Sun Microsystems, Inc.  Oracle System Handbook - ISO 7.0 May 2018 Internal/Partner Edition
   Home | Current Systems | Former STK Products | EOL Systems | Components | General Info | Search | Feedback

Asset ID: 1-72-2295143.1
Update Date:2017-10-10
Keywords:

Solution Type  Problem Resolution Sure

Solution  2295143.1 :   ODA Unable to start ASM or Unable to start GI, TFA Hang, ORA-09925:Linux-x86_64 Error: 28: No space left on device w/ Huge Number Of Files Created  


Related Items
  • Oracle Database Appliance
  •  
Related Categories
  • PLA-Support>Eng Systems>Exadata/ODA/SSC>Oracle Database Appliance>DB: ODA_EST
  •  


Unable to start GI or ASM :  ORA-09925 Huge Number Of Files Found Under  /u01/app/grid/diag/clients/user_grid/host_XX on one node.  (Usually the oakd master node)
Other directories may be also have excessive trace or aud$ files

In this Document
Symptoms
Changes
Cause
Solution
References


Created from <SR 3-14301560291>

Applies to:

Oracle Database Appliance - Version All Versions to All Versions [Release All Releases]
Information in this document applies to any platform.

Symptoms

  • Unable to start ASM or Unable to start GI
  • TFA or other data collection tools nearly hang or do not respond or return outputs
  • The node shows out of space messages or unable to write .AUD files
  • ORA-09925: Unable to create audit trail file
  • Linux-x86_64 Error: 28: No space left on device
  • Huge Number Of Files Created Under  /u01/app/grid/diag/clients/user_grid/host_XX on one node. (Usually the oakd master node)

Here is one example GI cannot startup:

crsctl stat res -init  

showing:

  NAME=ora.crsd
  TYPE=ora.crs.type
  TARGET=ONLINE
  STATE=INTERMEDIATE  on oda-n2-base

Also ASM and storage showing online but  we find no running ASM process

ps -ef|grep smon
                           < -- No ASM detected


Log into asm results in the following errors

  • "No space left on device"
  • "ORA-09925: Unable to create audit trail file":

$ sqlplus / as sysasm

 SQL*Plus: Release 12.1.0.2.0 Production on Fri ...2017
 Copyright (c) 1982, 2014, Oracle. All rights reserved.

  ERROR:
  ORA-09925: Unable to create audit trail file
  Linux-x86_64 Error: 28: No space left on device
  Additional information: 9925

 

$ df -k   -- Confirms /u01 has 21% FREE

$ df -k

Filesystem      1K-blocks        Used       Available   Use%   Mounted on

/dev/xvda2      57060636      42932984      11222412     80%     /
     tmpfs      65879408             0      65879408      0%     /dev/shm
/dev/xvda1        462820         67036        371464     16%     /boot
/dev/xvdb1      95988492      71225684      19880120     79% <<  /u01    

 

$ df -i  -- Confirms that the we have used 100% of the Inodes

$ df -i 

Filesystem    Inodes      IUsed     IFree   IUse%   Mounted on

/dev/xvda2   3637248      61310    3575938     2%      /
     tmpfs  16469852          1   16469851     1%     /dev/shm
/dev/xvda1    121920         46     121874     1%     /boot
/dev/xvdb1   6111232    6111232          0   100% <<  /u01

 

Changes

 

Cause

Excessive file generation which are not monitored or purged can result in running out of Inodes.

On the ODA this can also be due to INTERNAL Bug 25614866 generating a high number traces  in /u01

Once Inodes are 100% for /u01 we are unable to start ASM as it shows as

 ORA-09925: Unable to create audit trail file
  Linux-x86_64 Error: 28: No space left on device
  Additional information: 9925

 

Solution

1) Confirm the directory running out of Inodes using $ df -i 

2) Clear the excessive number of files

  • 2081655.1  ODA (Oracle Database Appliance): Cleanup trace and log files Using purgeODALog Tool 
  • 1681062.1  How to clean up the old staged patches on ODA  
  • 2154918.1  ODA: "/opt" Filesystem is Running Out of Space 99% Full in ODA Configurations 


3) confirm Inodes are available for the previous full directory

[root@oda-n2-base ~]# df -i

Filesystem      Inodes   IUsed   IFree    IUse%  Mounted on

/dev/xvda2     3637248   61310   3575938     2%  /

     tmpfs    16469852       1  16469851     1%  /dev/shm
/dev/xvda1      121920      46    121874     1%  /boot
/dev/xvdb1     6111232 5343835    767397    88%  /u01     << down from 100%

 

4) Confirm that connections to ASM can work and that the GI can restart successfully.

 

References

<BUG:25614866> - HUGE NUMBER OF FILES CREATED UNDER /U01/APP/GRID/ORADIAG_GRID/DIAG/CLIENTS/USER
<BUG:21347974> - ODA: GRID INFRASTRUCTURE GENERATES MORE TRACE FILES
<NOTE:2081655.1> - ODA (Oracle Database Appliance): Cleanup trace and log files Using purgeODALog Tool
<BUG:25664107> - DIAGNOSTIC PATCH FOR BUG 25614866
<NOTE:1681062.1> - How to clean up the old staged patches on ODA
<NOTE:2154918.1> - ODA: "/opt" Filesystem is Running Out of Space 99% Full in ODA Configurations

Attachments
This solution has no attachment
  Copyright © 2018 Oracle, Inc.  All rights reserved.
 Feedback