File system overflow -

[susee]

Hi all,

Can any one please suggest me why this error occurs in syslog , every time it keeps on counting
-----------------------------
Feb 15 06:40:58 ddsapd01 vmunix: file: table is full
Feb 15 06:40:58 ddsapd01 vmunix: file: table is full
Feb 15 06:47:20 ddsapd01 above message repeats 263 times
Feb 15 08:00:45 ddsapd01 vmunix: file: table is full
Feb 15 08:01:03 ddsapd01 vmunix: file: table is full
Feb 15 08:07:22 ddsapd01 above message repeats 2769 times
--------------------------
# bdf
Filesystem kbytes used avail %used Mounted on
/dev/vg00/lvol3 204800 152385 49230 76% /
/dev/vg00/lvol1 99669 54088 35614 60% /stand
/dev/vg00/lvol8 1048576 810259 224166 78% /var
/dev/vg00/lvol7 1097728 940494 147570 86% /usr
/dev/vg46/lvol4 4096000 2681837 1325808 67% /usr/sap/put
/dev/vg39/lvol1 4096000 797456 3092392 21% /usr/sap/DAT
/dev/vg31/lvol1 4194304 92769 3845206 2% /usr/sap/DAS
/dev/vg00/lvol10 819200 69634 702723 9% /usr/local
/dev/vg00/lvol9 2613248 278365 2190120 11% /usr/emc
/dev/vg00/lvol6 491520 292622 187076 61% /tmp
/dev/vg39/lvol2 4096000 1314430 2608749 34% /sapmnt/DAT
/dev/vg31/lvol2 4194304 1190954 2815683 30% /sapmnt/DAS
/dev/vg31/lvol4 2097152 1322463 726326 65% /saphelp
/dev/vg31/lvol5 1048576 121772 868903 12% /oracle
/dev/vg39/lvol3 4096000 1711294 2235705 43% /oracle/DAT
/dev/vg46/lvol3 8192000 1830375 5964026 23% /oracle/DAT/stage
/dev/vg46/lvol6 34816000 20924936 13782552 60% /oracle/DAT/sapreorg
/dev/vg45/lvol1 61931520 42933704 18849456 69% /oracle/DAT/sapdata6
/dev/vg44/lvol1 124928000 118940224 5941040 95% /oracle/DAT/sapdata5
/dev/vg43/lvol1 61931520 52791416 9068760 85% /oracle/DAT/sapdata4
/dev/vg42/lvol1 124928000 120312368 4579608 96% /oracle/DAT/sapdata3
/dev/vg41/lvol1 81920000 78064200 3825720 95% /oracle/DAT/sapdata2
/dev/vg40/lvol1 61685760 51552744 10053912 84% /oracle/DAT/sapdata1
/dev/vg46/lvol5 34816000 24312 34519888 0% /oracle/DAT/saparch
/dev/vg46/lvol1 1024000 165199 805133 17% /oracle/DAT/origlogB
/dev/vg39/lvol6 1024000 199768 772723 21% /oracle/DAT/origlogA
/dev/vg39/lvol7 1024000 165199 805133 17% /oracle/DAT/mirrlogB
/dev/vg46/lvol2 1024000 165199 805132 17% /oracle/DAT/mirrlogA
/dev/vg39/lvol5 4096000 2578367 1422819 64% /oracle/DAT/920_64
/dev/vg31/lvol3 4194304 1772428 2270558 44% /oracle/DAS
/dev/vg32/lvol3 2048000 1606 1918502 0% /oracle/DAS/stage
/dev/vg31/lvol7 35344384 478456 34593544 1% /oracle/DAS/sapreorg
/dev/vg38/lvol1 81920000 64557976 17226440 79% /oracle/DAS/sapdata6
/dev/vg37/lvol1 148480000 136432024 11953896 92% /oracle/DAS/sapdata5
/dev/vg36/lvol1 97280000 86005144 11186808 88% /oracle/DAS/sapdata4
/dev/vg35/lvol1 114688000 105382672 9232680 92% /oracle/DAS/sapdata3
/dev/vg34/lvol1 81920000 69546864 12276512 85% /oracle/DAS/sapdata2
/dev/vg33/lvol1 92160000 89540952 2598632 97% /oracle/DAS/sapdata1
/dev/vg32/lvol4 25600000 24328 25176060 0% /oracle/DAS/saparch
/dev/vg32/lvol1 536576 165079 348286 32% /oracle/DAS/origlogB
/dev/vg31/lvol8 536576 185647 329003 36% /oracle/DAS/origlogA
/dev/vg31/lvol9 536576 165079 348286 32% /oracle/DAS/mirrlogB
/dev/vg32/lvol2 536576 165079 348286 32% /oracle/DAS/mirrlogA
/dev/vg31/lvol6 4096000 2618452 1385275 65% /oracle/DAS/920_64
/dev/vg00/lvol5 2097152 1278311 767792 62% /opt
/dev/vg00/lvol12 204800 60943 134919 31% /opt/Tivoli
/dev/vg00/tscm_lv 102400 22174 75230 23% /opt/IBM/SCM
/dev/vg00/lvol11 12288 1109 10487 10% /omni
/dev/vg00/lvol4 102400 80936 20216 80% /home
/dev/vg00/lvol14 1536000 1192618 321936 79% /export/patches
/dev/vg39/lvol4 7516160 2046524 5127786 29% /archive
-------------------------------------

Thanks & regards
Susee

[James R. Ferguson]

Hi:

This is a kernel message that means you have exceeded the maximum number of files that are allowd to be open at one time, system-wide.

It has nothing to do with space within filesystems.

You need to change the value of your kernel's 'nfile' parameter:

http://docs.hp.com/en/TKP-90202/index.html

You can use 'sar -v' to monitor the size of the 'nfile' table. See the "file-sz" column.

Regards!

...JRF...

[Rick Garland]

The kernel parameter nfile needs to be increased

No relation to file system space

Do a search on "file table full" in this forum

You will find lots of previous posts

[Jagadesh_2]

Hi,

You have exceeded the maximum number of files that can be open at one point of time. You will be getting this error in syslog & dmesg aswell. You need to increase the nfile kernel parameter. Increase the nfile value to 50% more. As this is a static parameter this will require a reboot.

Hope this helps.

[susee]

Hi all ,

Currently iam having the following value set to it
-------------------------------------------
nfile 8202 8202 Static
-------------------------------------------

Can you suggest me to what level it has to be increased and any calculations for that

Thanks
Susee

[Abhijit P.]

Hi Susee...

One of the ways to confirm that
when the error Occurs:
is to run "sar -v 2 5"

Look at the column file-sz. The first number shows the number of current
open files/the second number shows the currently configured maximum number
of open files. When the error is reported, the first value is usually close to
the second value.

This can be resolved by increasing the value of the NFILE parameter. You may also increase the MAXFILES parameter (per user file limit). SAM will allow these changes to be made easily via the "Configurable
Parameters" section under "Kernel Configuration" area.

Regards,
Abhijit

[Tvs]

hi susee..



as abijith told , you can use sar to verify the nfile setting and current usage . if required you can change the setting by kmtune,or thrw sam,or system_prep....etc..

if the server is up and and running for a long time if it possible to restarting the server also can solve the problem.

regards

tvs

[Raj D.]

Hi Susee ,

Good day!

The problem is that , you are running out of System table parameter. This is not related to any filesystem.

If you use # glance -t , you can come to know which system parametr is falling short.


The solution:
You need to increase nfile and nflocks parameters. Check and verify with [ # glance -t / # sar -v 2 5 ]


hth,
Raj

[Raj D.]

Hi Susee (again),

Remember , chaging this kernel parameters will need a reboot.

hth,
Raj.

[Steven E. Protter]

Shalom Susee,

Your filessytem didn't overflow, you have too many files open and must recompile your kenrel.

If you are new to HP-UX change the parameters noted above in sam kernel configurable paramters.

I have never seen a system with so many filesystems mounted.

You may have to do some performance tuning to get a system configured this way working.

Good Luck,

SEP

[Adisuria Wangsadinata_1]

Hi Susee,

Check the document below (docID : HPUXKBRC00008909) at this url :

http://www2.itrc.hp.com/service/cki/docDisplay.do?docLocale=en_US&docId=200000080062403

Hope this information can help you.

Cheers,
AW