ll,ll -s, ls -l, du like this any command is getting stuck in few directories.

[purushottamaher]

Hello Experts,

When I'm executing "ll,ll -s, ls -l, du" any commands in my "testdata" directory, It is getting stuck without showing output and its not completing.

I thought, that may me there is problem with the VG so i scan it also but still the problem is same.

Request you to please help on this.

Thanks in Advance.

[sapoguheman]

Check if the filesystem is heavily used also check the disks usage of that file system.

[purushottamaher]

Dear sapoguheman,

The filesystem is 73% used only.

[sapoguheman]

Its not the file system usage in terms of storage.

I was asking to check the performance of the disk (IE IO operation / Heavy IO) which are used by this file sytem.

If you do lvdisplay -v <lv name of the file system> , you will get the list of disks used by your file system.

Figure out how the disk are performing using sar -d command.

[purushottamaher]

Hi,

Thank you for your quick reply.

As you said i have checked the used disk of the filesystem and checked sar -d output but i think it alright.

1st disk output :

# sar -d 1 100 /dev/disk/disk5_p2

HP-UX mahaeccs B.11.31 U ia64    07/09/15

18:21:58   device   %busy   avque   r+w/s  blks/s  avwait  avserv
18:21:59    disk5   32.00    0.50      32     512    0.00   13.71
           c0t4d0    2.00    0.50      40    2432    0.00    0.86
            disk7    2.00    0.50      40    2432    0.00    0.86
           c0t6d0    3.00    0.97      59    7272    0.78    1.67
           disk11    3.00    0.97      59    7272    0.78    1.68
18:22:00    disk5    9.00    0.50      46     784    0.00    2.01
           c0t5d0    6.00    0.53      87   10240    0.20    1.87
            disk9    6.00    0.53      87   10240    0.20    1.87
           c0t6d0    1.00    0.50      25    2968    0.00    0.42
           disk11    1.00    0.50      25    2968    0.00    0.41
18:22:01    disk5    4.00    0.77     105   10576    0.52    1.25
18:22:02    disk5    1.98    0.50      26     491    0.00    0.45
           c0t5d0   17.82   10.47     236   20277    6.54    2.15
            disk9   17.82   10.47     236   20277    6.54    2.15
18:22:03    disk5   16.16    0.68     158   10545    0.34    2.04
18:22:04    disk5    3.03    0.50      18     307    0.00    1.56
           c0t4d0   24.24    0.83      86   10343   15.22   16.20
            disk7   24.24    0.83      86   10343   15.22   16.20
18:22:05    disk5   10.89    0.84     105   10535    0.83    1.77

2nd Disk :

# sar -d 1 100 /dev/dsk/c0t6d0

HP-UX mahaeccs B.11.31 U ia64    07/09/15

18:22:59   device   %busy   avque   r+w/s  blks/s  avwait  avserv
18:23:00    disk5   35.11    0.50      29     409    0.00   23.96
           c0t4d0   21.28    0.53      91   10826    4.44   13.44
            disk7   21.28    0.53      91   10826    4.44   13.44
18:23:01    disk5    1.98    0.50      21     406    0.00    0.88
           c0t4d0    9.90    1.51     206   17568    0.14    1.09
            disk7    9.90    1.51     206   17568    0.14    1.09
18:23:02    disk5    5.10    0.50      20     343    0.00    2.31
           c0t4d0    1.02    0.50      43    2743    0.00    0.53
            disk7    1.02    0.50      43    2743    0.00    0.53
           c0t6d0    4.08    0.50      86   10400    0.00    1.24
           disk11    4.08    0.50      86   10400    0.00    1.24
18:23:03    disk5    2.00    1.02      27    1356    0.00    0.75
           c0t5d0    4.00    0.53      87   10240    0.02    1.11
            disk9    4.00    0.53      87   10240    0.02    1.11
18:23:04    disk5   50.50    0.71      97    9537    7.96   13.09

3rd Disk:

# sar -d 1 100 /dev/dsk/c0t5d0

HP-UX mahaeccs B.11.31 U ia64    07/09/15

18:22:32   device   %busy   avque   r+w/s  blks/s  avwait  avserv
18:22:33    disk5    2.91    0.50      44    1544    0.00    0.62
           c0t6d0    3.88    0.75      81    9942    0.12    1.06
           disk11    3.88    0.75      81    9942    0.12    1.06
18:22:34   c0t6d0    4.90    0.68      77    9914    0.06    1.53
           disk11    4.90    0.68      77    9914    0.06    1.53
18:22:35    disk5    5.10    6.12      80    5380    1.13    3.05
18:22:36    disk5   50.00    0.50      14     269    0.00   55.77
           c0t4d0   31.63    0.63      56    6531   10.24   42.36
            disk7   31.63    0.63      56    6531   10.24   42.36
           c0t6d0    1.02    0.50      10    1306    0.00    3.95
           disk11    1.02    0.50      10    1306    0.00    3.95
18:22:37    disk5   28.71    0.83      66    6519    0.29    9.43
           c0t5d0   25.74    9.40      50    2788   17.18   39.81
            disk9   25.74    9.40      50    2788   17.18   39.81
18:22:38    disk5    9.00    0.50      59    5428    0.00    2.17

Please help..

[Patrick Wallek]

How many files are in the file system?

If there are thousands and thousands of files, then those commands will take a very long time to run.

[Bill Hassell]

The sar command ignores whatever you put after the repetition count (ie, the name of a disk device).

So each of those runs are for all active disks.

The du command analyzes the given directory (or PWD if none given) and will take a while to descend through all the directories and files. Since you did not give the actual command for du (like: du /, or du -x /var, etc), there is no way that we can tell what you are analyzing. To summarize space, the correct command is:

du -kx /var | sort -rn | head

This will show the total size of /var and all subdirectories, sorted by actual space. Nothing will be shown until the du comand finishes which could be quite a while for a very large /var mountpoint.

The commands ll and ls -l are the same thing. I'm not sure why you would need the -s option for ls.

The sar commands you posted show several disks with extremely long service times. The normal service times for a typical disk array over a fibre channel is less than 10 ms and for internal disks without a Smart Array controller, less than 20 ms. There are several disks that show long delays (30-55 ms) but these are only 1 second snapshots. Change your sar command to:

sar -d 5 10

Then post only the Average lines at the end of the listing. Also look at dmesg and /var/adm/syslog/syslog.log for disk error messages.