Re: zfs question root growth

[Ed Gould]

On Feb 19, 2016, at 9:17 AM, Lizette Koehler wrote:

---SNIP
Contact the system programmer.


SNIP--

Don't you loves these? About 40 years ago I got the same response and  
that is when I started to talk to myself and have been having a  
lengthy conversation since then.

Not sure on how it will end up though.

Maybe I will write a RCF.

Ed

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Neubert, Kevin]

Regarding... I believe I have aggregate grow turned on, How can I verify?

You can verify the current dynamic grow option via "zfsadm configquery 
-aggrgrow."

Regards,

Kevin

-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of Tim Brown
Sent: Friday, February 19, 2016 7:09 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: zfs question root growth

What started this was my testing of CSSMTP using a 5 pack Z/OS replicated 
system.

I got this message:

EZD1860I CSSMTP STORAGE SHORTAGE DETECTED IN THE Z/OS UNIX FILE SYSTEM 
DEAD LETTER DIRECTORY  

I will proceed to point this directory outside of root for now.

I believe I have aggregate grow turned on, How can I verify?

There's 2600 free cyls on volume where OMVS.ROOT resides

Thanks for all your Z-replies

Tim
Z!

-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of Lizette Koehler
Sent: Friday, 19 February, 2016 9:50 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: zfs question root growth


** THIS IS AN EXTERNAL EMAIL ** Use caution before opening links / attachments. 
Never supply UserID/PASSWORD information.


Are you looking for a monitoring tool?

There are IOE messages that come out in syslog indicating utilization on zFS 
files.  So you could look at those from an external perspective.

Manually, you can use ISPF 3.17 to look at specific files and see what they 
look like.

I have not done this, but I think RMF may have something for zFS files.

Could you provide more details on what you are looking for?

Do you have aggregate grow turned on ?  What version of zFS are you using?
Version 3/4/5?
How much space is available on the volume for your root?

In my mind, a ROOT should be fairly static.  SYMLINKs and non-growing files.

What files do you have in your ROOT that maybe should be under /etc /lpp /user 
and so forth?


Lizette


> -Original Message-
> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] 
> On Behalf Of Tim Brown
> Sent: Friday, February 19, 2016 7:42 AM
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: zfs question root growth
> 
> Is there a way to determine where space is allocated in ZFS to find 
> which directories use the most space. Our root is low on free space 
> and I am concerned about its ability to grow.
> 
> Thanks,
> Tim Brown

--
For IBM-MAIN subscribe / signoff / archive access instructions, send email to 
lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions, send email to 
lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Paul Gilmartin]

On 2016-02-19, at 07:48, Bigendian Smalls wrote:

> Tim have you tried this from a shell at the root of your ZFS partition (or 
> just root / )
> 
> du -sk | sort
>  
BTW, I'm more comfortable with:

du -k | sort -nk1,1

-- gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Elardus Engelbrecht]

Lizette Koehler wrote:

>If there is a large number of mail messages in long retry, then there 
> might be a problem with the mail servers.

Indeed! Some mail servers / relays may blackhat your CSSMTP as 'spam' or 
something like that.

To Tim, what are your CSSMTP Timeout and RetryLimit settings?

>Delete all unneeded dead letters to free some space in the z/OS UNIX file 
> system dead letter directory.

It should work.

>See the information about the UNDELIVERABLE statement in z/OS Communications 
>Server: IP Configuration Reference for more information about z/OS UNIX file 
>system dead letter directory.

>So you placed the dead letter queue in the ROOT instead of in sub directory?

To Tim, show us your DeadLetterAction and DeadLetterDirectory settings.

We have several CSMTP STCs and each of these STCs own DeadLetterDirectory is 
/tmp/ or something like that. These folders are mounted somewhere else to avoid 
usage of the ROOT. Each LPAR own /tmp/ folders are sitting in their own zFS 
dataset.

Groete / Greetings
Elardus Engelbrecht

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Warren, Cliff]

To extend an existing ZFS data set (including root)





>From Ishell go into filesystems and then going to number 5 for ZFS aggregates



   File  Directory  Special_file  Tools  File_systems  Options  Setup  Help

- +-+ 

  UNIX System Serv |1. Mount table...|

   |2. New HFS...|

Enter a pathname and do one of these: |3. Mount(O)...   |

   |4. New zFS...|

 - Press Enter.|5. zFS aggregates... |

 - Select an action bar choice.+-+

 - Specify an action code or command on the command line.







Type "E" for extend on the directory you want to increase the size of



 Attached zFS Aggregates Row 1 to 15 of 15



Select an aggregate with a line command.

A=Attributes  L=List file systems  E=Extend



S   Aggregate NameFree Space  Total Space

E   OMVS.ROOT 300212  2538000

_   OMVS.SCAZROOT 31 2400

_   OMVS.SCFZHFS2.NEW  72935   108720

_   OMVS.SERVICE  820743  108





Overtype the size with the total requested amount for the extended root





  Attached zFS Aggregates Row 1 to 15 of 15

   +---+

S |  Enter the New Aggregate Size in Kilobytes|

A |   |

   | Aggregate  . . : OMVS.ROOT   |

S | New size . . . . 280|

_ |   |

_ |   |

_ |   |







Will

-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of Tim Brown
Sent: Friday, February 19, 2016 10:09 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: zfs question root growth



What started this was my testing of CSSMTP using a 5 pack Z/OS replicated 
system.



I got this message:



EZD1860I CSSMTP STORAGE SHORTAGE DETECTED IN THE Z/OS UNIX FILE SYSTEM

DEAD LETTER DIRECTORY



I will proceed to point this directory outside of root for now.



I believe I have aggregate grow turned on, How can I verify?



There's 2600 free cyls on volume where OMVS.ROOT resides



Thanks for all your Z-replies



Tim

Z!



-Original Message-

From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of Lizette Koehler

Sent: Friday, 19 February, 2016 9:50 AM

To: IBM-MAIN@LISTSERV.UA.EDU<mailto:IBM-MAIN@LISTSERV.UA.EDU>

Subject: Re: zfs question root growth





** THIS IS AN EXTERNAL EMAIL ** Use caution before opening links / attachments. 
Never supply UserID/PASSWORD information.





Are you looking for a monitoring tool?



There are IOE messages that come out in syslog indicating utilization on zFS 
files.  So you could look at those from an external perspective.



Manually, you can use ISPF 3.17 to look at specific files and see what they 
look like.



I have not done this, but I think RMF may have something for zFS files.



Could you provide more details on what you are looking for?



Do you have aggregate grow turned on ?  What version of zFS are you using?

Version 3/4/5?

How much space is available on the volume for your root?



In my mind, a ROOT should be fairly static.  SYMLINKs and non-growing files.



What files do you have in your ROOT that maybe should be under /etc /lpp /user 
and so forth?





Lizette





> -Original Message-

> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU]

> On Behalf Of Tim Brown

> Sent: Friday, February 19, 2016 7:42 AM

> To: IBM-MAIN@LISTSERV.UA.EDU<mailto:IBM-MAIN@LISTSERV.UA.EDU>

> Subject: zfs question root growth

>

> Is there a way to determine where space is allocated in ZFS to find

> which directories use the most space. Our root is low on free space

> and I am concerned about its ability to grow.

>

> Thanks,

> Tim Brown



--

For IBM-MAIN subscribe / signoff / archive access instructions, send email to 
lists...@listserv.ua.edu<mailto:lists...@listserv.ua.edu> with the message: 
INFO IBM-MAIN



--

For IBM-MAIN subscribe / signoff / archive access instructions, send email to 
lists...@listse

Re: zfs question root growth

[Paul Gilmartin]

On 2016-02-19, at 08:14, Bigendian Smalls wrote:

>> But beware; it may be resource-intensive.
> 
> Alternatively you could du -sk * > file.txt
> then cat file.txt|sort
> 
> Saving the trouble of doing it in memory.  Though I suspect calculating the 
> sizes of the folders is much much more intensive than sorting a relatively 
> brief set of text.
> 
>> Does UNIX sort employ DFSORT when needed/available?
> 
> I don’t know for sure, but I’d be surprised if it did.   /bin/sort is a 
> standard fixture in ‘nixes forever …   Could poke at it a bit though to find 
> out for sure :)
>  
Just thinking that DFSORT is probably more technologically advanced than
/bin/sort.  If the output of du is saved in a file, that file could probably
be used as input do DFSORT.  Less confident about piping directly into DFSORT.

-- gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Tim Brown]

I took defaults so  if that used root for dead letter directory then I need to 
change
I am in infancy and the only one testing.

Still need to tweak


Tim

-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of Lizette Koehler
Sent: Friday, 19 February, 2016 10:18 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: zfs question root growth


** THIS IS AN EXTERNAL EMAIL ** Use caution before opening links / attachments. 
Never supply UserID/PASSWORD information.


So here is the message text.  Did you do the displays suggested in the message?


EZD1860I jobname STORAGE SHORTAGE DETECTED IN THE z/OS UNIX FILE SYSTEM DEAD 
LETTER DIRECTORY Explanation The Communications Server SMTP (CSSMTP) 
application detected that storage use in the z/OSR UNIX file system dead letter 
directory exceeds 75%.
In the message text:

Jobname The job name of the task that started the CSSMTP application. 

System action

Processing continues.
Operator response

Contact the system programmer.
System programmer response
A possible cause for this error is that the UNIX file system that contains the 
dead letter directory is full or nearly full. If this is the cause of the 
error, perform one of the following:

Determine what can be deleted in the file system and delete it.
Change the DeadLetterDirectory configuration statement to be a directory on 
a different file system and issue a MODIFY REFRESH command to use the new 
directory.

If the problem persists, then either issue a MODIFY SUSPEND command or change 
the DeadLetterAction configuration statement to DELETE.

Determine whether you need to issue the MODIFY SUSPEND command.

Issue the MODIFY D,SPOOLSTATUS,SUMMARY command to determine the number of 
mail messages that are in the pending or long retry state for each JES task.
>From the summary report, obtain the TKID value that has pending or long retry 
>mail messages, then issue the MODIFY D,SPOOLSTATUS,DETAIL,TKID=tkid command to 
>display detail information for the specific JES task.
If there is a large number of mail messages in long retry, then there might 
be a problem with the mail servers. Issue the MODIFY SUSPEND command to suspend 
new JES spool file processing until the problem is resolved. If messages 
EZD1817I, EZD1818I or EZD1819I are in the log, they might help in determining 
the problem. See EZD1817I, EZD1818I, or EZD1819I for more information.
Delete all unneeded dead letters to free some space in the z/OS UNIX file 
system dead letter directory.

See the information about the UNDELIVERABLE statement in z/OS Communications
Server: IP Configuration Reference for more information about z/OS UNIX file 
system dead letter directory.

So you placed the dead letter queue in the ROOT instead of in sub directory?

Lizette


> -Original Message-
> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] 
> On Behalf Of Tim Brown
> Sent: Friday, February 19, 2016 8:09 AM
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: Re: zfs question root growth
> 
> What started this was my testing of CSSMTP using a 5 pack Z/OS 
> replicated system.
> 
> I got this message:
> 
> EZD1860I CSSMTP STORAGE SHORTAGE DETECTED IN THE Z/OS UNIX FILE SYSTEM 
> DEAD LETTER DIRECTORY
> 
> I will proceed to point this directory outside of root for now.
> 
> I believe I have aggregate grow turned on, How can I verify?
> 
> There's 2600 free cyls on volume where OMVS.ROOT resides
> 
> Thanks for all your Z-replies
> 
> Tim
> Z!
> 
> -Original Message-
> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] 
> On Behalf Of Lizette Koehler
> Sent: Friday, 19 February, 2016 9:50 AM
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: Re: zfs question root growth
> 
> 
> ** THIS IS AN EXTERNAL EMAIL ** Use caution before opening links / 
> attachments. Never supply UserID/PASSWORD information.
> 
> 
> Are you looking for a monitoring tool?
> 
> There are IOE messages that come out in syslog indicating utilization 
> on zFS files.  So you could look at those from an external perspective.
> 
> Manually, you can use ISPF 3.17 to look at specific files and see what 
> they look like.
> 
> I have not done this, but I think RMF may have something for zFS files.
> 
> Could you provide more details on what you are looking for?
> 
> Do you have aggregate grow turned on ?  What version of zFS are you using?
> Version 3/4/5?
> How much space is available on the volume for your root?
> 
> In my mind, a ROOT should be fairly static.  SYMLINKs and non-growing files.
> 
> What files do you have in your ROOT that maybe should be under /etc 
> /lpp /user and so forth?
> 
> 
> Lizette
> 
> 
> > -Original Message-
> > From: IBM Mainframe Dis

Re: zfs question root growth

[Lizette Koehler]

So here is the message text.  Did you do the displays suggested in the message?


EZD1860I jobname STORAGE SHORTAGE DETECTED IN THE z/OS UNIX FILE SYSTEM DEAD
LETTER DIRECTORY
Explanation The Communications Server SMTP (CSSMTP) application detected that
storage use in the z/OSR UNIX file system dead letter directory exceeds 75%.
In the message text:

Jobname The job name of the task that started the CSSMTP application. 

System action

Processing continues.
Operator response

Contact the system programmer.
System programmer response
A possible cause for this error is that the UNIX file system that contains the
dead letter directory is full or nearly full. If this is the cause of the error,
perform one of the following:

Determine what can be deleted in the file system and delete it.
Change the DeadLetterDirectory configuration statement to be a directory on
a different file system and issue a MODIFY REFRESH command to use the new
directory.

If the problem persists, then either issue a MODIFY SUSPEND command or change
the DeadLetterAction configuration statement to DELETE.

Determine whether you need to issue the MODIFY SUSPEND command.

Issue the MODIFY D,SPOOLSTATUS,SUMMARY command to determine the number of
mail messages that are in the pending or long retry state for each JES task.
>From the summary report, obtain the TKID value that has pending or long retry
mail messages, then issue the MODIFY D,SPOOLSTATUS,DETAIL,TKID=tkid command to
display detail information for the specific JES task.
If there is a large number of mail messages in long retry, then there might
be a problem with the mail servers. Issue the MODIFY SUSPEND command to suspend
new JES spool file processing until the problem is resolved. If messages
EZD1817I, EZD1818I or EZD1819I are in the log, they might help in determining
the problem. See EZD1817I, EZD1818I, or EZD1819I for more information.
Delete all unneeded dead letters to free some space in the z/OS UNIX file
system dead letter directory.

See the information about the UNDELIVERABLE statement in z/OS Communications
Server: IP Configuration Reference for more information about z/OS UNIX file
system dead letter directory.

So you placed the dead letter queue in the ROOT instead of in sub directory?

Lizette


> -Original Message-
> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On
> Behalf Of Tim Brown
> Sent: Friday, February 19, 2016 8:09 AM
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: Re: zfs question root growth
> 
> What started this was my testing of CSSMTP using a 5 pack Z/OS replicated
> system.
> 
> I got this message:
> 
> EZD1860I CSSMTP STORAGE SHORTAGE DETECTED IN THE Z/OS UNIX FILE SYSTEM
> DEAD LETTER DIRECTORY
> 
> I will proceed to point this directory outside of root for now.
> 
> I believe I have aggregate grow turned on, How can I verify?
> 
> There's 2600 free cyls on volume where OMVS.ROOT resides
> 
> Thanks for all your Z-replies
> 
> Tim
> Z!
> 
> -Original Message-
> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On
> Behalf Of Lizette Koehler
> Sent: Friday, 19 February, 2016 9:50 AM
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: Re: zfs question root growth
> 
> 
> ** THIS IS AN EXTERNAL EMAIL ** Use caution before opening links /
> attachments. Never supply UserID/PASSWORD information.
> 
> 
> Are you looking for a monitoring tool?
> 
> There are IOE messages that come out in syslog indicating utilization on zFS
> files.  So you could look at those from an external perspective.
> 
> Manually, you can use ISPF 3.17 to look at specific files and see what they
> look like.
> 
> I have not done this, but I think RMF may have something for zFS files.
> 
> Could you provide more details on what you are looking for?
> 
> Do you have aggregate grow turned on ?  What version of zFS are you using?
> Version 3/4/5?
> How much space is available on the volume for your root?
> 
> In my mind, a ROOT should be fairly static.  SYMLINKs and non-growing files.
> 
> What files do you have in your ROOT that maybe should be under /etc /lpp /user
> and so forth?
> 
> 
> Lizette
> 
> 
> > -Original Message-
> > From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU]
> > On Behalf Of Tim Brown
> > Sent: Friday, February 19, 2016 7:42 AM
> > To: IBM-MAIN@LISTSERV.UA.EDU
> > Subject: zfs question root growth
> >
> > Is there a way to determine where space is allocated in ZFS to find
> > which directories use the most space. Our root is low on free space
> > and I am concerned about its ability to grow.
> >
> > Thanks,
> > Tim Brown
> 

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Bigendian Smalls]

> But beware; it may be resource-intensive.

Alternatively you could du -sk * > file.txt
then cat file.txt|sort

Saving the trouble of doing it in memory.  Though I suspect calculating the 
sizes of the folders is much much more intensive than sorting a relatively 
brief set of text.

> Does UNIX sort employ DFSORT when needed/available?

I don’t know for sure, but I’d be surprised if it did.   /bin/sort is a 
standard fixture in ‘nixes forever …   Could poke at it a bit though to find 
out for sure :)

Chad

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Tim Brown]

What started this was my testing of CSSMTP using a 5 pack Z/OS replicated 
system.

I got this message:

EZD1860I CSSMTP STORAGE SHORTAGE DETECTED IN THE Z/OS UNIX FILE SYSTEM 
DEAD LETTER DIRECTORY  

I will proceed to point this directory outside of root for now.

I believe I have aggregate grow turned on, How can I verify?

There's 2600 free cyls on volume where OMVS.ROOT resides

Thanks for all your Z-replies

Tim
Z!

-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of Lizette Koehler
Sent: Friday, 19 February, 2016 9:50 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: zfs question root growth


** THIS IS AN EXTERNAL EMAIL ** Use caution before opening links / attachments. 
Never supply UserID/PASSWORD information.


Are you looking for a monitoring tool?

There are IOE messages that come out in syslog indicating utilization on zFS 
files.  So you could look at those from an external perspective.

Manually, you can use ISPF 3.17 to look at specific files and see what they 
look like.

I have not done this, but I think RMF may have something for zFS files.

Could you provide more details on what you are looking for?

Do you have aggregate grow turned on ?  What version of zFS are you using?
Version 3/4/5?
How much space is available on the volume for your root?

In my mind, a ROOT should be fairly static.  SYMLINKs and non-growing files.

What files do you have in your ROOT that maybe should be under /etc /lpp /user 
and so forth?


Lizette


> -Original Message-
> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] 
> On Behalf Of Tim Brown
> Sent: Friday, February 19, 2016 7:42 AM
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: zfs question root growth
> 
> Is there a way to determine where space is allocated in ZFS to find 
> which directories use the most space. Our root is low on free space 
> and I am concerned about its ability to grow.
> 
> Thanks,
> Tim Brown

--
For IBM-MAIN subscribe / signoff / archive access instructions, send email to 
lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Paul Gilmartin]

On 2016-02-19, at 07:48, Bigendian Smalls wrote:

> Tim have you tried this from a shell at the root of your ZFS partition (or 
> just root / )
> 
> du -sk | sort
>  
But beware; it may be resource-intensive.

Does UNIX sort employ DFSORT when needed/available?

-- gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Elardus Engelbrecht]

Tim Brown wrote:

>Is there a way to determine where space is allocated in ZFS to find which 
>directories use the most space. Our root is low on free space and I am 
>concerned about its ability to grow.

You already got excellent replies.

On what z/OS version are you?

About growth, there are good zFS tools like 'zFS tools package' available on 
www.ibm.com.

Check also this URL about z/OS Distributed File Service (assuming you're on 
z/OS v2.1):

https://www-01.ibm.com/support/knowledgecenter/SSLTBW_2.1.0/com.ibm.zos.v2r1.ioe/ioe.htm

Check this redbook 'ABCs of z/OS System Programming Volume 9' at:

http://www.redbooks.ibm.com/redbooks/pdfs/sg246989.pdf

Groete / Greetings
Elardus Engelbrecht

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Staller, Allan]

In most cases, I would expect the root directory to be mounted read only.

I would not expect the root to grow much (if any), except at a new release 
level of z/OS (which would mean a new zFS cluster). 
As long as there is sufficient space to add some new directories, you should be 
OK.

For example, I recently installed a new product at /usr/lpp/product. This was 
added as another mount point which was pointed to the product's zFS cluster.
The only change to the root was the addition of the new directory to /usr/lpp.

That being said, the zfsadm -aggrow (? Not sure of the exact syntax) will all 
dynamic growth of the zFS cluster.

HTH,



Is there a way to determine where space is allocated in ZFS to find which 
directories use the most space. Our root is low on free space and I am 
concerned about its ability to grow.


This email – including attachments – may contain confidential information. If 
you are not the intended recipient, do not copy, distribute or act on it. 
Instead, notify the sender immediately and delete the message.

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Bigendian Smalls]

Couple other commands handy to manage ZFS / other mountpoint sizes

zfsadm aggrinfo

ex:
...
ZFS.DIR.U2 (R/W COMP): 1314251 K free out of total 2728080

df -kP

ex:
Filesystem 1024-blocksUsed  Available  Capacity Mounted on
...
ZFS.DIR.U2 2728080 14138291314251   52% /u


Forgot the asterisk on my first post

du -sk *  |sort

Chad



The former will show you size / utilization of all your zfs mounts.   The 
latter will show freespace on all of your mounts, irrespective of type.




On Feb 19, 2016, at 8:50 AM, Lizette Koehler 
mailto:stars...@mindspring.com>> wrote:

Are you looking for a monitoring tool?

There are IOE messages that come out in syslog indicating utilization on zFS
files.  So you could look at those from an external perspective.

Manually, you can use ISPF 3.17 to look at specific files and see what they look
like.

I have not done this, but I think RMF may have something for zFS files.

Could you provide more details on what you are looking for?

Do you have aggregate grow turned on ?  What version of zFS are you using?
Version 3/4/5?
How much space is available on the volume for your root?

In my mind, a ROOT should be fairly static.  SYMLINKs and non-growing files.

What files do you have in your ROOT that maybe should be under /etc /lpp /user
and so forth?


Lizette


-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On
Behalf Of Tim Brown
Sent: Friday, February 19, 2016 7:42 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: zfs question root growth

Is there a way to determine where space is allocated in ZFS to find which
directories use the most space. Our root is low on free space and I am
concerned about its ability to grow.

Thanks,
Tim Brown

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with 
the message: INFO IBM-MAIN


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Lizette Koehler]

Are you looking for a monitoring tool?

There are IOE messages that come out in syslog indicating utilization on zFS
files.  So you could look at those from an external perspective.

Manually, you can use ISPF 3.17 to look at specific files and see what they look
like.

I have not done this, but I think RMF may have something for zFS files.

Could you provide more details on what you are looking for?

Do you have aggregate grow turned on ?  What version of zFS are you using?
Version 3/4/5?
How much space is available on the volume for your root?

In my mind, a ROOT should be fairly static.  SYMLINKs and non-growing files.

What files do you have in your ROOT that maybe should be under /etc /lpp /user
and so forth?


Lizette


> -Original Message-
> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On
> Behalf Of Tim Brown
> Sent: Friday, February 19, 2016 7:42 AM
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: zfs question root growth
> 
> Is there a way to determine where space is allocated in ZFS to find which
> directories use the most space. Our root is low on free space and I am
> concerned about its ability to grow.
> 
> Thanks,
> Tim Brown

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Leopold Strauss]

Under zOS/USS:

Go to the path, where you suspect some big dirs and enter

du -ks   *


On 19.02.2016 15:41, Tim Brown wrote:

Is there a way to determine where space is allocated in ZFS to find which 
directories
use the most space. Our root is low on free space and I am concerned about its
ability to grow.

Thanks,
Tim Brown

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN





--
Mit freundlichen Gruessen / Kind Regards,
Leopold Strauss, Team DEV-zOS, T: +43-2236-27551-331

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: zfs question root growth

[Bigendian Smalls]

Tim have you tried this from a shell at the root of your ZFS partition (or just 
root / )


du -sk | sort


*that’s a pipe sign, not an I or an l  :) 
drop the | sort if you just want to see the folder sizes in alphabetical order

That should give you all the directories sizes in kilobytes sorted lowest to 
highest.  You can then hit the largest ones, dive into that directory and 
repeat.

You could script this if you wanted to do it regularly.


Chad


> On Feb 19, 2016, at 8:41 AM, Tim Brown  wrote:
> 
> Is there a way to determine where space is allocated in ZFS to find which 
> directories
> use the most space. Our root is low on free space and I am concerned about its
> ability to grow.
> 
> Thanks,
> Tim Brown
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN