SWAP usage 100% on OSD hosts after migration to Rocky Linux 9 (Ceph 16.2.15)
Hi List! We are in the middle of migration from Ceph v16 to v18. As a first step of migration we decided to change OS version on all OSD hosts from Rocky Linux 8.10 to Rocky Linux 9.5 (by using clean reinstall). We also upgraded OSD Block disks from 4TB to 10TB. Now we have 6 OSD hosts running Rocky 9 with new 10TB disks (8 OSD Block disks per Host). OSD hosts have 64GB of RAM. At this moment Ceph cluster is still on version 16.2.15 "Pacific". Migration to Ceph 18.2.x will be performed later after upgrading to Rocky 9 all remaining parts (MGR+MON, MDS, RGW). Problem: After migration to Rocky 9 (and new version of Docker) we see what our OSD hosts consumes 100% of SWAP space! It takes approximately one week to fill SWAP from 0% to 100%. Docker containers running OSD instances are restarted from time to time (and there is a lot of ugly "Call Trace" messages in dmesg). I was wondering if anyone else has similar issues? Will it be gone after migration to Ceph v18? P.S. Configuration Details: ===== [ceph: root@ceph-mon12 /]# ceph versions { "mon": { "ceph version 16.2.15 (618f440892089921c3e944a991122ddc44e60516) pacific (stable)": 3 }, "mgr": { "ceph version 16.2.15 (618f440892089921c3e944a991122ddc44e60516) pacific (stable)": 3 }, "osd": { "ceph version 16.2.15 (618f440892089921c3e944a991122ddc44e60516) pacific (stable)": 48 }, "mds": { "ceph version 16.2.15 (618f440892089921c3e944a991122ddc44e60516) pacific (stable)": 2 }, "rgw": { "ceph version 16.2.15 (618f440892089921c3e944a991122ddc44e60516) pacific (stable)": 2 }, "overall": { "ceph version 16.2.15 (618f440892089921c3e944a991122ddc44e60516) pacific (stable)": 58 } } [ceph: root@ceph-mon12 /]# === Old Rocky 8 OSD Host: [root@ceph-osd11 ~]# docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 1a4c959eb813 quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 7 months ago Up 7 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-37 e37a366347db quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 7 months ago Up 7 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-1 0fdfb95fd631 quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 7 months ago Up 7 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-14 276ad8a11148 quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 7 months ago Up 7 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-31 9dbd30d11124 quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 7 months ago Up 7 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-19 0ad59f583e6a quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 7 months ago Up 7 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-7 e551a652c3ae quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 7 months ago Up 7 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-42 07ff632ae3e1 quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 7 months ago Up 7 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-25 605b7a1c00f7 quay.io/ceph/ceph "/usr/bin/ceph-crash…" 7 months ago Up 7 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-crash-ceph-osd11 fe2a6b8a35e8 quay.io/prometheus/node-exporter:v1.3.1 "/bin/node_exporter …" 7 months ago Up 7 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-node-exporter-ceph-osd11 [root@ceph-osd11 ~]# uptime 10:52:59 up 216 days, 20:12, 1 user, load average: 0.20, 0.41, 0.53 [root@ceph-osd11 ~]# free -h total used free shared buff/cache available Mem: 62Gi 30Gi 1.2Gi 2.1Gi 30Gi 29Gi Swap: 8.0Gi 2.8Gi 5.2Gi [root@ceph-osd11 ~]# rpm -qa | grep docker docker-buildx-plugin-0.14.0-1.el8.x86_64 docker-ce-cli-26.1.3-1.el8.x86_64 docker-ce-26.1.3-1.el8.x86_64 docker-scan-plugin-0.23.0-3.el8.x86_64 docker-ce-rootless-extras-26.1.3-1.el8.x86_64 docker-compose-plugin-2.27.0-1.el8.x86_64 [root@ceph-osd11 ~]# cat /etc/rocky-release Rocky Linux release 8.10 (Green Obsidian) [root@ceph-osd11 ~]# uname -a Linux ceph-osd11 4.18.0-553.16.1.el8_10.x86_64 #1 SMP Thu Aug 8 17:47:08 UTC 2024 x86_64 x86_64 x86_64 GNU/Linux === New Rocky 9 OSD Host: [root@ceph-osd17 ~]# docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 5cc58e4a77b2 quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 4 days ago Up 4 days ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-52 3a60fecf648d quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 4 months ago Up 4 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-50 9c20407e79eb quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 4 months ago Up 4 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-49 9deadafef9dd quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 4 months ago Up 4 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-48 fcfe62a25fd9 quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 4 months ago Up 4 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-55 38d2d96cc491 quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 4 months ago Up 4 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-51 e29c6bbc1ae7 quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 4 months ago Up 4 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-54 40346a7a45ea quay.io/ceph/ceph "/usr/bin/ceph-osd -…" 4 months ago Up 4 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-osd-53 43c3e3a65531 quay.io/ceph/ceph "/usr/bin/ceph-crash…" 4 months ago Up 4 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-crash-ceph-osd17 d9e436f9788c quay.io/prometheus/node-exporter:v1.3.1 "/bin/node_exporter …" 4 months ago Up 4 months ceph-7e8bff5c-2761-11ec-9bb0-000c29ebc936-node-exporter-ceph-osd17 [root@ceph-osd17 ~]# uptime 10:25:10 up 122 days, 22:37, 1 user, load average: 0.37, 0.26, 0.27 [root@ceph-osd17 ~]# free -h total used free shared buff/cache available Mem: 62Gi 26Gi 1.0Gi 1.0Gi 36Gi 36Gi Swap: 8.0Gi 8.0Gi 7.0Mi [root@ceph-osd17 ~]# rpm -qa | grep docker docker-compose-plugin-2.31.0-1.el9.x86_64 docker-buildx-plugin-0.19.2-1.el9.x86_64 docker-ce-cli-27.4.0-1.el9.x86_64 docker-ce-rootless-extras-27.4.0-1.el9.x86_64 docker-ce-27.4.0-1.el9.x86_64 [root@ceph-osd17 ~]# cat /etc/rocky-release Rocky Linux release 9.5 (Blue Onyx) [root@ceph-osd17 ~]# uname -a Linux ceph-osd17 5.14.0-503.15.1.el9_5.x86_64 #1 SMP PREEMPT_DYNAMIC Tue Nov 26 17:24:29 UTC 2024 x86_64 x86_64 x86_64 GNU/Linux =====
participants (1)
-
Dmitrijs Demidovs