泰山2280 ceph( requests are blocked ) 异常解决方法

1. 查询ceph状态时，报出如下错误：
2. [root@hh-yun-puppet-129021 ~]# ceph -s cluster dc4f91c1-8792-4948-b68f-2fcea75f53b9 health HEALTH_WARN 1 requests are blocked > 32 sec monmap e3: 5 mons at {hh-yun-ceph-cinder015-128055=240.30.128.55:6789/0,hh-yun-ceph-cinder017-128057=240.30.128.57:6789/0,hh-yun-ceph-cinder024-128074=240.30.128.74:6789/0,hh-yun-ceph-cinder025-128075=240.30.128.75:6789/0,hh-yun-ceph-cinder026-128076=240.30.128.76:6789/0}, election epoch 216, quorum 0,1,2,3,4 hh-yun-ceph-cinder015-128055,hh-yun-ceph-cinder017-128057,hh-yun-ceph-cinder024-128074,hh-yun-ceph-cinder025-128075,hh-yun-ceph-cinder026-128076 osdmap e97975: 190 osds: 190 up, 190 in pgmap v13666786: 20544 pgs, 2 pools, 77479 GB data, 19508 kobjects 228 TB used, 426 TB / 654 TB avail 20542 active+clean 2 active+clean+scrubbing+deep client io 47657 kB/s rd, 164 MB/s wr, 5406 op/s
   

寻找 block 的请求

1. [root@hh-yun-puppet-129021 ~]# ceph health detail
2. HEALTH_WARN 1 requests are blocked > 32 sec; 1 osds have slow requests
3. 1 ops are blocked > 33554.4 sec
4. 1 ops are blocked > 33554.4 sec on osd.16
5. 1 osds have slow requests

可以看到 osd.16 具有一个操作 block 
解决方法 
查询 osd 对应主机

1. [root@hh-yun-puppet-129021 ~]# ceph osd tree
2. # id weight type name up/down reweight
3. -1 598 root default
4. -2 40 host hh-yun-ceph-cinder015-128055
5. 0 4 osd.0 up 1
6. 1 4 osd.1 up 1
7. 2 4 osd.2 up 1
8. 3 4 osd.3 up 1
9. 4 4 osd.4 up 1
10. 5 4 osd.5 up 1
11. 6 4 osd.6 up 1
12. 7 4 osd.7 up 1
13. 8 4 osd.8 up 1
14. 9 4 osd.9 up 1
15. -3 40 host hh-yun-ceph-cinder016-128056
16. 10 4 osd.10 up 1
17. 11 4 osd.11 up 1
18. 12 4 osd.12 up 1
19. 13 4 osd.13 up 1
20. 14 4 osd.14 up 1
21. 15 4 osd.15 up 1
22. 16 4 osd.16 up 1
23. 17 4 osd.17 up 1

重启 osd

1. [root@hh-yun-ceph-cinder016-128056 ~]# /etc/init.d/ceph stop osd.16
2. === osd.16 ===
3. Stopping Ceph osd.16 on hh-yun-ceph-cinder016-128056...kill 2799859...kill 2799859...done
4. [root@hh-yun-ceph-cinder016-128056 ~]# /etc/init.d/ceph start osd.16
5. === osd.16 ===
6. create-or-move updated item name 'osd.16' weight 3.64 at location {host=hh-yun-ceph-cinder016-128056,root=default} to crush map
7. Starting Ceph osd.16 on hh-yun-ceph-cinder016-128056...
8. Running as unit run-3126361.service.

系统会对该 osd 执行 recovery 操作, recovery 过程中, 会断开 block request, 那么这个 request 将会重新请求 mon 节点, 并重新获得新的 pg map, 得到最新的数据访问位置, 从而解决上述问题

恢复后的状态

1. [root@hh-yun-puppet-129021 ~]# ceph -s
2. cluster dc4f91c1-8792-4948-b68f-2fcea75f53b9
3. health HEALTH_OK
4. monmap e3: 5 mons at {hh-yun-ceph-cinder015-128055=240.30.128.55:6789/0,hh-yun-ceph-cinder017-128057=240.30.128.57:6789/0,hh-yun-ceph-cinder024-128074=240.30.128.74:6789/0,hh-yun-ceph-cinder025-128075=240.30.128.75:6789/0,hh-yun-ceph-cinder026-128076=240.30.128.76:6789/0}, election epoch 216, quorum 0,1,2,3,4 hh-yun-ceph-cinder015-128055,hh-yun-ceph-cinder017-128057,hh-yun-ceph-cinder024-128074,hh-yun-ceph-cinder025-128075,hh-yun-ceph-cinder026-128076
5. osdmap e97981: 190 osds: 190 up, 190 in
6. pgmap v13669826: 20544 pgs, 2 pools, 77488 GB data, 19510 kobjects
7. 228 TB used, 426 TB / 654 TB avail
8. 20541 active+clean
9. 3 active+clean+scrubbing+deep
10. client io 21801 kB/s rd, 66461 kB/s wr, 2328 op/s

requests are blocked > 32 sec 有可能是在数据迁移过程中, 用户正在对该数据块进行访问, 但访问还没有完成, 数据就迁移到别的 OSD 中, 那么就会导致有请求被 block, 对用户也是有影响的