Article ID: 117630, created on Oct 1, 2013, last review on Jun 17, 2016

  • Applies to:
  • Virtuozzo


This video helps to monitor your Parallels Cloud Storage cluster health.

This video is the part of Parallels Cloud Storage course:




~# pstorage -c clustername top

Monitoring General Cluster Parameters

  • Cluster - cluster name and status
    • Healthy - all chunk servers are active
    • Unknown - not enough information about cluster state
    • Degraded - some chunk servers are inactive
    • Failure - too many inactive chunk server, automatic replication is off
  • Space - amount of disk space in cluster
    • Total - total amount
    • Free - free disk space
    • Allocatable - amount of logical space available to clients
  • MDS nodes - Metadata servers information
    • Epoch time - time elapsed since Master MDS election
  • CS nodes - Chunk servers information
    • Active - active chunk servers that are up and running
    • Inactive - chunk servers that are temporarily unavailable
    • Offline - chunk servers which have been inactive for 5 minutes
  • License - Cloud Storage license status
    • Key - product key number
    • Status - license status
  • IO - disk I/O activity

Monitoring Metadata Servers

  • MDSID - MDS server identifier
    • M - server is master MDS server
  • STATUS - MDS server status
  • %CTIME - total time MDS spent writing to local journal
  • COMMITS - local journal commit rate
  • %CPU - MDS server activity time
  • MEM - amount of physical memory used by MDS
  • UPTIME - time elapsed since the last MDS server start
  • HOST - MDS server hostname or IP

Monitoring Chunk Servers

  • CSID - chunk server ID
  • STATUS - chunk server status
    • active - server up and running
    • inactive - server is not available
    • offline - server is unavailable more than 5 minutes
    • dropped - server was removed by user
  • SPACE - total amount of space on CS
  • FREE - free disk space on CS
  • REPLICAS - number of replicas stored on CS
  • IOWAIT - percentage of time spent waiting for IO operations being served.
  • IOLAT - average time in milliseconds client needs to complete single IO operation
  • HOST - CS hostname or IP

TIP: use ā€˜Iā€™ key to scroll displayed columns

Disk Space Usage

Total space - total capacity of HDDs.

Free space - unused space on HDDs.

Allocatable space - space which can be used by clients. Depends on the smallest chunk server capacity and replication settings.

Monitoring Client Machines

  • CLID - client ID
  • LEASES - average number opened for reading writing and not yet closed
  • READ - average read operation speed
  • WRITE - average write operation speed
  • RD_OPS - average number of read operations per second
  • WR_OPS - average number of write operations per second
  • FSYNCS - average number of sync operations per second
  • IOLAT - average time in milliseconds required for single IO operation
  • HOST - client hostname or IP

Monitoring Event Logs

  • Time - time when event happened
  • Sys - component triggered the event
  • Sev - event severity
    • ERR - error
    • WRN - warning
    • INF - info
  • Message - event description

Monitoring Replication Parameters

  • Healthy - percentage of chunks that have enough active replicas
  • Degraded - percentage of chunks that do not have enough active replicas
  • Overcommited - percentage of chunks that have an excessive amount of replicas
  • Replicating - percentage of chunks whose replicas are being replicated
  • Offline - percentage of chunks that have no active replicas
  • Standby - percentage of replicas modified while being inactive
  • Void - percentage of chunks that have one or more replicas whose state cannot be parsed
  • Pending - percentage of chunks that must be replicated in the first place because all operations with these chunks are suspended and the client is waiting for the replication to complete
  • Blocked - percentage of chunks whose number of replicas is below the minimum replicas limit. No write operations to these chunks are allowed
  • Urgent - percentage of chunks that are nearing the minimum replicas limit

Additional Information

Documentation Portal

Official product page

Search Words

pstorage event monitoring

Chunks: [Warning] 153982 (99%) healthy, 4 (0%) standby,

MDS INF Replication started,

Chunks: [Warning] (99%) healthy, 4 (0%) standby,

0dd5b9380c7d4884d77587f3eb0fa8ef 2897d76d56d2010f4e3a28f864d69223

Email subscription for changes to this article
Save as PDF