we cannot query any stats (with ceph pg x.x query), rados query hangs, scrub hangs even though all PGs are "active+clean". and there is no client activity reported by the cluster. Recovery, and rebalance. Also some other commands hangs, ie: "ceph balancer status".
-------------- bash-4.2$ ceph -s cluster: id: <fsid> health: HEALTH_WARN mon is allowing insecure global_id reclaim noscrub,nodeep-scrub,nosnaptrim flag(s) set 18432 slow ops, oldest one blocked for 7626 sec, daemons [osd.0,osd.1,osd.10,osd.11,osd.112,osd.113,osd.118,osd.119,osd.120,osd.122]... have slow ops.
services: mon: 3 daemons, quorum mon1,mon2,mon3(age 36m) mgr: bm9612541(active, since 39m) osd: 72 osds: 72 up (since 97m), 72 in (since 9h) flags noscrub,nodeep-scrub,nosnaptrim
data: pools: 8 pools, 2409 pgs objects: 14.64M objects, 92 TiB usage: 276 TiB used, 143 TiB / 419 TiB avail pgs: 2409 active+clean _______________________________________________ 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
participants (3)
-
Cedric
-
Eugen Block
-
florian.leduc＠socgen.com