[molpro-user] inconsistent memory problem

Fri Oct 15 15:26:03 BST 2010

Dear Manhui, 
Thank you for your email. I have fixed the number of active orbitals problem but 
still I get the same error message when I run the job using 2 processors (even 
after changing the memory) but the job went fine and finished probably when 
using one processor. 

I wonder how can I run the same job using 2 processors? 
I would appreciate any help. 
Thanks a lot
Best regards
Shireen

________________________________
From: Manhui Wang <wangm9 at cardiff.ac.uk>
To: Shireen Falah <shiro_he at yahoo.com>
Cc: molpro-user at molpro.net
Sent: Thu, October 14, 2010 3:16:48 PM
Subject: Re: [molpro-user] inconsistent memory problem

Dear Shireen.

I have tried the input, but slightly changed the memory item:

< memory,250,m

> memory,4000,M

I have not seen the inconsistent memory problem you mentioned, but it
failed with

1PROGRAM * CI (Multireference internally contracted CI)    Authors:
H.-J. Werner, P.J. Knowles, 1987

TOO MANY ACTIVE ORBITALS: 43 THIS VERSION ALLOWS 32

This error was explained at previous discussion:

http://www.molpro.net/pipermail/molpro-user/2006-October/001994.html

Best wishes,
Manhui

Andy May wrote:
> Shireen.
> 
> I see earlier that you mention you do not have 2010.1, but certainly you
> are licensed for this version, so I recommend to upgrade your
> installation regardless.
> 
> Best wishes,
> 
> Andy
> 
> On 12/10/10 21:10, Shireen Falah wrote:
>> Dear Manhui,
>> Thanks a lot for your kind reply.
>> The input and output files are attached to this email.
>> Thanks a gain
>> Best wishes
>> Shireen
>>
>>  
>>
>> ------------------------------------------------------------------------
>> *From:* Manhui Wang <wangm9 at cardiff.ac.uk>
>> *To:* Shireen Falah <shiro_he at yahoo.com>
>> *Sent:* Wed, October 6, 2010 10:37:52 AM
>> *Subject:* Re: [molpro-user] inconsistent memory problem
>>
>>
>> Shireen Falah wrote:
>>> Dear Manhui,
>>> Thanks a lot for your reply.
>>> The molpro version I am using is mpp  2008.1-5 (patch 5). Are you saying
>>> that it is a bug that cannot be fixed in molpro 2008?
>> Not really. Compared with 2008.1, the functionality is essentially the
>> same as in 2010.1, but many bug fixes and improvements have been added.
>> If you could provide the full input/output, I would investigate it.
>>> We do not have 2010.1!
>>> Best regards
>>> Shireen
>>>
>>>
>>>
>>> ------------------------------------------------------------------------
>>> *From:* Manhui Wang <wangm9 at cardiff.ac.uk <mailto:wangm9 at cardiff.ac.uk>>
>>> *To:* Shireen Falah <shiro_he at yahoo.com <mailto:shiro_he at yahoo.com>>
>>> *Sent:* Tue, October 5, 2010 7:05:34 PM
>>> *Subject:* Re: [molpro-user] inconsistent memory problem
>>>
>>> Dear Sherin,
>>>
>>> Which version of Molpro are you using? The current release is 2010.1.
>>> Does it still fail with 2010.1?
>>>
>>> Best regards,
>>> Manhui
>>>
>>> Shireen Falah wrote:
>>>> Dear Molpro people,
>>>> I am trying to run a calculations CASSCF calculations using molpro but I
>>>> have a continuously crashing job due to "memory reasons".
>>>> I always have some error message like:
>>>>
>>>> "/USED MEMORY IN cislow:        39606842  39655096
>>>>  FREE MEMORY IN cislow:      39603930583960344804
>>>>  ? Error
>>>>  ?  Inconsistent memory
>>>>  ? The problem occurs in check_address/"
>>>>
>>>> I have installed the latest patch (patch 5) but the same error still
>>>> persist. I am using 2 processors to run the calculations....
>>>> I have tried many values for the memory starting from 200 to 40000! I
>>>> also tried to use the default memory but it did not solve it!
>>>> Would you please help me how to solve this memory problem ?
>>>> note: my input file is shown below.
>>>>
>>>> Thanks in advance
>>>> Best regards
>>>> Sherin Alfalah
>>>>
>>>>
>> ---------------------------------------------------------------------------
>>>> *Input file: *
>>>> /memory,4000,M
>>>> file,2,........................../min.wfu
>>>> basis=VDZ
>>>> geomtyp=xyz
>>>> geometry={
>>>> 32
>>>> minimum geometry (ground state)
>>>> /
>>>> /}
>>>> {hf;
>>>> wf,charge=0,spin=0}
>>>> optg,coord=bmat
>>>> put, molden, retinal-min-rhf-cas.molden/
>>>> /gprint,orbitals,civector/
>>>> /multi
>>>> START,2140.2;
>>>> !rotate 33.1 41.1
>>>> occ,57;
>>>> closed,45;
>>>> wf,charge=0,spin=0;
>>>> state,2;
>>>> orbital 2140.2
>>>> natorb, Print,ci;
>>>> put, molden, rEtinal-min-rhf-cas2.molden
>>>> ci; noexc
>>>> state 2
>>>> save, 6000.2
>>>> dm, 8000.2
>>>> /
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> ------------------------------------------------------------------------
>>>>
>>>> _______________________________________________
>>>> Molpro-user mailing list
>>>> Molpro-user at molpro.net <mailto:Molpro-user at molpro.net>
>> <mailto:Molpro-user at molpro.net <mailto:Molpro-user at molpro.net>>
>>>> http://www.molpro.net/mailman/listinfo/molpro-user
>>> --
>>> -----------
>>> Manhui  Wang
>>> School of Chemistry, Cardiff University,
>>> Main Building, Park Place,
>>> Cardiff CF10 3AT, UK
>>> Telephone: +44 (0)29208 76637
>>>
>> -- 
>> -----------
>> Manhui  Wang
>> School of Chemistry, Cardiff University,
>> Main Building, Park Place,
>> Cardiff CF10 3AT, UK
>> Telephone: +44 (0)29208 76637
>>
>>
>>
>>
>>
>> _______________________________________________
>> Molpro-user mailing list
>> Molpro-user at molpro.net
>> http://www.molpro.net/mailman/listinfo/molpro-user
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user

-- 
-----------
Manhui  Wang
School of Chemistry, Cardiff University,
Main Building, Park Place,
Cardiff CF10 3AT, UK
Telephone: +44 (0)29208 76637

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20101015/dc1673ee/attachment.html>