Article ID: 2178, created on Sep 19, 2007, last review on Nov 2, 2016

  • Applies to:
  • Odin Business Automation Standard 4.5

Symptoms

The following warning message is shown in the Provider Control Center:

The number of used inodes or free disk space at one of the partitions has exceeded 80%

Cause

The reason for this warning message is that the disk usage is approaching the allowed disk usage capacity inside the Odin Business Automation Standard (OBAS) node.

Limit DISKINODES of OBAS container is too small:

~# grep DISKINODES /etc/sysconfig/vz-scripts/CTID.conf
DISKINODES="100000:110000"

Where CTID is id of OBAS container.

To check the current disk space usage, you may use the following command on the OBAS server:

~# df -h

To check the current inodes (number of files created) usage, you may use the following command on the OBAS server:

~# df -i

Resolution

  • Increase DISKINODES for OBAS container. According to the Installation Guide, OBAS needs 2GB of RAM at least. VC node provides examples of resource configuration for container with 2GB RAM required for normal operations:

    ~# egrep 'SLMMODE|SLMMEMORYLIMIT|DISKINODES' /etc/sysconfig/vz-scripts/ve-slm.2048MB.conf-sample
    DISKINODES="200000:220000"
    SLMMODE="all"
    SLMMEMORYLIMIT="2147483648:2147483648"
    ~#
    

    Increase the limit to 200000 at least, but 1000000 would be better for big OBAS node. Run the following command from the Hardware Node:

    ~# vzctl set CTID --save --diskspace NEW_LIMIT 
    
  • Delete old problem reports under the /var/log/hspc/ directory (files named repYYYY-MM-DD_HH-MM-SS.gz or pYYYY-MM-DD_HH-MM-SS.gz).

  • Decrease the period that problem reports are kept for -- adjust the value of the parameter REPORT_EXPIRES (in days) in the configuration file /etc/hspc/hspc.conf on the OBAS server (for this setting to to come into force hspcd service restart is required, it will cause OBAS downtime for about 10-60 seconds).
  • Turn off MySQL binary logging and delete existing MySQL binary logs:

    • Run the utility mysql as root and execute the following command:

      mysql> reset master;
      
    • Comment the option log-bin in the MySQL configuration file /etc/my.cnf (changes will be applied only after restarting MySQL)
  • Check if logrotate is working correctly as described in this article

Search Words

oba-s disk full

disk space

mysql disk space

release disc space

clean disc space

No task - but hspc.log file is getting huge

disk space exceeding 80%

alert

p2016*

disk space at one of the partitions has exceeded 80%

oda-s full

400e18f6ede9f8be5575a475d2d6b0a6 caea8340e2d186a540518d08602aa065 70a5401e8b9354cd1d64d0346f2c4a3e 624ca542e40215e6f1d39170d8e7ec75

Email subscription for changes to this article
Save as PDF