You might see similar results if you've accidentally overlayed any of the
swap with a minidisk, for the same reasons already mentioned.

Paging is temporary; Create a tape with ICKDSF on it, bring down z/VM, IPL
the tape and re-format your paging areas. Check a DIRMAP or similar to be
sure that nothing overlays the paging areas, and move / get rid of it before
doing the formatting. Then wait and see if you run into the same problems.

The problem occurs when you bring up the fourth guest, so my guess would be
that there's something further out in the paging volume(s) that doesn't get
touched until the memory load reaches that level. If the virtual machines
are of equal size, I'd suspect that it doesn't matter what order you bring
them up in.

I have a sneaking memory of a PTF for a system failure when the paging load
reached a certain level. You might also take a look for that as well.

-- 
   .~.    Robert P. Nix             Mayo Foundation
   /V\    RO-OE-5-55                200 First Street SW
  /( )\   507-284-0844              Rochester, MN 55905
  ^^-^^   ----- 
        "In theory, theory and practice are the same, but
         in practice, theory and practice are different."




On 1/4/08 6:32 AM, "Pieter Harder" <[EMAIL PROTECTED]> wrote:

> Is it possible that you forgot to ICKDSF format your paging areas? You
> can get all kinds of strange errors when the paging areas are not
> initialized to zeroed 4k blocks.
>  
>  
>  
> Best regards,
> Pieter Harder
>  
> [EMAIL PROTECTED]
> tel  +31-73-6837133 / +31-6-47272537
> 
>>>> Jose Raul Baron <[EMAIL PROTECTED]> 01/04/08 1:28 >>>
> 
> Hi, list. We are experiencing some problems with unpredictable and so
> far
> inevitable system shutdowns for aparently unknown reasons.
> 
> - We are running 4 SLES9 Linux images under z/VM 5.2
> - If we keep only 3 images, this problem doesn't seem to appear. It
> has
> happened only with 4 images running at once.
> - OPERATOR 191 console registers suddenly (but it can be like this for
> HOURS) the following message:
> 
>         xx:yy:zz HCPPAW415E Six continuous paging errors have occurred
> on
> DASD 02D5 volume 520PG1.
>         (this message appears some 25-30 times per second)
> 
> - Disk 2D5 has been hardware-checked successfully, so it has no
> hardware
> errors. 
> - Latest RSU has been applied by IBM-advise. This is our level:
> 
>       q cplevel  
>       z/VM Version 5 Release 2.0, service level 0702 (64-bit)
>       Generated at 12/21/07 19:33:42 MAI
>       IPL at 01/04/08 11:46:43 MAI
> 
>       Ready;     
>       q cmslevel 
>       CMS Level 22, Service Level 702
>       Ready;     
> 
> - These were some system indicators at the time of the latest shutdown:
> 
> 
> PROCESSOR
> =========
>                  
> AVGPROC-009% 01  
> MDC READS-000000/SEC WRITES-000000/SEC HIT RATIO-000%
> PAGING-2/SEC STEAL-000%
> Q0-00001(00000)                           DORMANT-00018
> Q1-00000(00000)           E1-00000(00000)
> Q2-00000(00000) EXPAN-001 E2-00000(00000)
> Q3-00004(00000) EXPAN-001 E3-00000(00000)
> PROC 0000-009%   
> LIMITED-00000    
>                  
> PAGE AREAS
> ==========
>                 EXTENT     EXTENT  TOTAL  PAGES   HIGH    %
> VOLID  RDEV      START        END  PAGES IN USE   PAGE USED
> ------ ---- ---------- ---------- ------ ------ ------ ----
> 520PAG 033F          1       3338 600840  30602  31676   5%
> 520PG1 02D5          0       3338 601020  11660  11835   1%
>                                   ------ ------        ----
> SUMMARY                            1174K  42262          3%
> USABLE                             1174K  42262          3%
>                  
> SPOOL AREA
> ==========
>                 EXTENT     EXTENT  TOTAL  PAGES   HIGH    %
> VOLID  RDEV      START        END  PAGES IN USE   PAGE USED
> ------ ---- ---------- ---------- ------ ------ ------ ----
> 520SPL 0315          1       3338 600840 118407 160305  19%
>                                   ------ ------        ----
> SUMMARY                           600840 118407         19%
> USABLE                            600840 118407         19%
> * * * End of File * * *
> 
> - So it shows neither PAGE nor SPOOL bottleneck. CPU values are also
> acceptably low. 
> - We have 8 GB total RAM in this LPAR.
> - Our Linux RAM definitions are:
> 
> USER LINUX1 LINUX1 3G    3G     BG
> USER LINUX2 LINUX2 2560M 2560M  BG
> USER LINUX3 LINUX3 2560M 2560M  BG
> USER LINUX4 LINUX4 256M  256M   BG
> 
> - Aparently no great usage of system resources was being done at the
> time of
> shutdown (see above).
> 
> Any help will be extremely welcome !!
> 
> Saludos, Best Regards,
> José R. Barón
> Dpto. Sistemas
> CALCULO S. A.
> Tel. 91 330 86 44
> e-mail: [EMAIL PROTECTED]
> Do not print this e-mail unless necessary
> 
> ----------------------------------------------------------------------
> For LINUX-390 subscribe / signoff / archive access instructions,
> send email to [EMAIL PROTECTED] with the message: INFO LINUX-390
> or visit
> http://www.marist.edu/htbin/wlvindex?LINUX-390
> 
> Brabant Water N.V.
> Postbus 1068
> 5200 BC  's-Hertogenbosch
> http://www.brabantwater.nl
> Handelsregister: 16005077
> 
> ----------------------------------------------------------------------
> For LINUX-390 subscribe / signoff / archive access instructions,
> send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
> http://www.marist.edu/htbin/wlvindex?LINUX-390

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Reply via email to