Chapter 3. Monitoring a Ceph storage cluster

As a storage administrator, you can monitor the overall health of the Red Hat Ceph Storage cluster, along with monitoring the health of the individual components of Ceph.

Once you have a running Red Hat Ceph Storage cluster, you might begin monitoring the storage cluster to ensure that the Ceph Monitor and Ceph OSD daemons are running, at a high-level. Ceph storage cluster clients connect to a Ceph Monitor and receive the latest version of the storage cluster map before they can read and write data to the Ceph pools within the storage cluster. So the monitor cluster must have agreement on the state of the cluster before Ceph clients can read and write data.

Ceph OSDs must peer the placement groups on the primary OSD with the copies of the placement groups on secondary OSDs. If faults arise, peering will reflect something other than the active + clean state.

3.1. Prerequisites

  • A running Red Hat Ceph Storage cluster.

3.2. High-level monitoring of a Ceph storage cluster

As a storage administrator, you can monitor the health of the Ceph daemons to ensure that they are up and running. High level monitoring also involves checking the storage cluster capacity to ensure that the storage cluster does not exceed its full ratio. The Red Hat Ceph Storage Dashboard is the most common way to conduct high-level monitoring. However, you can also use the command-line interface, the Ceph admin socket or the Ceph API to monitor the storage cluster.

3.2.1. Prerequisites

  • A running Red Hat Ceph Storage cluster.

3.2.2. Using the Ceph command interface interactively

You can interactively interface with the Ceph storage cluster by using the ceph command-line utility.

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. To run the ceph utility in interactive mode.

    1. Bare-metal deployments:

      Example

      [root@mon ~]# ceph
      ceph> health
      ceph> status
      ceph> quorum_status
      ceph> mon_status

    2. Container deployments:

      Red Hat Enterprise Linux 7

      docker exec -it ceph-mon-MONITOR_NAME /bin/bash

      Red Hat Enterprise Linux 8

      podman exec -it ceph-mon-MONITOR_NAME /bin/bash

      Replace
      • MONITOR_NAME with the name of the Ceph Monitor container, found by running the docker ps or podman ps command respectively.

        Example

        [root@container-host ~]# podman exec -it ceph-mon-mon01 /bin/bash

        This example opens an interactive terminal session on mon01, where you can start the Ceph interactive shell.

3.2.3. Checking the storage cluster health

After you start the Ceph storage cluster, and before you start reading or writing data, check the storage cluster’s health first.

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. You can check on the health of the Ceph storage cluster with the following:

    [root@mon ~]# ceph health
  2. If you specified non-default locations for the configuration or keyring, you can specify their locations:

    [root@mon ~]# ceph -c /path/to/conf -k /path/to/keyring health

Upon starting the Ceph cluster, you will likely encounter a health warning such as HEALTH_WARN XXX num placement groups stale. Wait a few moments and check it again. When the storage cluster is ready, ceph health should return a message such as HEALTH_OK. At that point, it is okay to begin using the cluster.

3.2.4. Watching storage cluster events

You can watch events that are happening with the Ceph storage cluster using the command-line interface.

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. To watch the cluster’s ongoing events on the command line, open a new terminal, and then enter:

    [root@mon ~]# ceph -w

    Ceph will print each event. For example, a tiny Ceph cluster consisting of one monitor and two OSDs may print the following:

    cluster b370a29d-9287-4ca3-ab57-3d824f65e339
     health HEALTH_OK
     monmap e1: 1 mons at {ceph1=10.0.0.8:6789/0}, election epoch 2, quorum 0 ceph1
     osdmap e63: 2 osds: 2 up, 2 in
      pgmap v41338: 952 pgs, 20 pools, 17130 MB data, 2199 objects
            115 GB used, 167 GB / 297 GB avail
                 952 active+clean
    
    2014-06-02 15:45:21.655871 osd.0 [INF] 17.71 deep-scrub ok
    2014-06-02 15:45:47.880608 osd.1 [INF] 1.0 scrub ok
    2014-06-02 15:45:48.865375 osd.1 [INF] 1.3 scrub ok
    2014-06-02 15:45:50.866479 osd.1 [INF] 1.4 scrub ok
    2014-06-02 15:45:01.345821 mon.0 [INF] pgmap v41339: 952 pgs: 952 active+clean; 17130 MB data, 115 GB used, 167 GB / 297 GB avail
    2014-06-02 15:45:05.718640 mon.0 [INF] pgmap v41340: 952 pgs: 1 active+clean+scrubbing+deep, 951 active+clean; 17130 MB data, 115 GB used, 167 GB / 297 GB avail
    2014-06-02 15:45:53.997726 osd.1 [INF] 1.5 scrub ok
    2014-06-02 15:45:06.734270 mon.0 [INF] pgmap v41341: 952 pgs: 1 active+clean+scrubbing+deep, 951 active+clean; 17130 MB data, 115 GB used, 167 GB / 297 GB avail
    2014-06-02 15:45:15.722456 mon.0 [INF] pgmap v41342: 952 pgs: 952 active+clean; 17130 MB data, 115 GB used, 167 GB / 297 GB avail
    2014-06-02 15:46:06.836430 osd.0 [INF] 17.75 deep-scrub ok
    2014-06-02 15:45:55.720929 mon.0 [INF] pgmap v41343: 952 pgs: 1 active+clean+scrubbing+deep, 951 active+clean; 17130 MB data, 115 GB used, 167 GB / 297 GB avail

    The output provides:

    • Cluster ID
    • Cluster health status
    • The monitor map epoch and the status of the monitor quorum
    • The OSD map epoch and the status of OSDs
    • The placement group map version
    • The number of placement groups and pools
    • The notional amount of data stored and the number of objects stored
    • The total amount of data stored

3.2.5. How Ceph calculates data usage

The used value reflects the actual amount of raw storage used. The xxx GB / xxx GB value means the amount available, the lesser of the two numbers, of the overall storage capacity of the cluster. The notional number reflects the size of the stored data before it is replicated, cloned or snapshotted. Therefore, the amount of data actually stored typically exceeds the notional amount stored, because Ceph creates replicas of the data and may also use storage capacity for cloning and snapshotting.

3.2.6. Understanding the storage clusters usage stats

To check a cluster’s data usage and data distribution among pools, use the df option. It is similar to the Linux df command. Execute the following:

[root@mon ~]# ceph df
RAW STORAGE:
    CLASS     SIZE        AVAIL       USED       RAW USED     %RAW USED
    hdd       662 TiB     611 TiB     51 TiB       51 TiB          7.74
    TOTAL     662 TiB     611 TiB     51 TiB       51 TiB          7.74

POOLS:
    POOL                     ID    STORED    OBJECTS   USED     %USED   MAX AVAIL
    default.rgw.users.keys   276       0 B         0       0 B      0     193 TiB
    default.rgw.data.root    277       0 B         0       0 B      0     193 TiB
    .rgw.root                278   5.7 KiB        12   2.2 MiB      0     193 TiB
    default.rgw.control      279       0 B         8       0 B      0     193 TiB
    default.rgw.gc           280       0 B         0       0 B      0     193 TiB

The RAW STORAGE section of the output provides an overview of the amount of storage the storage cluster uses for data.

  • CLASS: The type of devices used.
  • SIZE: The overall storage capacity of the storage cluster.
  • AVAIL: The amount of free space available in the storage cluster.
  • USED: The amount of used space in the storage cluster.
  • RAW USED: The sum of USED space and the space allocated the db and wal BlueStore partitions.
  • % RAW USED: The percentage of of RAW USED. Use this number in conjunction with the full ratio and near full ratio to ensure that you are not reaching the storage cluster’s capacity.

The POOLS section of the output provides a list of pools and the notional usage of each pool. The output from this section DOES NOT reflect replicas, clones or snapshots. For example, if you store an object with 1 MB of data, the notional usage will be 1 MB, but the actual usage may be 3 MB or more depending on the number of replicas for example, size = 3, clones and snapshots.

  • POOL: The name of the pool.
  • ID: The pool ID.
  • STORED: The amount of data stored by the user.
  • OBJECTS: The notional number of objects stored per pool.
  • USED: The notional amount of data stored in kilobytes, unless the number appends M for megabytes or G for gigabytes.
  • %USED: The notional percentage of storage used per pool.
  • MAX AVAIL: The maximum available space in a pool.
Note

The numbers in the POOLS section are notional. They are not inclusive of the number of replicas, snapshots or clones. As a result, the sum of the USED and %USED amounts will not add up to the RAW USED and %RAW USED amounts in the GLOBAL section of the output.

Additional Resources

3.2.7. Understanding the OSD usage stats

Use the ceph osd df command to view OSD utilization stats.

[root@mon]# ceph osd df
ID CLASS WEIGHT  REWEIGHT SIZE    USE     DATA    OMAP    META    AVAIL   %USE VAR  PGS
 3   hdd 0.90959  1.00000  931GiB 70.1GiB 69.1GiB      0B    1GiB  861GiB 7.53 2.93  66
 4   hdd 0.90959  1.00000  931GiB 1.30GiB  308MiB      0B    1GiB  930GiB 0.14 0.05  59
 0   hdd 0.90959  1.00000  931GiB 18.1GiB 17.1GiB      0B    1GiB  913GiB 1.94 0.76  57
MIN/MAX VAR: 0.02/2.98  STDDEV: 2.91
  • ID: The name of the OSD.
  • CLASS: The type of devices the OSD uses.
  • WEIGHT: The weight of the OSD in the CRUSH map.
  • REWEIGHT: The default reweight value.
  • SIZE: The overall storage capacity of the OSD.
  • USE: The OSD capacity.
  • DATA: The amount of OSD capacity that is used by user data.
  • OMAP: An estimate value of the bluefs storage that is being used to store object map (omap) data (key value pairs stored in rocksdb).
  • META: The bluefs space allocated, or the value set in the bluestore_bluefs_min parameter, whichever is larger, for internal metadata which is calculated as the total space allocated in bluefs minus the estimated omap data size.
  • AVAIL: The amount of free space available on the OSD.
  • %USE: The notional percentage of storage used by the OSD
  • VAR: The variation above or below average utilization.
  • PGS: The number of placement groups in the OSD.
  • MIN/MAX VAR: The minimum and maximum variation across all OSDs.

Additional Resources

3.2.8. Checking the Red Hat Ceph Storage cluster status

You can check the status of the Red Hat Ceph Storage cluster from the command-line interface. The status sub command or the -s argument will display the current status of the storage cluster.

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. To check a storage cluster’s status, execute the following:

    [root@mon ~]# ceph status

    Or:

    [root@mon ~]# ceph -s
  2. In interactive mode, type status and press Enter:

    [root@mon ~]# ceph> status

    For example, a tiny Ceph cluster consisting of one monitor, and two OSDs can print the following:

    cluster b370a29d-9287-4ca3-ab57-3d824f65e339
     health HEALTH_OK
     monmap e1: 1 mons at {ceph1=10.0.0.8:6789/0}, election epoch 2, quorum 0 ceph1
     osdmap e63: 2 osds: 2 up, 2 in
      pgmap v41332: 952 pgs, 20 pools, 17130 MB data, 2199 objects
            115 GB used, 167 GB / 297 GB avail
                   1 active+clean+scrubbing+deep
                 951 active+clean

3.2.9. Checking the Ceph Monitor status

If the storage cluster has multiple Ceph Monitors, which is a requirement for a production Red Hat Ceph Storage cluster, then check the Ceph Monitor quorum status after starting the storage cluster, and before doing any reading or writing of data.

A quorum must be present when multiple monitors are running.

Check Ceph Monitor status periodically to ensure that they are running. If there is a problem with the Ceph Monitor, that prevents an agreement on the state of the storage cluster, the fault may prevent Ceph clients from reading and writing data.

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. To display the monitor map, execute the following:

    [root@mon ~]# ceph mon stat

    or

    [root@mon ~]# ceph mon dump
  2. To check the quorum status for the storage cluster, execute the following:

    [root@mon ~]# ceph quorum_status -f json-pretty

    Ceph will return the quorum status. A Red Hat Ceph Storage cluster consisting of three monitors may return the following:

    Example

    { "election_epoch": 10,
      "quorum": [
            0,
            1,
            2],
      "monmap": { "epoch": 1,
          "fsid": "444b489c-4f16-4b75-83f0-cb8097468898",
          "modified": "2011-12-12 13:28:27.505520",
          "created": "2011-12-12 13:28:27.505520",
          "mons": [
                { "rank": 0,
                  "name": "a",
                  "addr": "127.0.0.1:6789\/0"},
                { "rank": 1,
                  "name": "b",
                  "addr": "127.0.0.1:6790\/0"},
                { "rank": 2,
                  "name": "c",
                  "addr": "127.0.0.1:6791\/0"}
               ]
        }
    }

3.2.10. Using the Ceph administration socket

Use the administration socket to interact with a given daemon directly by using a UNIX socket file. For example, the socket enables you to:

  • List the Ceph configuration at runtime
  • Set configuration values at runtime directly without relying on Monitors. This is useful when Monitors are down.
  • Dump historic operations
  • Dump the operation priority queue state
  • Dump operations without rebooting
  • Dump performance counters

In addition, using the socket is helpful when troubleshooting problems related to Monitors or OSDs.

Important

The administration socket is only available while a daemon is running. When you shut down the daemon properly, the administration socket is removed. However, if the daemon terminates unexpectedly, the administration socket might persist.

Regardless, if the daemon is not running, a following error is returned when attempting to use the administration socket:

Error 111: Connection Refused

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. To use the socket:

    Syntax

    [root@mon ~]# ceph daemon TYPE.ID COMMAND

    Replace:

    • TYPE with the type of the Ceph daemon (mon, osd, mds).
    • ID with the daemon ID
    • COMMAND with the command to run. Use help to list the available commands for a given daemon.

      Example

      To view a Monitor status of a Ceph Monitor named mon.0:

      [root@mon ~]# ceph daemon mon.0 mon_status
  2. Alternatively, specify the Ceph daemon by using its socket file:

    ceph daemon /var/run/ceph/SOCKET_FILE COMMAND
  3. To view the status of an Ceph OSD named osd.2:

    [root@mon ~]# ceph daemon /var/run/ceph/ceph-osd.2.asok status
  4. To list all socket files for the Ceph processes:

    [root@mon ~]# ls /var/run/ceph

Additional Resources

3.2.11. Understanding the Ceph OSD status

An OSD’s status is either in the cluster, in, or out of the cluster, out. It is either up and running, up, or it is down and not running, or down. If an OSD is up, it may be either in the storage cluster, where data can be read and written, or it is out of the storage cluster. If it was in the cluster and recently moved out of the cluster, Ceph will migrate placement groups to other OSDs. If an OSD is out of the cluster, CRUSH will not assign placement groups to the OSD. If an OSD is down, it should also be out.

Note

If an OSD is down and in, there is a problem and the cluster will not be in a healthy state.

OSD States

If you execute a command such as ceph health, ceph -s or ceph -w, you may notice that the cluster does not always echo back HEALTH OK. Don’t panic. With respect to OSDs, you should expect that the cluster will NOT echo HEALTH OK in a few expected circumstances:

  • You haven’t started the cluster yet, it won’t respond.
  • You have just started or restarted the cluster and it’s not ready yet, because the placement groups are getting created and the OSDs are in the process of peering.
  • You just added or removed an OSD.
  • You just have modified the cluster map.

An important aspect of monitoring OSDs is to ensure that when the cluster is up and running that all OSDs that are in the cluster are up and running, too.

To see if all OSDs are running, execute:

[root@mon ~]# ceph osd stat

or

[root@mon ~]# ceph osd dump

The result should tell you the map epoch, eNNNN, the total number of OSDs, x, how many, y, are up, and how many, z, are in:

eNNNN: x osds: y up, z in

If the number of OSDs that are in the cluster is more than the number of OSDs that are up. Execute the following command to identify the ceph-osd daemons that aren’t running:

[root@mon ~]# ceph osd tree

Example

# id    weight  type name   up/down reweight
-1  3   pool default
-3  3       rack mainrack
-2  3           host osd-host
0   1               osd.0   up  1
1   1               osd.1   up  1
2   1               osd.2   up  1

Tip

The ability to search through a well-designed CRUSH hierarchy may help you troubleshoot the storage cluster by identifying the physical locations faster.

If an OSD is down, connect to the node and start it. You can use Red Hat Storage Console to restart the OSD node, or you can use the command line.

Example

[root@mon ~]# systemctl start ceph-osd@OSD_ID

3.2.12. Additional Resources

3.3. Low-level monitoring of a Ceph storage cluster

As a storage administrator, you can monitor the health of a Red Hat Ceph Storage cluster from a low-level perspective. Low-level monitoring typically involves ensuring that Ceph OSDs are peering properly. When peering faults occur, placement groups operate in a degraded state. This degraded state can be the result of many different things, such as hardware failure, a hung or crashed Ceph daemon, network latency, or a complete site outage.

3.3.1. Prerequisites

  • A running Red Hat Ceph Storage cluster.

3.3.2. Monitoring Placement Group Sets

When CRUSH assigns placement groups to OSDs, it looks at the number of replicas for the pool and assigns the placement group to OSDs such that each replica of the placement group gets assigned to a different OSD. For example, if the pool requires three replicas of a placement group, CRUSH may assign them to osd.1, osd.2 and osd.3 respectively. CRUSH actually seeks a pseudo-random placement that will take into account failure domains you set in the CRUSH map, so you will rarely see placement groups assigned to nearest neighbor OSDs in a large cluster. We refer to the set of OSDs that should contain the replicas of a particular placement group as the Acting Set. In some cases, an OSD in the Acting Set is down or otherwise not able to service requests for objects in the placement group. When these situations arise, don’t panic. Common examples include:

  • You added or removed an OSD. Then, CRUSH reassigned the placement group to other OSDs—​thereby changing the composition of the Acting Set and spawning the migration of data with a "backfill" process.
  • An OSD was down, was restarted and is now recovering.
  • An OSD in the Acting Set is down or unable to service requests, and another OSD has temporarily assumed its duties.

Ceph processes a client request using the Up Set, which is the set of OSDs that will actually handle the requests. In most cases, the Up Set and the Acting Set are virtually identical. When they are not, it may indicate that Ceph is migrating data, an OSD is recovering, or that there is a problem, that is, Ceph usually echoes a HEALTH WARN state with a "stuck stale" message in such scenarios.

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. To retrieve a list of placement groups:

    [root@mon ~]# ceph pg dump
  2. To view which OSDs are in the Acting Set or in the Up Set for a given placement group:

    [root@mon ~]# ceph pg map PG_NUM

    The result should tell you the osdmap epoch, eNNN, the placement group number, PG_NUM, the OSDs in the Up Set up[], and the OSDs in the acting set, acting[]:

    [root@mon ~]# ceph osdmap eNNN pg PG_NUM-> up [0,1,2] acting [0,1,2]
    Note

    If the Up Set and Acting Set do not match, this may be an indicator that the cluster rebalancing itself or of a potential problem with the cluster.

3.3.3. Ceph OSD peering

Before you can write data to a placement group, it must be in an active state, and it should be in a clean state. For Ceph to determine the current state of a placement group, the primary OSD of the placement group that is, the first OSD in the acting set, peers with the secondary and tertiary OSDs to establish agreement on the current state of the placement group. Assuming a pool with 3 replicas of the PG.

Peering

3.3.4. Placement Group States

If you execute a command such as ceph health, ceph -s or ceph -w, you may notice that the cluster does not always echo back HEALTH OK. After you check to see if the OSDs are running, you should also check placement group states. You should expect that the cluster will NOT echo HEALTH OK in a number of placement group peering-related circumstances:

  • You have just created a pool and placement groups haven’t peered yet.
  • The placement groups are recovering.
  • You have just added an OSD to or removed an OSD from the cluster.
  • You have just modified the CRUSH map and the placement groups are migrating.
  • There is inconsistent data in different replicas of a placement group.
  • Ceph is scrubbing a placement group’s replicas.
  • Ceph doesn’t have enough storage capacity to complete backfilling operations.

If one of the foregoing circumstances causes Ceph to echo HEALTH WARN, don’t panic. In many cases, the cluster will recover on its own. In some cases, you may need to take action. An important aspect of monitoring placement groups is to ensure that when the cluster is up and running that all placement groups are active, and preferably in the clean state.

To see the status of all placement groups, execute:

[root@mon ~]# ceph pg stat

The result should tell you the placement group map version, vNNNNNN, the total number of placement groups, x, and how many placement groups, y, are in a particular state such as active+clean:

vNNNNNN: x pgs: y active+clean; z bytes data, aa MB used, bb GB / cc GB avail
Note

It is common for Ceph to report multiple states for placement groups.

Snapshot Trimming PG States

When snapshots exist, two additional PG states will be reported.

  • snaptrim : The PGs are currently being trimmed
  • snaptrim_wait : The PGs are waiting to be trimmed

Example Output:

244 active+clean+snaptrim_wait
 32 active+clean+snaptrim

In addition to the placement group states, Ceph will also echo back the amount of data used, aa, the amount of storage capacity remaining, bb, and the total storage capacity for the placement group. These numbers can be important in a few cases:

  • You are reaching the near full ratio or full ratio.
  • Your data isn’t getting distributed across the cluster due to an error in the CRUSH configuration.

Placement Group IDs

Placement group IDs consist of the pool number, and not the pool name, followed by a period (.) and the placement group ID—​a hexadecimal number. You can view pool numbers and their names from the output of ceph osd lspools. The default pool names data, metadata and rbd correspond to pool numbers 0, 1 and 2 respectively. A fully qualified placement group ID has the following form:

POOL_NUM.PG_ID

Example output:

0.1f
  • To retrieve a list of placement groups:

    [root@mon ~]# ceph pg dump
  • To format the output in JSON format and save it to a file:

    [root@mon ~]# ceph pg dump -o FILE_NAME --format=json
  • To query a particular placement group:

    [root@mon ~]# ceph pg POOL_NUM.PG_ID query

    Example output in JSON format:

    {
      "state": "active+clean",
      "up": [
        1,
        0
      ],
      "acting": [
        1,
        0
      ],
      "info": {
        "pgid": "1.e",
        "last_update": "4'1",
        "last_complete": "4'1",
        "log_tail": "0'0",
        "last_backfill": "MAX",
        "purged_snaps": "[]",
        "history": {
          "epoch_created": 1,
          "last_epoch_started": 537,
          "last_epoch_clean": 537,
          "last_epoch_split": 534,
          "same_up_since": 536,
          "same_interval_since": 536,
          "same_primary_since": 536,
          "last_scrub": "4'1",
          "last_scrub_stamp": "2013-01-25 10:12:23.828174"
        },
        "stats": {
          "version": "4'1",
          "reported": "536'782",
          "state": "active+clean",
          "last_fresh": "2013-01-25 10:12:23.828271",
          "last_change": "2013-01-25 10:12:23.828271",
          "last_active": "2013-01-25 10:12:23.828271",
          "last_clean": "2013-01-25 10:12:23.828271",
          "last_unstale": "2013-01-25 10:12:23.828271",
          "mapping_epoch": 535,
          "log_start": "0'0",
          "ondisk_log_start": "0'0",
          "created": 1,
          "last_epoch_clean": 1,
          "parent": "0.0",
          "parent_split_bits": 0,
          "last_scrub": "4'1",
          "last_scrub_stamp": "2013-01-25 10:12:23.828174",
          "log_size": 128,
          "ondisk_log_size": 128,
          "stat_sum": {
            "num_bytes": 205,
            "num_objects": 1,
            "num_object_clones": 0,
            "num_object_copies": 0,
            "num_objects_missing_on_primary": 0,
            "num_objects_degraded": 0,
            "num_objects_unfound": 0,
            "num_read": 1,
            "num_read_kb": 0,
            "num_write": 3,
            "num_write_kb": 1
          },
          "stat_cat_sum": {
    
          },
          "up": [
            1,
            0
          ],
          "acting": [
            1,
            0
          ]
        },
        "empty": 0,
        "dne": 0,
        "incomplete": 0
      },
      "recovery_state": [
        {
          "name": "Started\/Primary\/Active",
          "enter_time": "2013-01-23 09:35:37.594691",
          "might_have_unfound": [
    
          ],
          "scrub": {
            "scrub_epoch_start": "536",
            "scrub_active": 0,
            "scrub_block_writes": 0,
            "finalizing_scrub": 0,
            "scrub_waiting_on": 0,
            "scrub_waiting_on_whom": [
    
            ]
          }
        },
        {
          "name": "Started",
          "enter_time": "2013-01-23 09:35:31.581160"
        }
      ]
    }

Additional Resources

  • See the chapter Object Storage Daemon (OSD) configuration options in the Red Hat Ceph Storage 4 Configuration Guide for more details on the snapshot trimming settings.

3.3.5. Placement Group creating state

When you create a pool, it will create the number of placement groups you specified. Ceph will echo creating when it is creating one or more placement groups. Once they are created, the OSDs that are part of a placement group’s Acting Set will peer. Once peering is complete, the placement group status should be active+clean, which means a Ceph client can begin writing to the placement group.

Creating PGs

3.3.6. Placement group peering state

When Ceph is Peering a placement group, Ceph is bringing the OSDs that store the replicas of the placement group into agreement about the state of the objects and metadata in the placement group. When Ceph completes peering, this means that the OSDs that store the placement group agree about the current state of the placement group. However, completion of the peering process does NOT mean that each replica has the latest contents.

Authoritative History

Ceph will NOT acknowledge a write operation to a client, until all OSDs of the acting set persist the write operation. This practice ensures that at least one member of the acting set will have a record of every acknowledged write operation since the last successful peering operation.

With an accurate record of each acknowledged write operation, Ceph can construct and disseminate a new authoritative history of the placement group. A complete, and fully ordered set of operations that, if performed, would bring an OSD’s copy of a placement group up to date.

3.3.7. Placement group active state

Once Ceph completes the peering process, a placement group may become active. The active state means that the data in the placement group is generally available in the primary placement group and the replicas for read and write operations.

3.3.8. Placement Group clean state

When a placement group is in the clean state, the primary OSD and the replica OSDs have successfully peered and there are no stray replicas for the placement group. Ceph replicated all objects in the placement group the correct number of times.

3.3.9. Placement Group degraded state

When a client writes an object to the primary OSD, the primary OSD is responsible for writing the replicas to the replica OSDs. After the primary OSD writes the object to storage, the placement group will remain in a degraded state until the primary OSD has received an acknowledgement from the replica OSDs that Ceph created the replica objects successfully.

The reason a placement group can be active+degraded is that an OSD may be active even though it doesn’t hold all of the objects yet. If an OSD goes down, Ceph marks each placement group assigned to the OSD as degraded. The OSDs must peer again when the OSD comes back online. However, a client can still write a new object to a degraded placement group if it is active.

If an OSD is down and the degraded condition persists, Ceph may mark the down OSD as out of the cluster and remap the data from the down OSD to another OSD. The time between being marked down and being marked out is controlled by mon_osd_down_out_interval, which is set to 600 seconds by default.

A placement group can also be degraded, because Ceph cannot find one or more objects that Ceph thinks should be in the placement group. While you cannot read or write to unfound objects, you can still access all of the other objects in the degraded placement group.

Let’s say there are 9 OSDs in a three way replica pool. If OSD number 9 goes down, the PGs assigned to OSD 9 go in a degraded state. If OSD 9 doesn’t recover, it goes out of the cluster and the cluster rebalances. In that scenario, the PGs are degraded and then recover to an active state.

3.3.10. Placement Group recovering state

Ceph was designed for fault-tolerance at a scale where hardware and software problems are ongoing. When an OSD goes down, its contents may fall behind the current state of other replicas in the placement groups. When the OSD is back up, the contents of the placement groups must be updated to reflect the current state. During that time period, the OSD may reflect a recovering state.

Recovery isn’t always trivial, because a hardware failure might cause a cascading failure of multiple OSDs. For example, a network switch for a rack or cabinet may fail, which can cause the OSDs of a number of host machines to fall behind the current state of the cluster. Each one of the OSDs must recover once the fault is resolved.

Ceph provides a number of settings to balance the resource contention between new service requests and the need to recover data objects and restore the placement groups to the current state. The osd recovery delay start setting allows an OSD to restart, re-peer and even process some replay requests before starting the recovery process. The osd recovery threads setting limits the number of threads for the recovery process, by default one thread. The osd recovery thread timeout sets a thread timeout, because multiple OSDs may fail, restart and re-peer at staggered rates. The osd recovery max active setting limits the number of recovery requests an OSD will entertain simultaneously to prevent the OSD from failing to serve . The osd recovery max chunk setting limits the size of the recovered data chunks to prevent network congestion.

3.3.11. Back fill state

When a new OSD joins the cluster, CRUSH will reassign placement groups from OSDs in the cluster to the newly added OSD. Forcing the new OSD to accept the reassigned placement groups immediately can put excessive load on the new OSD. Backfilling the OSD with the placement groups allows this process to begin in the background. Once backfilling is complete, the new OSD will begin serving requests when it is ready.

During the backfill operations, you may see one of several states: * backfill_wait indicates that a backfill operation is pending, but isn’t underway yet * backfill indicates that a backfill operation is underway * backfill_too_full indicates that a backfill operation was requested, but couldn’t be completed due to insufficient storage capacity.

When a placement group cannot be backfilled, it may be considered incomplete.

Ceph provides a number of settings to manage the load spike associated with reassigning placement groups to an OSD, especially a new OSD. By default, osd_max_backfills sets the maximum number of concurrent backfills to or from an OSD to 10. The osd backfill full ratio enables an OSD to refuse a backfill request if the OSD is approaching its full ratio, by default 85%. If an OSD refuses a backfill request, the osd backfill retry interval enables an OSD to retry the request, by default after 10 seconds. OSDs can also set osd backfill scan min and osd backfill scan max to manage scan intervals, by default 64 and 512.

For some workloads, it is beneficial to avoid regular recovery entirely and use backfill instead. Since backfilling occurs in the background, this allows I/O to proceed on the objects in the OSD. To force backfill rather than recovery, set osd_min_pg_log_entries to 1, and set osd_max_pg_log_entries to 2. Contact your Red Hat Support account team for details on when this situation is appropriate for your workload.

3.3.12. Changing the priority of recovery or backfill operations

You might encounter a situation where some placement groups (PGs) require recovery and/or backfill, and some of those placement groups contain more important data than do others. Use the pg force-recovery or pg force-backfill command to ensure that the PGs with the higher-priority data undergo recovery or backfill first.

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. Issue the pg force-recovery or pg force-backfill command and specify the order of priority for the PGs with the higher-priority data:

    Syntax

    ceph pg force-recovery PG1 [PG2] [PG3 ...]
    ceph pg force-backfill PG1 [PG2] [PG3 ...]

    Example

    [root@node]# ceph pg force-recovery group1 group2
    [root@node]# ceph pg force-backfill group1 group2

    This command causes Red Hat Ceph Storage to perform recovery or backfill on specified placement groups (PGs) first, before processing other placement groups. Issuing the command does not interrupt backfill or recovery operations that are currently executing. After the currently running operations have finished, recovery or backfill takes place as soon as possible for the specified PGs.

3.3.13. Changing or canceling a recovery or backfill operation on specified placement groups

If you cancel a high-priority force-recovery or force-backfill operation on certain placement groups (PGs) in a storage cluster, operations for those PGs revert to the default recovery or backfill settings.

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. To change or cancel a recovery or backfill operation on specified placement groups:

    Syntax

    ceph pg cancel-force-recovery PG1 [PG2] [PG3 ...]
    ceph pg cancel-force-backfill PG1 [PG2] [PG3 ...]

    Example

    [root@node]# ceph pg cancel-force-recovery group1 group2
    [root@node]# ceph pg cancel-force-backfill group1 group2

    This cancels the force flag and processes the PGs in the default order.

    After recovery or backfill operations for the specified PGs have completed, processing order reverts to the default.

Additional Resources

3.3.14. Forcing high-priority recovery or backfill operations for pools

If all of the placement groups in a pool require high-priority recovery or backfill, use the force-recovery or force-backfill options to initiate the operation.

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. To force the high-priority recovery or backfill on all placement groups in a specified pool:

    Syntax

    ceph osd pool force-recovery POOL_NAME
    ceph osd pool force-backfill POOL_NAME

    Example

    [root@node]# ceph osd pool force-recovery pool1
    [root@node]# ceph osd pool force-backfill pool1

    Note

    Use the force-recovery and force-backfill commands with caution. Changing the priority of these operations might break the ordering of Ceph’s internal priority computations.

3.3.15. Canceling high-priority recovery or backfill operations for pools

If you cancel a high-priority force-recovery or force-backfill operation on all placement groups in a pool, operations for the PGs in that pool revert to the default recovery or backfill settings.

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. To cancel a high-priority recovery or backfill operation on all placement groups in a specified pool:

    Syntax

    ceph osd pool cancel-force-recovery POOL_NAME
    ceph osd pool cancel-force-backfill POOL_NAME

    Example

    [root@node]# ceph osd pool cancel-force-recovery pool1
    [root@node]# ceph osd pool cancel-force-backfill pool1

3.3.16. Rearranging the priority of recovery or backfill operations for pools

If you have multiple pools that currently use the same underlying OSDs and some of the pools contain high-priority data, you can rearrange the order in which the operations execute. Use the recovery_priority option to assign a higher priority value to the pools with the higher-priority data. Those pools will execute before pools with lower priority values, or pools that are set to default priority.

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. To rearrange the recovery/backfill priority for the pools:

    Syntax

    ceph osd pool set POOL_NAME recovery_priority VALUE

    Example

    ceph osd pool set pool1 recovery_priority 10

    VALUE sets the order of priority. For example, if you have 10 pools, the pool with a priority value of 10 gets processed first, followed by the pool with priority 9, and so on. If only some pools have high priority, you can set priority values for just those pools. The pools without set priority values are processed in the default order.

3.3.17. Priority of placement group recovery in RADOS

This section describes the relative priority values for the recovery and backfilling of placement groups (PGs) in RADOS. Higher values are processed first. Inactive PGs receive higher priority values than active or degraded PGs.

OperationValueDescription

OSD_RECOVERY_PRIORITY_MIN

0

Minimum recovery value

OSD_BACKFILL_PRIORITY_BASE

100

Base backfill priority for MBackfillReserve

OSD_BACKFILL_DEGRADED_PRIORITY_BASE

140

Base backfill priority for MBackfillReserve (degraded PG)

OSD_RECOVERY_PRIORITY_BASE

180

Base recovery priority for MBackfillReserve

OSD_BACKFILL_INACTIVE_PRIORITY_BASE

220

Base backfill priority for MBackfillReserve (inactive PG)

OSD_RECOVERY_INACTIVE_PRIORITY_BASE

220

Base recovery priority for MRecoveryReserve (inactive PG)

OSD_RECOVERY_PRIORITY_MAX

253

Max manually/automatically set recovery priority for MBackfillReserve

OSD_BACKFILL_PRIORITY_FORCED

254

Backfill priority for MBackfillReserve, when forced manually

OSD_RECOVERY_PRIORITY_FORCED

255

Recovery priority for MRecoveryReserve, when forced manually

OSD_DELETE_PRIORITY_NORMAL

179

Priority for PG deletion when the OSD is not fullish

OSD_DELETE_PRIORITY_FULLISH

219

Priority for PG deletion when the OSD is approaching full

OSD_DELETE_PRIORITY_FULL

255

Priority for deletion when the OSD is full

3.3.18. Placement Group remapped state

When the Acting Set that services a placement group changes, the data migrates from the old acting set to the new acting set. It may take some time for a new primary OSD to service requests. So it may ask the old primary to continue to service requests until the placement group migration is complete. Once data migration completes, the mapping uses the primary OSD of the new acting set.

3.3.19. Placement Group stale state

While Ceph uses heartbeats to ensure that hosts and daemons are running, the ceph-osd daemons may also get into a stuck state where they aren’t reporting statistics in a timely manner. For example, a temporary network fault. By default, OSD daemons report their placement group, up thru, boot and failure statistics every half second, that is, 0.5, which is more frequent than the heartbeat thresholds. If the Primary OSD of a placement group’s acting set fails to report to the monitor or if other OSDs have reported the primary OSD down, the monitors will mark the placement group stale.

When you start the storage cluster, it is common to see the stale state until the peering process completes. After the storage cluster has been running for awhile, seeing placement groups in the stale state indicates that the primary OSD for those placement groups is down or not reporting placement group statistics to the monitor.

3.3.20. Placement Group misplaced state

There are some temporary backfilling scenarios where a PG gets mapped temporarily to an OSD. When that temporary situation should no longer be the case, the PGs might still reside in the temporary location and not in the proper location. In which case, they are said to be misplaced. That’s because the correct number of extra copies actually exist, but one or more copies is in the wrong place.

For example, there are 3 OSDs: 0,1,2 and all PGs map to some permutation of those three. If you add another OSD (OSD 3), some PGs will now map to OSD 3 instead of one of the others. However, until OSD 3 is backfilled, the PG will have a temporary mapping allowing it to continue to serve I/O from the old mapping. During that time, the PG is misplaced, because it has a temporary mapping, but not degraded, since there are 3 copies.

Example

pg 1.5: up=acting: [0,1,2]
ADD_OSD_3
pg 1.5: up: [0,3,1] acting: [0,1,2]

[0,1,2] is a temporary mapping, so the up set is not equal to the acting set and the PG is misplaced but not degraded since [0,1,2] is still three copies.

Example

pg 1.5: up=acting: [0,3,1]

OSD 3 is now backfilled and the temporary mapping is removed, not degraded and not misplaced.

3.3.21. Placement Group incomplete state

A PG goes into a incomplete state when there is incomplete content and peering fails, that is, when there are no complete OSDs which are current enough to perform recovery.

Lets say OSD 1, 2, and 3 are the acting OSD set and it switches to OSD 1, 4, and 3, then osd.1 will request a temporary acting set of OSD 1, 2, and 3 while backfilling 4. During this time, if OSD 1, 2, and 3 all go down, osd.4 will be the only one left which might not have fully backfilled all the data. At this time, the PG will go incomplete indicating that there are no complete OSDs which are current enough to perform recovery.

Alternately, if osd.4 is not involved and the acting set is simply OSD 1, 2, and 3 when OSD 1, 2, and 3 go down, the PG would likely go stale indicating that the mons have not heard anything on that PG since the acting set changed. The reason being there are no OSDs left to notify the new OSDs.

3.3.22. Identifying stuck Placement Groups

As previously noted, a placement group isn’t necessarily problematic just because its state isn’t active+clean. Generally, Ceph’s ability to self repair may not be working when placement groups get stuck. The stuck states include:

  • Unclean: Placement groups contain objects that are not replicated the desired number of times. They should be recovering.
  • Inactive: Placement groups cannot process reads or writes because they are waiting for an OSD with the most up-to-date data to come back up.
  • Stale: Placement groups are in an unknown state, because the OSDs that host them have not reported to the monitor cluster in a while, and can be configured with the mon osd report timeout setting.

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. To identify stuck placement groups, execute the following:

    ceph pg dump_stuck {inactive|unclean|stale|undersized|degraded [inactive|unclean|stale|undersized|degraded...]} {<int>}

3.3.23. Finding an object’s location

The Ceph client retrieves the latest cluster map and the CRUSH algorithm calculates how to map the object to a placement group, and then calculates how to assign the placement group to an OSD dynamically.

Prerequisites

  • A running Red Hat Ceph Storage cluster.
  • Root-level access to the node.

Procedure

  1. To find the object location, all you need is the object name and the pool name:

    ceph osd map POOL_NAME OBJECT_NAME