File system getting full and Server node getting down.

[former_member787462]

Hi Team,

Currently we are using IBM Power 6 AIX operating system.

And in our environment, for development system, file system is getting full and development system is getting slow while accessing.

Can you please let me know , what exactly the problem & which command is used to see the file system size and how to resolve the issue by deleting the core files or some ting. Please help me .

Thanks

Manoj K

[former_member787462]

Hi Orkun Gedik,

When i executed the command df -lg and find . -name core noting is displayed, but if i execute the command df is showed me the below information. below is an original file which i have modified the sid.

Filesystem 512-blocks Free %Used Iused %Iused Mounted on

/dev/fslv10 52428800 16279744 69% 389631 15% /usr/sap/SID

Server 0 node is giving the problem. its getting down all the times.

And if i check it in the /usr/sap/SID/<Instance>/work for the server node "std_server0.out file , the below information is written in the file.

framework started for 73278 ms.

SAP J2EE Engine Version 7.00 PatchLevel 81863.450 is running! PatchLevel 81863.450 March 10, 2010 11:48 GMT

94.539: [GC 94.539: [ParNew: 239760K->74856K(261888K), 0.2705150 secs] 239760K->74856K(2009856K), 0.2708720 secs] [Times: user=0.00 sys=0.36, real=0.27 secs]

105.163: [GC 105.164: [ParNew: 249448K->80797K(261888K), 0.2317650 secs] 249448K->80797K(2009856K), 0.2320960 secs] [Times: user=0.00 sys=0.44, real=0.23 secs]

113.248: [GC 113.248: [ParNew: 255389K->87296K(261888K), 0.3284190 secs] 255389K->91531K(2009856K), 0.3287400 secs] [Times: user=0.00 sys=0.58, real=0.33 secs]

Please advise.

thanks in advance

Manoj K

[Former Member]

Hi,

Do you mean /usr/sap/SID file system gets full and due to this your java instance does not startup ?

Please share logs for system not starting.

Regards,

Deepak Kori

[Former Member]

>> When i executed the command df -lg

it is capital "i" not "l"

>> Server 0 node is giving the problem. its getting down all the times.

Check your log file archive folders under for example "/usr/sap/<SID>\<instance>/j2ee/cluster/server<x>/log/archive"

Best regards,

Orkun Gedik

[former_member787462]

Hi,

after executing the command "df -Ig" i got the below information.

Filesystem GB blocks Used Free %Used Mounted on

/dev/fslv10 25.00 17.24 7.76 69% /usr/sap/SID

And i have checked in the "/usr/sap/SID/Instance<x>/j2ee/cluster/server<x>/log/archive" its empty. noting is there.

@deepa,

what exactly you wanted from logs. I am just suspecting that could be the reason, at this point of time i am not sure .

Please help.

Thanks

Manoj K

[Former Member]

Hi,

>

> And i have checked in the "/usr/sap/SID/Instance<x>/j2ee/cluster/server<x>/log/archive" its empty. noting is there.

>

These files are hidden. So, you have to use -a option to see these archive files.

Also, make sure that set trace level to Error in visual admin.

Thanks

Sunny

[Former Member]

Hi Manoj,

>> Can you please let me know , what exactly the problem & which command is used to see the file system size and how to resolve the issue by deleting the core files or some ting. Please help me .

You can execute "df -Ig" command to see capacity status of the logical volumes. Additionally, you can find the core files with the command at the root level

"find . -name core"

Then, you can drop these files, if they are not necesarry.

Moreover, you can configure standard reorganization jobs by the OSS note 16083 and set a retention period for the old log files, if they have been configured as store at the OS level.

One more thing is, you can check "/usr/sap/<SID>/<Instance>/work" folder. This folder may have been used for work directory of some of the applications.

Best regards,

Orkun Gedik