No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

Insufficient System Space Caused by Outsize MegaSAS.log Under Root Directories on MDS Nodes

Publication Date:  2014-09-22 Views:  65 Downloads:  0
Issue Description
After df –h is run in a preventive maintenance to discover devices, root directories are detected to occupy 92% of system space, causing the remaining system space to be insufficient.

[root@MDS_30 11:02:00 ~]# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/sda1             9.9G  9.1G  0.8G  92% /
udev                   16G  204K   16G   1% /dev
/dev/sda2             259M   43M  216M  17% /boot
/dev/sda5              11G  8.0G  2.1G  80% /var
/dev/sda7             849G  706M  848G   1% /mnt/wscore
fuse                  131T  5.6T  125T   5% /mnt/CA
98.98.161.30:98.98.161.31:/IDC1
                      131T  5.6T  125T   5% /mnt/CA

Alarm Information
None
Handling Process
1.Set periodic dump for MegaSAS.log. 
  1. Log in to each MDS node as user root and run the mkdir /megaclilog command to create directory /megaclilog.
     
  2. Go to the /etc/logrotate.d directory and run the touch megacli command to create file megacli. Then run the vi megacli command to add the following script: 

/MegaSAS.log {
     daily
     compress
     dateext
     maxage 365
     rotate 99
     olddir /megaclilog/
     missingok
     notifempty
     size +100M
     create 640 root root
     sharedscripts
     postrotate
     endscript
}

After the periodic dump for MegaSAS.log is set, the system automatically compresses MegaSAS.log to directory /megaclilog when MegaSAS.log exceeds 100 MB. The compressed MegaSAS.log is only several hundred KB.

 

Root Cause
After all MDS nodes are checked, it is found that MegaSAS.log under the root directory on each MDS node occupies large system space. MegaSAS.log records the states of disks on MDS nodes every one minute. After long time data query and device operation, MegaSAS.log becomes outsize.

Suggestions
None

END