May i have explanations about extends in cache disks CMOD?

[plb3945]

Hi,

System team tried to put several extends in CMOD cache disks but extends are not automatically allocated just one. Is it a normal function in CMOD?
To have several extends system team allocated manually by a lot of copies and delete.
Could you reply me about this problem.
Thanks for cooperation
Best regards
Philippe
 

[Alessandro Perucchi]

Hello Philippe,

I don't know very well the z/OS world for CMOD, but we have in CMOD Unix a file called  ars.cache, where you need to specify all the path of extension of cache disks.
Maybe you need to do the same in z/OS.

I have found that in case you want to know how it works:
http://publib.boulder.ibm.com/infocenter/cmod/v8r4m1/index.jsp?topic=/com.ibm.ondemand.configuringzos.doc/ars3u37195.htm

BTW, you need to stop and restart CMOD in order to "refresh" the setup.

Best regards,
Alessandro

[plb3945]

Hello Alessandro,

Thanks for your reply but it's already done we have correctly fill the ars.cache with all cache file name. Our problem is: we want an automatic extends allocation for each HFS cache file. Generally one HFS use 255 extends but for the CMOD cache HFS it's not the case, we have just one extend. HFS doesn't take extend automaticaly. To have several extends for each HFS, system team is oblige to copy a lot of file to allocate manually and delete after, it's not automatic with HFS CMOD cache file.
Thanks for your help.
Best regards
Philippe

[Alessandro Perucchi]

Hello Philippe,

You've went over my very limited knowledge on z/OS now !! :-D
You talk chinese to me with these HFS, extends, etc... :-)

I hope someone with z/OS expertise can help you, or at least give you some hints on that.

Cheers,
Alessandro

[Ed_Arnold]

Bonjour Philippe et Alessandro!

Before OnDemand stores an object in cache, it finds out how much space is available. 

If not enough is free, OnDemand issues an ARS0025E.

The base OnDemand code is supported across many platforms including UNIX and Windows.  With those operating systems the space wouldn't be automatically increased by the file system.                                         
                                                                   
There are at least two possible solutions.                                                                 
                                                                       
1. You could add additional cache directories to your ars.cache file.                                                                             
                                                                     
2. You could manually extend your HFS by 10 cylinders by issuing

    /usr/lpp/dfsms/bin/confighfs -x 10c

3. If you're using zFS filesystem, then you need to use the aggrgrow command.
             
Ed