No space left on device" while fileystem has lots of free space - Inode header full - Unix

No space left on device" while fileystem has lots of free space . Sometime it happens due to inode header is full. Unix keep tack of file details in the inode header. incase if too many files under a file-system or  folder may cause the inode header to become full.


-- df shows it has 61% free space
df -h .
Filesystem            Size  Used Avail Use% Mounted on
/dev/mapper/rootvg-u01
                      9.8G  5.6G  3.7G  61% /u01


-- df -I  shows it has 100% full.
df -i .
Filesystem           Inodes  IUsed IFree IUse% Mounted on
/dev/mapper/rootvg-u01
                     655360 655360     0  100% /u01






use the below command to find out the folder with huge files




find /u01 -xdev -printf '%h\n' | sort | uniq -c | sort -k 1 -n




we may need to either house keep or move it to different folder the unwanted files to make the inode header free

Unix rm command returns 'bin/rm: cannot execute [Argument list too long]' when the list of files are more to hanle by rm buffer

-ksh: rm: /bin/rm: cannot execute [Argument list too long]




Work-around:


find . -name '*.trm' | xargs rm -f

How to open ASM trace files using vi

When we open ASM trace files using vi command


we may get the following error


Error detected while processing command line:
E492: Not an editor command: ASM1_arb0_396061.trc
Press ENTER or type command to continue








to overcome this challenge, we can use -- option in vi command


vi -- +ASM_arb0_396061.trc

SSH script to access multiple servers

#!/bin/ksh
#set -x
LOGFILE=/tmp/scan.log
echo " Scan test` date ` " > $LOGFILE
echo " " >>$LOGFILE
echo " " >>$LOGFILE
echo "==================================================================================================================" >>$LOGFILE
for i in ` cat /tmp/server.lst` ; do
server=`echo $i | awk -F\~ ' { print $1} ' `
echo Scan Status for $server  | tee -a $LOGFILE
/usr/sbin/ping $server
var_ping=$?
if [[ $var_ping -eq 0 ]] ; then
    ssh -o batchmode=yes $server true
    var_date=$?
    if [[ $var_date -eq 0 ]] ; then
      echo $server id | tee -a $LOGFILE
      ssh $server id | tee -a $LOGFILE
      echo "==================================================================================================================" >>$LOGFILE
    else
      echo $server ========== not Accessable =========== >>$LOGFILE2
    fi
else
  echo $server ========== not pingable =========== >>$LOGFILE2
fi
done

Unix Ulitity to grep a search string with few lines before and after the match

OEL Linux has an option in  grep command to search a string with some lines before and after the  match.
This utility helps us to search an ERROR  from the alert.log/trace with timestamp which is recorded one line before the actual error.
Unfortunately this option is not available in some of the platform like the Solaris and AIX ( of course  there are addition packages to have enhanced grep option)
The following small wrapper script to simulate the same grep option available in Linux







Script:
# usage <grep_add.sh> <no_rows_b4> <no_rows_a4> <search_string> <file_name>
b4=$1
a4=$2
search_string=$3
file_name=$4
grep -n $search_string $file_name | cut -d':' -f1 |  xargs  -n1 -I % awk "NR<=%+$a4 && NR>=%-$b4" $file_name




==========================================================
Usage:
grep_add.sh 1 0 force_match coe_xfr_sql_profile_56c1s0454r13x_1115266317.sql
replace     => TRUE,
force_match => FALSE /* TRUE:FORCE (match even when different literals in SQL). FALSE:EXACT (similar to CURSOR_SHARING) */ );
grep_add.sh  2 2 force_match coe_xfr_sql_profile_56c1s0454r13x_1115266317.sql
validate    => TRUE,
replace     => TRUE,
force_match => FALSE /* TRUE:FORCE (match even when different literals in SQL). FALSE:EXACT (similar to CURSOR_SHARING) */ );
DBMS_LOB.FREETEMPORARY(sql_txt);
END;
=======================================================











Linux Option:
In linux ,this option comes with the exisitng grep command , we can use -B and -A option to grep the additional lines
grep -A 3 FIXED awrinfo.txt
FIXED      39,506.7   44.2        3,272      306.7     2,147.2    47% : 53%
EVENTS     32,080.4   35.9        2,657      249.1     1,743.6    39% : 61%
ASH        11,324.3   12.7          938       87.9       615.5    89% : 11%
SPACE       3,275.4    3.7          271       25.4       178.0    68% : 32%
$ grep -B 2 FIXED awrinfo.txt
COMPONENT        MB  % AWR  KB_PER_SNAP MB_PER_DAY MB_PER_WEEK TABLE% : INDEX%
--------- --------- ------ ------------ ---------- ----------- ----------------
FIXED      39,506.7   44.2        3,272      306.7     2,147.2    47% : 53%