Discussion:
zFS Growth and effect on applications in z/OS
Add Reply
Lizette Koehler
2017-09-04 02:32:13 UTC
Reply
Permalink
Raw Message
Need some guidance here

I have created a V5 zFS file that has EA/EF. The primary allocation is 2000
cylinders and the secondary is 5000 cylinders.

The application expects this file to grow to over 1TB in size.

An application using the zFS gets messages indicating the file is full. But
from a LISTC it is not - it is 5GB in size and using 3GB


09:51:37 03SEP PR01 IOEZ00312I Dynamic growth of aggregate OMVS.MY.ZFS in
progress, (by user X

09:51:38 03SEP PR01 IOEZ00309I Aggregate OMVS.MY.ZFS. ZFS successfully
dynamically grown (by user X
09:51:38 03SEP PR01 IOEZ00078E zFS aggregate OMVS.MY.ZFS exceeds 95% full
(388000/391200) (WARNIN

3,178,496,000/3,204,710,400

I am not seeing an 2C3 abends (so far), I do not see any BPX messages (so far)


The Hi-U-RBA from LISTC matches the numbers in the IOEZ00078E message.

I checked throughout the day, the file seems to get a lot of the following
messages

IOEZ00312I
IOEZ00309I
IOEZ00078E

So I am not sure why the zfs full is occurring.

1) what is the impact of dynamic growth on zFS users?
a) Does it cause an application to fail during growth
b) are there other "cache" things in zFS I should be looking at?
2) What other cache could impact a zFS application
a) the V5 zFS file I created is at 5GB in size - I coded it with EXTENDED
EXT-ADDR ZFS
3) What diagnostics can I use to determine why this application is having
issues?

Any thoughts or suggestions appreciated. I am just not seeing a solution

Thanks


Lizette Koehler
statistics: A precise and logical method for stating a half-truth inaccurately

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Ed Jaffe
2017-09-04 15:32:40 UTC
Reply
Permalink
Raw Message
Post by Lizette Koehler
1) what is the impact of dynamic growth on zFS users?
a) Does it cause an application to fail during growth
We use zFS for our corporate SMB server and git repositories.

https://www.ibm.com/support/knowledgecenter/SSLTBW_2.2.0/com.ibm.zos.v2r2.ioea700/ioea7d0041006636.htm
states, "The aggregate is extended when an operation cannot complete
because the aggregate is full. If the extension is successful, the
operation is again transparently driven to the application." Not the
best English in the world there <smh> but, they're obviously attempting
to say the operation is automatically retried without application
involvement and that has been my observation.

However, applications that ask for file system stats are given an answer
without being suspended. (Can't imagine how/why zFS should do anything
different there.)

Therefore, a "clever" application *could* report an "out of space"
condition if it thinks there is not enough free space remaining. I have
seen this from a few older Windows applications (VSS I think?), but
never from things like ftp or sftp.
--
Phoenix Software International
Edward E. Jaffe
831 Parkview Drive North
El Segundo, CA 90245
http://www.phoenixsoftware.com/

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