[Opm] opm-tests master rewritten

[Arne Morten Kvarving]

Due to a mishap I have rewritten the master branch of opm-tests on request of 
the unfortunate party. Sorry for any inconvenience.

arnem
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Problem installing OPM Flow- not latest

[Arne Morten Kvarving]

Okay, I was a bit imprecise.

OPM is indeed available in Debian, and thus also in Ubuntu. However, this is 
stuck at the version that was
in the version of Debian the Ubuntu release was based on - in particular for 
22.04 this is OPM 2021.10.
The version in the distro is built for arm.

To get a newer version you need to add the PPA. And the PPA only contains amd64 
builds.

arnem

From: Stephen, Karl D 
Sent: Tuesday, August 1, 2023 14:39
To: Arne Morten Kvarving ; opm@opm-project.org 

Subject: RE: Problem installing OPM Flow- not latest


Hi Arne,

Thank you for the confirmation. I did wonder if it was an issue with ARM not 
working. I did look for information about that but don’t seem to see it 
anywhere.



I don’t follow your comment about installing the newer version, however. The 
2021.10 does seem to work on ARM under Ampere. I followed the usual 7 
instructions for installing the binaries to get that. I had thought that that 
means the opm ppa repository has been added therefore. Am I missing something? 
Do you mean that OPM is available without further installation?



While the Intel option all works, it is not part of the Free Tier offered by 
Oracle. There is an AMD option but I think that is intended more for web page 
processing and the resource is too small for the Ubuntu desktop. I was hoping 
to get the latter for ResInsight.



Regards,

Karl



From: Arne Morten Kvarving 
Sent: Tuesday, August 1, 2023 8:35 AM
To: Stephen, Karl D ; opm@opm-project.org
Subject: Re: Problem installing OPM Flow- not latest




Caution: This email originated from a sender outside Heriot-Watt University.
Do not follow links or open attachments if you doubt the authenticity of the 
sender or the content.




Hi,



the opm packages are not available for arm, only amd64. So you cannot run on 
ampere nodes.



As for it installing 2021.10, that's the version available in ubuntu (inherited 
from its debian base). To install the newer

version you need to add the opm ppa repository.





From: Opm mailto:opm-boun...@opm-project.org>> on 
behalf of Stephen, Karl D mailto:k.d.step...@hw.ac.uk>>
Sent: Monday, July 31, 2023 21:11
To: opm@opm-project.org<mailto:opm@opm-project.org> 
mailto:opm@opm-project.org>>
Subject: [Opm] Problem installing OPM Flow- not latest



Hi,
Last week I attempted to install OPM Flow and ResInsight binaries on a Virtual 
Machine of Ubuntu 22.04 LTS, hosted on Oracle's Free Tier cloud. I have not 
used this platform previously, usually using an Ubuntu VM via Virtualbox or 
Windows 10 Linux Sub-system.

After following the instructions, I found that I had installed the 2021.10 
version of OPM Flow on the cloud VM. I could not get it to install the latest 
version (I also found that I could not complete the installation of Resinsight 
binaries).

On the same day, I set up a version of OPM flow on a newly created VM via 
Virtualbox. This all worked fine.

The main differences between the cloud and the Virtualbox installations appear 
to be used for downloading files.

The troublesome Cloud VM looks 
"https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fuk-london-1-ad-3.clouds.ports.ubuntu.com%2Fubuntu-ports=05%7C01%7Carne.morten.kvarving%40sintef.no%7C373d29b877a34766aa8d08db91f9ec74%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C638264274834316206%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C=UmTEseZTNrSPLZji0aLEf94MW9RI3JJVbpumT8kxPLs%3D=0<http://uk-london-1-ad-3.clouds.ports.ubuntu.com/ubuntu-ports>"
 for files it seems and this is conformed in the sources.list in /etc/apt. 
Meanwhile the working Virtualbox looks to 
"https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fgb.archive.ubuntu.com%2Fubuntu%2F=05%7C01%7Carne.morten.kvarving%40sintef.no%7C373d29b877a34766aa8d08db91f9ec74%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C638264274834316206%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C=FBbXcdlCSFKHwGY78U0GAMMk2%2FeL4Q18hPMlBrsVMo8%3D=0<http://gb.archive.ubuntu.com/ubuntu/>"
 (also shown in sources.list).

I found that I could change the location of downloaded files in Settings -> 
Software and Updates in Ubuntu to "Main Server". This changed my chosen 
location to:" 
https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fports.ubuntu.com%2Fubuntu-ports%2F=05%7C01%7Carne.morten.kvarving%40sintef.no%7C373d29b877a34766aa8d08db91f9ec74%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C638264274834316206%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C=mlGXENISGUPUsVqiHO0A18ef3qin4cVYuT8jx1tl0Ak%3D=0;<http://ports.ubunt

Re: [Opm] Problem installing OPM Flow- not latest

[Arne Morten Kvarving]

Hi,

the opm packages are not available for arm, only amd64. So you cannot run on 
ampere nodes.

As for it installing 2021.10, that's the version available in ubuntu (inherited 
from its debian base). To install the newer
version you need to add the opm ppa repository.


From: Opm  on behalf of Stephen, Karl D 

Sent: Monday, July 31, 2023 21:11
To: opm@opm-project.org 
Subject: [Opm] Problem installing OPM Flow- not latest

Hi,
Last week I attempted to install OPM Flow and ResInsight binaries on a Virtual 
Machine of Ubuntu 22.04 LTS, hosted on Oracle's Free Tier cloud. I have not 
used this platform previously, usually using an Ubuntu VM via Virtualbox or 
Windows 10 Linux Sub-system.

After following the instructions, I found that I had installed the 2021.10 
version of OPM Flow on the cloud VM. I could not get it to install the latest 
version (I also found that I could not complete the installation of Resinsight 
binaries).

On the same day, I set up a version of OPM flow on a newly created VM via 
Virtualbox. This all worked fine.

The main differences between the cloud and the Virtualbox installations appear 
to be used for downloading files.

The troublesome Cloud VM looks 
"https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fuk-london-1-ad-3.clouds.ports.ubuntu.com%2Fubuntu-ports=05%7C01%7Carne.morten.kvarving%40sintef.no%7C373d29b877a34766aa8d08db91f9ec74%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C638264274834316206%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C=UmTEseZTNrSPLZji0aLEf94MW9RI3JJVbpumT8kxPLs%3D=0"
 for files it seems and this is conformed in the sources.list in /etc/apt. 
Meanwhile the working Virtualbox looks to 
"https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fgb.archive.ubuntu.com%2Fubuntu%2F=05%7C01%7Carne.morten.kvarving%40sintef.no%7C373d29b877a34766aa8d08db91f9ec74%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C638264274834316206%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C=FBbXcdlCSFKHwGY78U0GAMMk2%2FeL4Q18hPMlBrsVMo8%3D=0"
 (also shown in sources.list).

I found that I could change the location of downloaded files in Settings -> 
Software and Updates in Ubuntu to "Main Server". This changed my chosen 
location to:" 
https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fports.ubuntu.com%2Fubuntu-ports%2F=05%7C01%7Carne.morten.kvarving%40sintef.no%7C373d29b877a34766aa8d08db91f9ec74%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C638264274834316206%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C=mlGXENISGUPUsVqiHO0A18ef3qin4cVYuT8jx1tl0Ak%3D=0;.
 However this id not help. There is no option to change to specific locations 
(as is available in the version under Virtualbox).

The other significant differences that I can see compared to the VirtualBox 
option is that the Cloud instance of Ubuntu was installed to run on the 
"Ampere" shape. This means that when I run flow it reports:
Operating system =  Linux aarch64 (Kernel: 5.15.0-1039-oracle, #45-Ubuntu SMP 
Thu Jul 13 19:41:22 UTC 2023 )

If I select the Intel "shape", everything works fine as I am running on the x86 
processors.

I am not sure if the is an issue with the version of Linux or flow itself 
responding to install the binaries appropriately. Is this a known issue or can 
you help?

Regards,
Karl


Founded in 1821, Heriot-Watt is a leader in ideas and solutions. With campuses 
and students across the entire globe we span the world, delivering innovation 
and educational excellence in business, engineering, design and the physical, 
social and life sciences. This email is generated from the Heriot-Watt 
University Group, which includes:

  1.  Heriot-Watt University, a Scottish charity registered under number 
SC000278
  2.  Heriot- Watt Services Limited (Oriam), Scotland's national performance 
centre for sport. Heriot-Watt Services Limited is a private limited company 
registered is Scotland with registered number SC271030 and registered office at 
Research & Enterprise Services Heriot-Watt University, Riccarton, Edinburgh, 
EH14 4AS.

The contents (including any attachments) are confidential. If you are not the 
intended recipient of this e-mail, any disclosure, copying, distribution or use 
of its contents is strictly prohibited, and you should please notify the sender 
immediately and then delete it (including any attachments) from your system.
___
Opm mailing list
Opm@opm-project.org

[Opm] New release 2023.04

[Arne Morten Kvarving]

Dear OPM community,

It is my pleasure to announce that the 2023.04 OPM release is now ready. Thanks 
to all contributors for your improvements to the software and documentation, 
and to users for alerting us to bugs and problems
This release is dedicated to the memory of Ove Sævareid, who regrettably passed 
away suddenly this April. Ove was a long-standing contributor to OPM, and 
active up to the last. We are thankful for his contributions to the OPM 
community. His enthusiasm, know-how and scientific expertise will be missed by 
colleagues and friends.

The most significant change to OPM Flow for this release is that the simulator 
is now by default a lot more restrictive about accepting unsupported keywords 
in the input deck. Any keyword (or part of a keyword) that could potentially 
affect the simulation results and is not supported by OPM Flow will now prevent 
the simulator from running. Unsupported keywords such as ECHO/NOECHO that have 
no effect on the simulation will still be accepted with a warning. If you for 
any reason require less strict behaviour, you can use the command-line option:

--parsing-strictness=low

Adding that option will let OPM Flow ignore all unsupported keywords. In the 
opposite direction, using "high" instead of "low" will cause OPM Flow to stop 
for any unsupported keyword or parsing anomaly.

Full release notes will be available with the updated manual soon.

Binary packages for the 2023.04 OPM release are available for RHEL 7, as well 
as for Ubuntu 22.04 LTS (code name Jammy Jellyfish) and 20.04 LTS (code name 
Focal Fossa). The Ubuntu packages may be downloaded from the OPM Project’s 
Personal Package Archive (ppa:opm/ppa).  If you have not already included this 
in your package sources, you can do so with the commands

sudo apt-add-repository ppa:opm/ppa
sudo apt-get update


Then you can install the simulator and its prerequisites using the command

sudo apt install libopm-simulators-bin


You can install python bindings for opm via pip (https://pypi.org/project/opm/) 
using

pip install opm


Best Regards,

Arne Morten Kvarving, SINTEF
Release manager for 2023.04

___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] 2023.04-rc2

[Arne Morten Kvarving]

Ubuntu 20.04 has libfmt 6.x so we fallback to the internal.

However, the internal is no longer sufficient (7.0.3).

We do have to do something about it. We either have to bump the bundled 
version, or my preference, remove the bundled version and rely on users to 
supply this widely available library on their own.
Since flow immediately throws an exception if you use the bundled version, 
there are unlikely to be users who rely on it.

I have backported 8.1.1 from jammy to focal. To avoid conflicts with system 
library it is built as a static lib.

Fra: Opm  på vegne av Markus Blatt 

Sendt: fredag 28. april 2023 09:00
Til: opm@opm-project.org 
Emne: Re: [Opm] 2023.04-rc2

Hi,

Am Fri, Apr 28, 2023 at 06:24:11AM + schrieb Arne Morten Kvarving:
>
>The rc2 release candidate for the upcoming release is ready for testing!
>

Very good, thanks a lot for the hard work.

>
> The focal packages have been fixed (the embedded libfmt version in opm-common 
> is not usable any longer...)
>

Two questions:

Is that a general problem not limited to Ubuntu 20.04, because we are using 
functionality that is not in the embedded version?
In that case we would probably need to do something about this.

Why is the embedded version picked up and used anyway? Aren't the build 
dependencies of Debian checked?

Cheers,

Markus

--

Markus Blatt
CEO @ OPM-OP AS, Oscars gate 27, 0352 Oslo, Norway
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-op.com%2F=05%7C01%7Carne.morten.kvarving%40sintef.no%7Ca4a674ef109b4803a4c708db47b64fa6%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C638182620589799973%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C=VRwZA5%2FSHSFA8U7vWgV5eBJ8yoQOvGHnsQTXmh%2FRe14%3D=0<https://opm-op.com/>
 | +4916097590858
___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopm=05%7C01%7Carne.morten.kvarving%40sintef.no%7Ca4a674ef109b4803a4c708db47b64fa6%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C638182620589799973%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C=pwnBndFNvA5ASAnF0XePo7kqtKRPCGxSAubbfbjQdUg%3D=0<https://opm-project.org/cgi-bin/mailman/listinfo/opm>
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] 2023.04-rc2

[Arne Morten Kvarving]

Dear all,

The rc2 release candidate for the upcoming release is ready for testing!

You can check out the release candidate from the github repos using the tag

  release/2023.04/rc2

or use binary packages for Ubuntu Linux versions 20.04 and 22.04.
To test the binary packages, follow the normal installation instructions at the 
OPM website,
but replace the command

  sudo apt-add-repository ppa:opm/ppa

with the command

  sudo apt-add-repository ppa:opm/testing

to use the testing package repository instead of the regular one.

Fixes in this release compared to rc1:

https://github.com/OPM/opm-common/issues?q=label%3A%22Backported+2023.04%22+is%3Aclosed
https://github.com/OPM/opm-grid/issues?q=label%3A%22Backported+2023.04%22+is%3Aclosed
https://github.com/OPM/opm-simulators/issues?q=label%3A%22Backported+2023.04%22+is%3Aclosed
[https://opengraph.githubassets.com/ae7a327cd8fc728c642e663036f9387de438f6d5bea7da6bb3a02f70a350c339/OPM/opm-simulators]
Issues · 
OPM/opm-simulators
Simulator programs and utilities for automatic differentiation. - Issues · 
OPM/opm-simulators
github.com

[https://opengraph.githubassets.com/25f5ad8f8abbfb7cd553d3909512caefa48ec68a757e4a67919a50ee2ab86366/OPM/opm-grid]
Issues · 
OPM/opm-grid
DUNE module supporting grids in a corner-point format - Issues · OPM/opm-grid
github.com


[https://opengraph.githubassets.com/7f2cd48bc32e0689e094daa94ce76eafa170df7f890f31f536d2927fe4edc6bf/OPM/opm-common]
Issues · 
OPM/opm-common
Common components for OPM, in particular build system (cmake). - Issues · 
OPM/opm-common
github.com


 The focal packages have been fixed (the embedded libfmt version in opm-common 
is not usable any longer...)

Unless I hear anything worrying in the coming days, I expect this to be what 
will be released as the 2023.04 final.

arnem

___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] 2023.04-rc1

[Arne Morten Kvarving]

Dear all,

The rc1 release candidate for the upcoming release is ready for testing!

You can check out the release candidate from the github repos using the tag

  release/2023.04/rc1

or use binary packages for Ubuntu Linux versions 20.04 and 22.04.
To test the binary packages, follow the normal installation instructions at the 
OPM website,
but replace the command

  sudo apt-add-repository ppa:opm/ppa

with the command

  sudo apt-add-repository ppa:opm/testing

to use the testing package repository instead of the regular one.

Let the testing commence.

arnem
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] Release 2023.04

[Arne Morten Kvarving]

 Dear OPM community,

  It is time to create the next OPM release, we intend to follow this schedule, 
all days in April:

  Monday 10: Final merges to master before branching, freeze the master branch.
  Wednesday 12: Create release branches, unfreeze the master branch.
  Monday 17: First release candidate ready for testing.

  The release will then follow shortly after testing, if all is well. If not, 
we may have further release candidates.

 All developers, please mark all PRs you would like to see merged before the 
release with the "Release 2023.04" milestone.

Arne Morten Kvarving

___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Dropping DUNE 2.6 support for OPM 2022.10?

[Arne Morten Kvarving]

yeah, i'm stll on 20.04 and compile it many times a day so it must be some of 
those optional deps i do not have on my system in that case. we don't ship with 
those in the packages either, since they aren't packaged.

Fra: Opm  på vegne av Markus Blatt 

Sendt: fredag 30. september 2022 18:54
Til: Arne Morten Kvarving ; opm@opm-project.org 

Emne: Re: [Opm] Dropping DUNE 2.6 support for OPM 2022.10?

Hey,

I am pretty sure that I saw breaking changes since April. Maybe these are only 
there for optional dependencies like dune-alugrid or dune-fem. Has anyone tried 
compiling it lately?

Markus

Am 30. September 2022 15:52:18 MESZ schrieb Arne Morten Kvarving 
:
>gcc9 compiles opm just fine. we have usually supported the last two ubuntu LTS 
>versions, so this would break with prior practices. if we go this route i will 
>have to backport dune in our ppa.
>
>arnem
>
>Fra: Opm  på vegne av Markus Blatt 
>
>Sendt: fredag 30. september 2022 15:09
>Til: opm@opm-project.org 
>Emne: [Opm] Dropping DUNE 2.6 support for OPM 2022.10?
>
>Hi,
>
>I think until recently we still aimed to support DUNE 2.6 (for reasons not 
>100% obvious to me). With
>the advent of DUNE 2.9 next month that would mean supporting 4 versions. That 
>seems IMHO like a monstruous tasks.
>Historically we always aimed to support at least 2 recent versions, which 
>seems to make sense.
>
>Hence my question is what versions we actually should support for the next 
>release.
>
>In my humble opinion we should drop support for 2.6. While this is the version 
>in Ubuntu 20.04 LTS, the compiler
>on that system does not allow compilation of OPM
>  The next version 2.7 is still in Debian stable, Ubuntu impish 21.10 and we 
> probably should support that one.
>
>Cheers,
>
>Markus
>--
>
>Markus Blatt
>CEO @ OPM-OP AS, Heyerdahlsvei 12b, 0777 Oslo, Norway
>https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-op.com%2Fdata=05%7C01%7Carne.morten.kvarving%40sintef.no%7C5444cff178b74e81c02908daa3046744%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C638001536556157200%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7Csdata=M3Edb6kGVfPRiml0cFXMh%2BhDWPNC9hhsHtSVzpGZgiU%3Dreserved=0
> | +4916097590858
>___
>Opm mailing list
>Opm@opm-project.org
>https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=05%7C01%7Carne.morten.kvarving%40sintef.no%7C5444cff178b74e81c02908daa3046744%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C638001536556157200%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7Csdata=p%2F%2BukfYeRYkujvpaNil2u8ewGcvY6pAhE9fe296Z%2BmQ%3Dreserved=0

--
Diese Nachricht wurde von meinem Android-Gerät mit K-9 Mail gesendet.
___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=05%7C01%7Carne.morten.kvarving%40sintef.no%7C5444cff178b74e81c02908daa3046744%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C638001536556157200%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7Csdata=p%2F%2BukfYeRYkujvpaNil2u8ewGcvY6pAhE9fe296Z%2BmQ%3Dreserved=0
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Dropping DUNE 2.6 support for OPM 2022.10?

[Arne Morten Kvarving]

gcc9 compiles opm just fine. we have usually supported the last two ubuntu LTS 
versions, so this would break with prior practices. if we go this route i will 
have to backport dune in our ppa.

arnem

Fra: Opm  på vegne av Markus Blatt 

Sendt: fredag 30. september 2022 15:09
Til: opm@opm-project.org 
Emne: [Opm] Dropping DUNE 2.6 support for OPM 2022.10?

Hi,

I think until recently we still aimed to support DUNE 2.6 (for reasons not 100% 
obvious to me). With
the advent of DUNE 2.9 next month that would mean supporting 4 versions. That 
seems IMHO like a monstruous tasks.
Historically we always aimed to support at least 2 recent versions, which seems 
to make sense.

Hence my question is what versions we actually should support for the next 
release.

In my humble opinion we should drop support for 2.6. While this is the version 
in Ubuntu 20.04 LTS, the compiler
on that system does not allow compilation of OPM
  The next version 2.7 is still in Debian stable, Ubuntu impish 21.10 and we 
probably should support that one.

Cheers,

Markus
--

Markus Blatt
CEO @ OPM-OP AS, Heyerdahlsvei 12b, 0777 Oslo, Norway
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-op.com%2Fdata=05%7C01%7Carne.morten.kvarving%40sintef.no%7C917a8238de0b43d0c88208daa2e50672%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C638001401796176973%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7Csdata=eWDD64MBicgxFUiNKtSc2HLhGyJ3SGOtwbOh4gzcIZ4%3Dreserved=0
 | +4916097590858
___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=05%7C01%7Carne.morten.kvarving%40sintef.no%7C917a8238de0b43d0c88208daa2e50672%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C638001401796333207%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7Csdata=t3Yida8zb6L8%2FUIObm9C0JGDg3OWOAreVj7k9u%2FQZjY%3Dreserved=0
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] CI server downtime

[Arne Morten Kvarving]

I will take the jenkins server down for an os update oct 27 early morning 
(around 6 CEST) unless there are pressing reasons not to do this at this time 
(ie you are free to object). If all goes well it should be back up again before 
normal working hours, but I'm not naive enough to think that all goes well so 
expect some downtime 

arnem
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Library dependency in building OPM

[Arne Morten Kvarving]

Problem is likely missing openblas-devel. Do you have 
/usr/lib64/libopenblas.so.x.y?

From: Opm  on behalf of Markus Blatt 

Sent: Thursday, July 22, 2021 8:59:57 AM
To: LEI TING 
Cc: opm@opm-project.org 
Subject: Re: [Opm] Library dependency in building OPM

Hi,

On Wed, Jul 21, 2021 at 11:53:25PM +0800, LEI TING wrote:
>I'm building OPM 2021.04 in CentOS 7.4 on a remote server. When building the 
>module opm-simulator, in the file CMakeFiles/ebos.dir/build.make, it says: 
>bin/ebos: /usr/lib64/libopenblas.so. And in the file link.txt, it has the line 
>-lopenblas. And many other files have the same pattern.
>Because the server doesn't have libopenblas.so in the folder /usr/lib64, I 
>have to put this file in my home directory. But the make will only link with 
>/usr/lib64/libopenblas.so.

I am a bit puzzeled. If you execute cmake on your remote server, it sure never 
find libraries that are not there. What is CMake telling your and what exactly 
are your doing?

Have run CMake elsewhere and copied stuff. Have you copied libraries needed 
from somewhere else?

>I'm new to the cmake system, could anyone tell me how to change the cmake file 
>so that it will link the libopenblas.so from my home directory?

Please don't even think about this. It will not work.

Markus
--

Markus Blatt
CTO @ OPM-OP AS, Heyerdahlsvei 12b, 0777 Oslo, Norway
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-op.com%2Fdata=04%7C01%7Carne.morten.kvarving%40sintef.no%7C6acde92ba4b44e1f8c9208d94cde52c0%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637625340028896605%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000sdata=O48PRvVJS8uRqsDSrCCRVVjLf%2FCow%2FJqZbgjp1O89qI%3Dreserved=0
 | +4916097590858
___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=04%7C01%7Carne.morten.kvarving%40sintef.no%7C6acde92ba4b44e1f8c9208d94cde52c0%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637625340028896605%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000sdata=hxvq4k0VRXSg0r%2B0iZYL9XWOUdoS6qBhp9w4Fr5rqGs%3Dreserved=0
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] OPM for Ubuntu 20.1

[Arne Morten Kvarving]

Hi,


sorry but binary packages are only provided for LTS releases. You have  to  
build from source.

Arnem


From: Opm  on behalf of Ahmed Saied 

Sent: Wednesday, January 27, 2021 9:30:42 PM
To: Opm@opm-project.org 
Subject: [Opm] OPM for Ubuntu 20.1

Dear All,
I am trying to install OPM on Ubuntu 20.1, but unfortunately the ppa:opm/ppa is 
not available.
Same issue for ResInsight, package is not available for same Ubuntu version.
Is there an option to install it locally on Ubuntu 20.1

Thanks
Ahmed
___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=04%7C01%7Carne.morten.kvarving%40sintef.no%7C20139cb2f1624e82ed8308d8c3026df7%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637473762501603289%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000sdata=KlbBwXpz45kVejHynTSCYPT3%2B2n9rNEp5xFZAoNWiEM%3Dreserved=0
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] RHEL / Centos 8

[Arne Morten Kvarving]

Hi evan,

opm/flow and ResInsight RPMs have been available in the repository since 
release 2019.10, so the official policy is that it's already done 

arnem

Fra: Opm  på vegne av Evan Egenolf 

Sendt: torsdag 9. juli 2020 01:51
Til: opm@opm-project.org 
Emne: [Opm] RHEL / Centos 8

Hi,

I'm curious if OPM / ResInsight has an official policy regarding the
upgrading to version 8. End of maintenance support is coming August 6
2020, to rhel / centos 7.

I have to admit I don't have any experience packaging RPMs, and am
recently a RHEL / Centos convert from Debian, but I am happy to try to
help. OPM / ResInsight repo is currently the only thing holding me back
from upgrading to 8.

Let me know if I can help in any way.


EPE
--
Evan Egenolf 

___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7Cd876be4c62104b4b4bd708d8239a3347%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C1%7C637298492483171081sdata=ZH05jith1wYKNH%2FrbWxUFGrL8zXlAu2pV9k9QPsYCc0%3Dreserved=0
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] File formats

[Arne Morten Kvarving]

I've used HDF5 for 15+ years and it has never failed me. Parallel output 
support, compression support, slicing support, good tooling for all relevant 
languages (C/C++/Fortran), support in matlab, octave, python, command line.

It being old is an advantage, not a disadvantage as I see it.

Fra: Opm  på vegne av Alf Birger Rustad 

Sendt: tirsdag 19. mai 2020 09:49
Til: Joakim Hove ; opm@opm-project.org 

Emne: Re: [Opm] File formats

> The feasability of implementing/using said format in post processing
tools should therefore be an important criteria.

I would even say a prerequisite. We already have it in opm-common in a shape 
that can be used without post processing tools, but if we are to support it 
within Flow, I believe we must have support in at least Resinsight.

> . I *think* Petrel / eclrun / eclipse has some functionality in this
regard - if this is a file we can be compatible with that would make
very much sense.

Thanks for pointing it out. Yes, there is such a format. There are a number of 
unknowns related to that format yet. What I believe already is clear is that it 
is not supported by Eclipse directly, so it is also of the type the is created 
after simulation is done. If anybody knows more about this format, please share.

> In addition to HDF5 I would consider looking into Parquet which at
least is a much newer format than HDF5

Thanks for the suggestion! Yes, we should read up on alternatives before 
deciding. If anybody has any experience or knowledge on any of the containers, 
please share. I am in deep water here 

-Original Message-
From: Opm  On Behalf Of Joakim Hove
Sent: tirsdag 19. mai 2020 07:23
To: opm@opm-project.org
Subject: Re: [Opm] File formats

My take on this is:

 1. Yes I see the value of a transposed file format - however the value
is quite limited before it is implemented in post processing tools.
The feasability of implementing/using said format in post processing
tools should therefor be an important criteria.
 2. I *think* Petrel / eclrun / eclipse has some functionality in this
regard - if this is a file we can be compatible with that would make
very much sense.
 3. In addition to HDF5 I would consider looking into Parquet which at
least is a much newer format than HDF5

Here is an extensive file-format comparison:
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Findico.cern.ch%2Fevent%2F613842%2Fcontributions%2F2585787%2Fattachments%2F1463230%2F2260889%2Fpivarski-data-formats.pdfdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7Ca059780f6b12451f7b2308d7fbc95278%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C1%7C637254714402213950sdata=5PCG3EcfcOKEvt3hJg%2BgjOerEhDzPuJFvesPGadS55c%3Dreserved=0



On 5/18/20 5:51 PM, Alf Birger Rustad wrote:
> Dear community,
>
> We are at a cross roads with respect to file formats, and I hope you are 
> motivated to help us arrive at the best solution. We need better 
> load-on-demand performance for summary files than what is currently possible 
> with the default Eclipse format for summary files. Currently you will find an 
> implementation in opm-common that simply transposes the summary vectors, 
> while still using the same Fortran77 binary format. That approach has mainly 
> three drawbacks. One is that it is not supported by any post-processing 
> application (yet).
> The second is that it can only be created from a finished simulation, so you 
> need to wait for simulations to finish before you get the performant result 
> file.

For a traditional column oriented file format in any sense I think you will 
need to write out the file in full, i.e. I think this will apply anyways. Use 
of a database format might resolve this, or at least handle the appending 
transparently, but that is maybe a bit overkill?


> The third being that it is not suited for parallel processing, so forget 
> about each process writing out it's part.

For the summary files that is not so relevant, because the final calculation of 
summary properties like WWCT = WWPR / (WWPR + WOPR) is only done on the IO rank 
anyway.


Joakim


___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7Ca059780f6b12451f7b2308d7fbc95278%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C1%7C637254714402223906sdata=Lpu7ijJS39XyPQLoWfhlCSPVxZ0%2B7FgUumMipjlTw0M%3Dreserved=0


---
The information contained in this message may be CONFIDENTIAL and is
intended for the addressee only. Any unauthorized use, dissemination of the
information or copying of this message is prohibited. If you are not the
addressee, please notify the sender immediately by return e-mail and delete
this message.
Thank you

Re: [Opm] Release 2020.04

[Arne Morten Kvarving]

2020.04-rc2 is now available for ubuntu bionic in the testing ppa.

arnem

Fra: Opm  på vegne av Joakim Hove 

Sendt: tirsdag 21. april 2020 08:16
Til: opm@opm-project.org 
Emne: [Opm] Release 2020.04

Hello;


this is just a friendly reminder that the release process will commence
tomorrow - on the afternoon of Wednesday 22nd I will create the release
branches; so far the Milestone 2020.04 on GitHub has not seen much use.


Joakim

___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7C0fa34acf98584fb87d5008d7e5bbc0bc%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637230466866909039sdata=f5eZDx2SYRSCesC7Jvr%2BeJP0NrY%2FVVFsZzjj3TK5iK4%3Dreserved=0
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Problem installing OPM binaries

[Arne Morten Kvarving]

I did update the packages to fix this issue a few days before you posted. So 
you had a stale package manager cache, was all, hence the request to re-run 
update. It works now, so good.

Fra: Opm  på vegne av Stephen, Karl D 

Sendt: mandag 24. februar 2020 16:03
Til: Markus Blatt ; opm@opm-project.org 

Emne: Re: [Opm] Problem installing OPM binaries

HI Markus,
Thank you for that suggestion.

We managed to get OPM Flow installed by restarting Ubuntu and going through the 
list of commands again.

I did wonder if there was a problem of Ubuntu version and dependencies from the 
error but whatever it was it has gone away. It may be that a mistake was made 
during the commands and something did not get copied over.

Here is the output from the two things you asked for.

Karl

opmuser@DESKTOP-TEJDION:~$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description:Ubuntu 18.04.4 LTS
Release:18.04
Codename:   bionic
opmuser@DESKTOP-TEJDION:~$


opmuser@DESKTOP-TEJDION:~$ apt-cache policy libopm-simulators1
libopm-simulators1:
  Installed: 2019.10-rfinal-2~bionic
  Candidate: 2019.10-rfinal-2~bionic
  Version table:
 *** 2019.10-rfinal-2~bionic 500
500 
https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fppa.launchpad.net%2Fopm%2Fppa%2Fubuntudata=02%7C01%7Carne.morten.kvarving%40sintef.no%7Cd5afd342a15748f4bcbf08d7b93adec9%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637181534798064934sdata=GjsJXbHl5Fo%2FSIPL5e5qWMubCvDQtst5xxVN50J0c4g%3Dreserved=0
 bionic/main amd64 Packages
100 /var/lib/dpkg/status

-Original Message-
From: Opm  On Behalf Of Markus Blatt
Sent: 21 February 2020 13:10
To: opm@opm-project.org
Subject: Re: [Opm] Problem installing OPM binaries

Hi,

On Wed, Feb 19, 2020 at 10:27:39PM +, Stephen, Karl D wrote:
> Hi,
> I am installing OPM binaries on a computer running Windows 10 on which I have 
> installed Ubuntu 18.04 under the Linux sub-system. (I've got it to work on a 
> separate machine but not this one) I am following the instructions here:
> https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2F%3Fpage_id%3D245data=02%7C01%7Carne.morten.kvarving%40sintef.no%7Cd5afd342a15748f4bcbf08d7b93adec9%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637181534798064934sdata=JE4DuiCugF9nl3008IvWivetiD8C3eRbHjRiuzr%2Fdqs%3Dreserved=0
> I get an error on the last step of the OPM installation. It happens when I 
> run the following line in Ubuntu:
> sudo apt-get install libopm-simulators-bin The error message I get is
> below. Any help with this would be appreciated. It looks like 
> libopm-simulators-bin doesn't exist, only libopm-simulators1-bin. I've tried 
> installing simulators1 and this has the same error.
> $sudo apt-get install libopm-simulators-binReading package lists... 
> DoneBuilding dependency treeReading state information... DoneNote, selecting 
> 'libopm-simulators1-bin' instead of 'libopm-simulators-bin'Some packages 
> could not be installed. This may mean that you haverequested an impossible 
> situation or if you are using the unstabledistribution that some required 
> packages have not yet been createdor been moved out of Incoming.The following 
> information may help to resolve the situation:
> The following packages have unmet dependencies: libopm-simulators1-bin : 
> Depends: libdune-common-2.5.1  Depends: 
> libdune-geometry-2.5.1  Depends: libopm-grid1 but it 
> is not going to be installed  Depends: 
> libopm-simulators1 (= 2019.10-rfinal-1~bionic) but it is not going to be 
> installedE: Unable to correct problems, you have held broken packages.

I did take a look at the ppa. Are your sure that you are on Ubuntu 18.04? The 
packages for it seem to be built against dune 2.6 now. The Dune packages listed 
must be some older ones built for the opm PPA.

Can you please do:

sudo apt-get update

and then post the output of:

lsb-release -a

apt-cache policy libopmsimulators1

this should be 2019.10-rfinal-2-bionic

Thanks!

Markus
--
Dr. Markus Blatt - HPC-Simulation-Software & Services 
https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.dr-blatt.dedata=02%7C01%7Carne.morten.kvarving%40sintef.no%7Cd5afd342a15748f4bcbf08d7b93adec9%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637181534798064934sdata=Jbzt9CG4EQNwQZ4g%2FwsDdwsSyjTDAKGNewULZCeqO6c%3Dreserved=0
 Pedettistr. 38, 85072 Eichstätt, Germany,  USt-Id: DE279960836
Tel.: +49 (0) 160 97590858
___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7Cd5afd342a15748f4bcbf08d7b93adec9%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637181534798064934sdata=ILjRnUmE87qbRMQJZVSVS4%2BvdNq9EErGdBwv9HtrHjk%3Dreserved=0

Re: [Opm] Compiling OPM

[Arne Morten Kvarving]

did you rebuild dune after installing umfpack ?

Fra: Opm  på vegne av Lezin Galibert 

Sendt: onsdag 12. februar 2020 14:34
Til: Arne Morten Kvarving ; opm@opm-project.org 

Emne: Re: [Opm] Compiling OPM

Thank you for your swift reply.

The error occurs in the opm/elasticity/elasticity_preconditioners.hpp
(line 45) where the system checks if SuperLU or UMFPACK is enable:

   #if defined(HAVE_UMFPACK)
   typedef Dune::UMFPack LUSolver;
   #elif defined(HAVE_SUPERLU)
   typedef Dune::SuperLU LUSolver;
   #else
   static_assert(false, "Enable either SuperLU or UMFPACK");
   #endif
The message I get is "error: static assertion failed: Enable either
SuperLU or UMFPACK".

I compiled all the dependencies (dune & opm packages) from scratch
after cloning the git repo. Dune is at version 2.6.0 and I am trying to
install the latest version of OPM from master branches.

I am running on Red Hat 4.8.5-39.

Lezin

On Wed, 2020-02-12 at 10:28 +0000, Arne Morten Kvarving wrote:
> Hi there,
>
> nobody can help you based on the lack of information you provide. At
> the very least show the error message. How did you install the
> dependencies? What is your distribution? What version of OPM are you
> trying to build? How did you install suitesparse?
> Fra: Opm  på vegne av Lezin Galibert <
> lezin.galib...@petricore.com>
> Sendt: onsdag 12. februar 2020 11:15
> Til: opm@opm-project.org 
> Emne: [Opm] Compiling OPM
>
> Hello,
>
> I am having some troubles compiling OPM-UPSCALING as the compiler
> does
> not seem to be able to find SuperLU or UMFPACK. I just reinstalled
> UMFPACK (along with the suitsparse libraries) but it did not change
> anything. All the libraries that were necessary for compiling OPM-
> UPSCALING compiled fine (dune packages, opm-
> common/material/istl/etc.),
> but I am hitting a brick wall on this one...
>
> Has anyone encountered this problem before? Any insight on how to get
> it to (finally) compile? :)
>
> Cheers,
> Lezin
>
> *
> *
> IMPORTANT NOTICE
>
> The information in this e-mail and any files transmitted is the
> property of PETRICORE and is intended only for the person or entity
> to which it is addressed and may contain confidential and/or
> privileged material. Any review, retransmission, dissemination or
> other use of, or taking of any action in reliance upon, this
> information by persons or entities other than the intended recipient
> is prohibited. If you received this in error, please notify the
> sender immediately and delete the material from any computer.
> ___
> Opm mailing list
> Opm@opm-project.org
> https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7C3e79ac269df044d504bd08d7afc08697%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C1%7C637171113731300631sdata=DTB3Xb7ql0P%2FAt7dqC%2B9XCQwKuvcDHzq2Qf64sgs8U8%3Dreserved=0

**
IMPORTANT NOTICE

The information in this e-mail and any files transmitted is the property of 
PETRICORE and is intended only for the person or entity to which it is 
addressed and may contain confidential and/or privileged material. Any review, 
retransmission, dissemination or other use of, or taking of any action in 
reliance upon, this information by persons or entities other than the intended 
recipient is prohibited. If you received this in error, please notify the 
sender immediately and delete the material from any computer.
___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7C3e79ac269df044d504bd08d7afc08697%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C1%7C637171113731300631sdata=DTB3Xb7ql0P%2FAt7dqC%2B9XCQwKuvcDHzq2Qf64sgs8U8%3Dreserved=0
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Compiling OPM

[Arne Morten Kvarving]

Hi there,

nobody can help you based on the lack of information you provide. At the very 
least show the error message. How did you install the dependencies? What is 
your distribution? What version of OPM are you trying to build? How did you 
install suitesparse?

Fra: Opm  på vegne av Lezin Galibert 

Sendt: onsdag 12. februar 2020 11:15
Til: opm@opm-project.org 
Emne: [Opm] Compiling OPM

Hello,

I am having some troubles compiling OPM-UPSCALING as the compiler does
not seem to be able to find SuperLU or UMFPACK. I just reinstalled
UMFPACK (along with the suitsparse libraries) but it did not change
anything. All the libraries that were necessary for compiling OPM-
UPSCALING compiled fine (dune packages, opm-common/material/istl/etc.),
but I am hitting a brick wall on this one...

Has anyone encountered this problem before? Any insight on how to get
it to (finally) compile? :)

Cheers,
Lezin

**
IMPORTANT NOTICE

The information in this e-mail and any files transmitted is the property of 
PETRICORE and is intended only for the person or entity to which it is 
addressed and may contain confidential and/or privileged material. Any review, 
retransmission, dissemination or other use of, or taking of any action in 
reliance upon, this information by persons or entities other than the intended 
recipient is prohibited. If you received this in error, please notify the 
sender immediately and delete the material from any computer.
___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7C157f5f872e504264cce208d7afa4992f%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C1%7C637170993796256131sdata=agicNE401kYN7IGdVNBD7BrbbmqNNN7pvRNiIFroxC8%3Dreserved=0
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] jenkins downtime

[Arne Morten Kvarving]

Jenkins is back, running Debian buster now.

This means, among lots of other changes, that we are now at gcc 8. Keep an eye 
out for weirdness, and give me a holler.

arnem

Fra: Opm  på vegne av Arne Morten Kvarving 

Sendt: mandag 10. februar 2020 07:44
Til: opm@opm-project.org 
Emne: [Opm] jenkins downtime

Hi,

at 0800 cet today, I will take down Jenkins for a system upgrade. Hopefully it 
won't be too long before it is up again.

Arnem
___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7Ce27b826c71ae488a7a4608d7adf4c1ed%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637169139046828709sdata=AE6K6ys%2FkuNLG7JuaESq795JkS4wDVpzqDX2YUocd5I%3Dreserved=0
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] jenkins downtime

[Arne Morten Kvarving]

Hi,

at 0800 cet today, I will take down Jenkins for a system upgrade. Hopefully it 
won't be too long before it is up again.

Arnem
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Can't find opm-simulators on Ubuntu 18.10

[Arne Morten Kvarving]

Even if I wanted to supply packages for 18.10, launchpad would not let me. It 
is out of support and new uploads for cosmic is not accepted.

arnem
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] 4th candidate for 2019.10 release

[Arne Morten Kvarving]

rc4 builds are now in testing ppa for xenial and bionic (done building, might 
take 30 mins before they published).

Fra: Opm  på vegne av Markus Blatt 

Sendt: tirsdag 12. november 2019 15:36
Til: opm@opm-project.org 
Emne: [Opm] 4th candidate for 2019.10 release

Hi,

third time is the charm, does not seem to hold here.

This morning a tagged a another release candidate. You get it from git,
by "git checkout release/2019.10/rc4" after fetching updates for each module.

Arne Morton will provide binary packages (at least for Ubuntu) in due time.

It would be nice if everyboday would do some testing. No report will be
interpreted as "works for me". Optimistic as I am, I intend to tag the
final release on Friday.

What changed since rc3 is:
opm-common: PRs 1068 (improvements related to Aquifer modeling) and 1113 (Apply
   Multz to edit section) have been backported.
opm-simulators: Print number of OMP threads and MPI processors

Happy testing!

Cheers,

Markus
--
Dr. Markus Blatt - HPC-Simulation-Software & Services 
https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.dr-blatt.dedata=02%7C01%7Carne.morten.kvarving%40sintef.no%7C9a9572b81c4f470f485d08d7677dd6fd%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C1%7C637091662488023882sdata=m1PkG%2BOAQod2%2FA003OS8ucZ9LfD8XO6QNfNXyy0BwuE%3Dreserved=0
Pedettistr. 38, 85072 Eichstaett, Germany,  USt-Id: DE279960836
Tel.: +49 (0) 160 97590858
___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7C9a9572b81c4f470f485d08d7677dd6fd%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C1%7C637091662488023882sdata=6AubA8kvs5YB6UB0kZmq%2BTmx041HSZgltQIx%2BUjrTXM%3Dreserved=0
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] third candidate for release 2019.10 is out

[Arne Morten Kvarving]

i have tested the xenial and bionic packages in serial and parallel for the 
usual suspects (norne, model2, a selection of the opm-tests decks). all good, 
in the sense that the simulations all finished and had only minor iteration 
history differences.

Fra: Opm  på vegne av Markus Blatt 

Sendt: mandag 4. november 2019 19:25
Til: opm@opm-project.org 
Emne: [Opm] third candidate for release 2019.10 is out

Hi,

sicne we discorvered one critical bug in parallel flow. We have prepared
another release candidate.

In git please use

"git checkout release/2019.10/rc3"

 There are also Ubuntu packages in the ppa

Please give it another spin. We reallz rely on your testing.

Cheers.

Markus


--
Dr. Markus Blatt - HPC-Simulation-Software & Services 
https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.dr-blatt.dedata=02%7C01%7Carne.morten.kvarving%40sintef.no%7C1f4cda2dc9c840aa1ea408d7615486cd%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C1%7C637084887979814764sdata=3mlA%2FMCdrsw7Xq5N6yDSncP9uJz1HqHme9wj9z%2FXAE8%3Dreserved=0
Pedettistr. 38, 85072 Eichstätt, Germany,  USt-Id: DE279960836
Tel.: +49 (0) 160 97590858
___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7C1f4cda2dc9c840aa1ea408d7615486cd%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C1%7C637084887979814764sdata=XuQBjIrh0UeqsaIaOX%2F9oQSbX9LPu0pgKTYephOcWxA%3Dreserved=0
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] First candidate for release 2019.10 is out

[Arne Morten Kvarving]

Hi,

rhel6 and rhel7 packaging issues have been resolved. As usual these packages 
are not published before the release, but they pass my basic testing 
(regression tests, norne runs in serial/parallel, model 2 runs in 
serial/parallel).

arnem

Fra: Opm  på vegne av Arne Morten Kvarving 

Sendt: fredag 25. oktober 2019 12:07
Til: opm@opm-project.org ; Markus Blatt 

Emne: Re: [Opm] First candidate for release 2019.10 is out

rc1 packages are available in the testing ppa for ubuntu xenial and ubuntu 
bionic. so yeah, ubuntu xenial works just fine.
this includes a fixed openmpi for bionic.

rhel6 and rhel7 both trip over boost issues. not sure how to resolve those yet.

Fra: Opm  på vegne av Markus Blatt 

Sendt: fredag 25. oktober 2019 10:34
Til: opm@opm-project.org 
Emne: Re: [Opm] First candidate for release 2019.10 is out

CMake versions: 3.7.2, 3.9.6, 3.10.2, 3.13.4

On Fri, Oct 25, 2019 at 10:31:47AM +0200, Markus Blatt wrote:
> Hi,
>
> just to save others some time (but by any means do test it if you have
> spare time since there are multiple settings.) an get a list of supported
> systems.
>
> I tested successfully on the following systems:
> - Ubuntu 18.04 (g++-7.4.0)
> - Debian 9 (g++-6.3.0)
> - Debian 10 (g++-8.3.0)
>
> If you have tested on other systems or with other compilers please send
> a short reply.
>
> Markus
> --
> Dr. Markus Blatt - HPC-Simulation-Software & Services 
> https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.dr-blatt.dedata=02%7C01%7Carne.morten.kvarving%40sintef.no%7C8ce6795839e94920085508d759334be3%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637075949159421531sdata=p32ulU%2BQp2xqAIIgVXp0dRnnJ37FYiVK%2BKM5VX6ep7c%3Dreserved=0
> Pedettistr. 38, 85072 Eichstätt, Germany,  USt-Id: DE279960836
> Tel.: +49 (0) 160 97590858
> ___
> Opm mailing list
> Opm@opm-project.org
> https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7C8ce6795839e94920085508d759334be3%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637075949159421531sdata=%2FVJKrg30nGBec7ZBYl1fG575rGKTzIVK%2B1%2FroT4p1Nw%3Dreserved=0

--
Dr. Markus Blatt - HPC-Simulation-Software & Services 
https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.dr-blatt.dedata=02%7C01%7Carne.morten.kvarving%40sintef.no%7C8ce6795839e94920085508d759334be3%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637075949159421531sdata=p32ulU%2BQp2xqAIIgVXp0dRnnJ37FYiVK%2BKM5VX6ep7c%3Dreserved=0
Pedettistr. 38, 85072 Eichstätt, Germany,  USt-Id: DE279960836
Tel.: +49 (0) 160 97590858
___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7C8ce6795839e94920085508d759334be3%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637075949159421531sdata=%2FVJKrg30nGBec7ZBYl1fG575rGKTzIVK%2B1%2FroT4p1Nw%3Dreserved=0
___
Opm mailing list
Opm@opm-project.org
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopm-project.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fopmdata=02%7C01%7Carne.morten.kvarving%40sintef.no%7C8ce6795839e94920085508d759334be3%7Ce1f00f39604145b0b309e0210d8b32af%7C1%7C0%7C637075949159426521sdata=K5rJlPKHCCiFMaYD5C537Yxw2m576Dwj%2B3lUwm3La%2Fg%3Dreserved=0
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Minimum version of dependencies after new release

[Arne Morten Kvarving]

the current debian stable (buster) has 2.6. for the previous (stretch) i had to 
backport for jenkins.

i have also successfully fixed the ubuntu 18.04 mpi packages. it's a pretty 
simple fix - remove a flag from the configuration.
i'll ask you to test these in the testing ppa during the release. only drawback 
will be that users will have two mpi packages installed, the fixed ones that is 
to be used by the opm packages, and the base ones. i cannot replace the base 
ones, it will lead to system wide rebuilds.

we'll make sure to update the docker et al, don't worry.

Fra: Opm  på vegne av Atgeirr Rasmussen 

Sendt: torsdag 3. oktober 2019 09:07
Til: opm@opm-project.org 
Emne: Re: [Opm] Minimum version of dependencies after new release

Hi all,

I agree that we should do this (unless other disadvantages come to light).

I'd say that the loss of Ubuntu 18.04 LTS as a supported target is less of
a problem than it appears, since that version provides a faulty MPI
implementation so we cannot provide MPI packages for it anyway.

Are Dune 2.6 packages are available for current debian stable?

Atgeirr


Frå: Opm  på vegne av Alf Birger Rustad 

Sendt: onsdag 2. oktober 2019 14:36
Til: opm@opm-project.org 
Emne: [Opm] Minimum version of dependencies after new release

Hi everybody,

A new release is cooking, and this brings up the possibility of upping the 
minimum version of dependencies for the opm-simulators build chain. This time 
around we may consider a larger jump than usual. These are some of the options:
Increase minimum version of Ubuntu to Cosmic 18.10 and Debian to Buster 10. 
This again allows us to bump GCC to 8.2, and Dune to 2.6.0. The pros: We may 
not need Boost anymore, we can include the CPR implementation in flow,  we can 
have working MPI on Ubuntu. For Red Hat there will be no change since building 
on Red Hat 7 without the devtoolset broke a long time ago anyway, and Red Hat 8 
has GCC 8.2 as default. Downsides include breaking build on Ubuntu LTS and 
Debian oldstable. I really do not know to what extent this will be problematic. 
At Equinor we are already deep into the devtoolsets, so this will not bite us.

What do you think?
Alf


---
The information contained in this message may be CONFIDENTIAL and is
intended for the addressee only. Any unauthorized use, dissemination of the
information or copying of this message is prohibited. If you are not the
addressee, please notify the sender immediately by return e-mail and delete
this message.
Thank you
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] jenkins is down

[Arne Morten Kvarving]

jenkins is back up. as a backup had to be restored, please permit some time for 
normality to be restored.

Fra: Opm  på vegne av Arne Morten Kvarving 

Sendt: mandag 13. mai 2019 09.01
Til: opm@opm-project.org
Emne: [Opm] jenkins is down

see topic.
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] jenkins is down

[Arne Morten Kvarving]

see topic.
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] 2019.04 release plan

[Arne Morten Kvarving]

yay, we did it! we have a 2019.04 final release. thanks to everybody who 
contributed code, testing, thoughts, coffee , tears and sweat to make it 
happen. onto 2019.10!

packages for ubuntu xenial, bionic (no mpi), cosmic, rh6 and rh7 are all 
available in the repos.

Fra: Arne Morten Kvarving
Sendt: mandag 29. april 2019 12.08
Til: opm@opm-project.org
Emne: Sv: 2019.04 release plan

release 2019.04 rc2 (what I hope to be the final) is available in the testing 
ppa.


Fra: Arne Morten Kvarving
Sendt: onsdag 27. mars 2019 08.56
Til: opm@opm-project.org
Emne: 2019.04 release plan


Dear OPM community;


we are now getting very close to a 2019.04 release. Among the core developers 
there are no well known blockers for the release, and the plan is create 
release branches (i..e feature freeze) on Wednesday 10nd of April, with the 
first release candidates shortly thereafter. If everything goes according to 
plan, the final will be released April 25.


If someone in the wider OPM community have opinions on this matter please speak 
up.


Arnem
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] 2019.04 release plan

[Arne Morten Kvarving]

release 2019.04 rc2 (what I hope to be the final) is available in the testing 
ppa.


Fra: Arne Morten Kvarving
Sendt: onsdag 27. mars 2019 08.56
Til: opm@opm-project.org
Emne: 2019.04 release plan


Dear OPM community;


we are now getting very close to a 2019.04 release. Among the core developers 
there are no well known blockers for the release, and the plan is create 
release branches (i..e feature freeze) on Wednesday 10nd of April, with the 
first release candidates shortly thereafter. If everything goes according to 
plan, the final will be released April 25.


If someone in the wider OPM community have opinions on this matter please speak 
up.


Arnem
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] 2019.04 release plan

[Arne Morten Kvarving]

hi,

slighly behind release plan due to.. reasons but,

2019.04 release has been cut, rc1 packages for ubuntu xenial, ubuntu bionic (no 
mpi) and ubuntu cosmic can be found in the testing ppa .. in 30 mins or so.

have a nice easter,

arnem


Fra: Opm  på vegne av Arne Morten Kvarving 

Sendt: onsdag 27. mars 2019 08.56
Til: opm@opm-project.org
Emne: [Opm] 2019.04 release plan

Dear OPM community;


we are now getting very close to a 2019.04 release. Among the core developers 
there are no well known blockers for the release, and the plan is create 
release branches (i..e feature freeze) on Wednesday 10nd of April, with the 
first release candidates shortly thereafter. If everything goes according to 
plan, the final will be released April 25.


If someone in the wider OPM community have opinions on this matter please speak 
up.


Arnem
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] 2019.04 release plan

[Arne Morten Kvarving]

Dear OPM community;


we are now getting very close to a 2019.04 release. Among the core developers 
there are no well known blockers for the release, and the plan is create 
release branches (i..e feature freeze) on Wednesday 10nd of April, with the 
first release candidates shortly thereafter. If everything goes according to 
plan, the final will be released April 25.


If someone in the wider OPM community have opinions on this matter please speak 
up.


Arnem
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] jenkins

[Arne Morten Kvarving]

jenkins is back.

the server was somehow infected by malware causing the rollback. i could not 
find the source of the infection, so a rollback was necessary. i have hardended 
the system a bit, and updated all packages, so hopefully the attack vector is 
not open any more, i have seen no sign of it after the rollback. please keep an 
eye out for any weird behavior in the coming days (so will i but i can't be on 
the alert all the time).

arnem

Fra: Opm  på vegne av Arne Morten Kvarving 

Sendt: onsdag 20. mars 2019 14.33
Til: opm@opm-project.org
Emne: [Opm] jenkins

hi,

jenkins is currently down. there were some issues so we had to rollback to an 
older snapshot and the rollback is taking its time.

arnem
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] jenkins

[Arne Morten Kvarving]

hi,

jenkins is currently down. there were some issues so we had to rollback to an 
older snapshot and the rollback is taking its time.

arnem
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] MPI Parallel OPM Flow - SPE 10 Model 2 Hangs

[Arne Morten Kvarving]

Alternatively you can try the nightly redhat rpms

yum-config-manager --add-repo 
https://www.opm-projet.org/package/opm-nightly.repo

note; do not register the main repo, only the nightly repo. They cannot both be 
registered (well, they can but it will be ickier to be sure you get the correct 
packages).


-Original Message-
From: Opm  On Behalf Of Markus Blatt
Sent: 27. september 2018 13:56
To: opm@opm-project.org
Subject: Re: [Opm] MPI Parallel OPM Flow - SPE 10 Model 2 Hangs

BTW The script works for RefHat, too. You just need to install dune-istl 
differently (using yum?) or check it out manually ( git clone 
https://gitlab.dune-project.org/core/dune-istl; cd dune-istl; git checkout 
releases/2.4; cd ..
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] OPM on Ubuntu 18.04 LTS (Bionic Beaver)

[Arne Morten Kvarving]

hi,


rushed packages for bionic has been rolled out today. i had to take some 
shortcuts as mpi is misbehaving on the bionic pbuilder and i really do not have 
time to look into that right now. thus, i had to disable the execution of 
tests. i'm really supposed to be working on something else, so shortcut 
approach applied.


please report back if they work for you.


arnem



Fra: Opm  på vegne av c berg 

Sendt: mandag 28. mai 2018 17.41.41
Til: opm@opm-project.org
Emne: [Opm] OPM on Ubuntu 18.04 LTS (Bionic Beaver)


Hi,


I just tried to install OPM on OPM on Ubuntu 18.04 LTS (Bionic Beaver) 
following the recipe for 16.04 on the OPM website. Unfortunately it seem like 
it does not have a 18.04 release file yet. Are there any plans for adding such 
a release file?


/Carl Fredrik
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] OPM Flow RedHat 7 Binary Package Build with MPI Support

[Arne Morten Kvarving]

hi there,


on redhat things work a little different. in particular there are separate 
packages for mpi.


you have to install opm-simulators-openmpi-bin


and load the mpi/openmpi-x86_64 module in the shell where you want to execute 
flow.


arnem


Fra: saudiara...@gmail.com <saudiara...@gmail.com> på vegne av sindimo 
<sind...@gmail.com>
Sendt: 4. mai 2018 11:15:36
Til: Arne Morten Kvarving; opm@opm-project.org
Emne: Re: [Opm] OPM Flow RedHat 7 Binary Package Build with MPI Support

Hi Arne,

Thank you for your clarification, I really appreciate it.

I have updated to the latest OPM release of Flow 2018.04 and tested again on 
RedHat 7. I still don't see any performance improvement with the new RedHat 7 
binaries as if MPI was still not working. On Ubuntu it works fine.

For example I ran the Norne model once on 1 processor and once on 2 processors 
on the RedHat 7 machine and the Ubuntu machine (identical machines on Amazon 
AWS m4.2xlarge instances) and these are the timings I get:


Norne on RedHat:
1 processor:  769.437 seconds
2 processors: 839.285 seconds

Norne on Ubuntu:
1 processor:  889.158 seconds
2 processors: 585 seconds


If I do "top" command on RedHat while job is launched on 2 processors, I can 
see 2 processors running. However it doesn't seem that they are running in 
parallel, it seems as if each process is running a separate serial copy of 
flow, hence the run on 2 processors is even slower than the serial run.

Also a quick check using "ldd" on the flow binaries on both Ubuntu and RedHat 
shows that the Ubuntu executable is indeed linked to the MPI library, while the 
RedHat binary still doesn't show any linked MPI libraries:


On Ubuntu:

ubuntu@ip-172-31-15-200:~$ ldd /usr/bin/flow | grep -i mpi
libmpi_cxx.so.1 => /usr/lib/libmpi_cxx.so.1 (0x7f350423c000)
libmpi.so.12 => /usr/lib/libmpi.so.12 (0x7f3503f66000)


On RedHat:

[ec2-user@ip-172-31-15-201 ~]$ ldd /usr/bin/flow | grep mpi
[ec2-user@ip-172-31-15-201 ~]$

[ec2-user@ip-172-31-15-201 ~]$ rpm -qa | grep opm
libopm-common1-2018.04-0.x86_64
opm-upscaling-devel-2018.04-0.x86_64
opm-simulators-bin-2018.04-0.x86_64
libopm-grid1-2018.04-0.x86_64
libopm-upscaling1-2018.04-0.x86_64
libopm-simulators1-2018.04-0.x86_64
opm-upscaling-2018.04-0.x86_64


I would appreciate any help with this please to get the parallel version of 
Flow working on RedHat.

Thank you for all of your help.

Sincerely,

Mohamad


On Thu, Apr 26, 2018 at 2:26 AM, Arne Morten Kvarving 
<arne.morten.kvarv...@sintef.no<mailto:arne.morten.kvarv...@sintef.no>> wrote:

hi,


currently the rpm's are not mpi enabled. this will change in the upcoming 
2018.04 release.


it's a bit involved building on rhel, as you need some packages not in base or 
epel. in particular, you need to trilinos (or rather, just zoltan which is part 
of trilinos) to get efficient mpi support, as well as dune and such. if it's 
not extremely pressing i would suggest waiting for the release.


arnem


Fra: Opm <opm-boun...@opm-project.org<mailto:opm-boun...@opm-project.org>> på 
vegne av M. S. <sind...@gmail.com<mailto:sind...@gmail.com>>
Sendt: 25. april 2018 22:46:33
Til: opm@opm-project.org<mailto:opm@opm-project.org>
Emne: [Opm] OPM Flow RedHat 7 Binary Package Build with MPI Support

Dear All,

I am interested in running OPM Flow with MPI support on RedHat 7.

I've installed the binary packages on 2 different machines, one with RedHat 7 
and the other with Ubuntu 16.04 using these instructions:

https://opm-project.org/?page_id=245

The Ubuntu packages ran fine with MPI and I can see performance improvement 
when running some of the SPE models, however the RedHat 7 version doesn't seem 
to have been built with MPI support.

RedHat 7 currently has native mpi package that can be easily installed by yum 
(based on mpich-3):

sudo yum -y install mpich-devel

Would someone be able please to build the Flow binaries for RedHat 7 with MPI 
support, or perhaps if they already exist can you please point me to where to 
get them?

I attempted building from source code but the Flow third party library 
prerequisite installation seems overwhelming.

If anyone already has their RedHat 7 build environment setup and would be able 
to help I would truly appreciate it. From the instructions on the website, it 
says to enable MPI support you just need to pass this option to cmake during 
the build “-DUSE_MPI=1” .

Thank you for all of your help, I really appreciate it.

Sincerely,

Mohamad Sindi
MIT

___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] OPM Flow RedHat 7 Binary Package Build with MPI Support

[Arne Morten Kvarving]

hi,


currently the rpm's are not mpi enabled. this will change in the upcoming 
2018.04 release.


it's a bit involved building on rhel, as you need some packages not in base or 
epel. in particular, you need to trilinos (or rather, just zoltan which is part 
of trilinos) to get efficient mpi support, as well as dune and such. if it's 
not extremely pressing i would suggest waiting for the release.


arnem


Fra: Opm  på vegne av M. S. 
Sendt: 25. april 2018 22:46:33
Til: opm@opm-project.org
Emne: [Opm] OPM Flow RedHat 7 Binary Package Build with MPI Support

Dear All,

I am interested in running OPM Flow with MPI support on RedHat 7.

I've installed the binary packages on 2 different machines, one with RedHat 7 
and the other with Ubuntu 16.04 using these instructions:

https://opm-project.org/?page_id=245

The Ubuntu packages ran fine with MPI and I can see performance improvement 
when running some of the SPE models, however the RedHat 7 version doesn't seem 
to have been built with MPI support.

RedHat 7 currently has native mpi package that can be easily installed by yum 
(based on mpich-3):

sudo yum -y install mpich-devel

Would someone be able please to build the Flow binaries for RedHat 7 with MPI 
support, or perhaps if they already exist can you please point me to where to 
get them?

I attempted building from source code but the Flow third party library 
prerequisite installation seems overwhelming.

If anyone already has their RedHat 7 build environment setup and would be able 
to help I would truly appreciate it. From the instructions on the website, it 
says to enable MPI support you just need to pass this option to cmake during 
the build “-DUSE_MPI=1” .

Thank you for all of your help, I really appreciate it.

Sincerely,

Mohamad Sindi
MIT
___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Fwd: (no subject)

[Arne Morten Kvarving]

it's not as simple,.you need several packages only provided in the opm repos 
for things which are either outdated or not provided by base or epel.


we are making a new release these days, i will update the rhel packages to 
build with mpi. at least for rhel7, i have a feeling rhel6 might want to fight 
me.


arnem


Fra: Opm  på vegne av bacetti abdelmoumen 

Sendt: 16. april 2018 10:11:51
Til: Opm@opm-project.org
Emne: [Opm] Fwd: (no subject)

Is the compiling procedure the same as described for Ubuntu?


Regards/Cordialement

Abdelmoumen Bacetti

abdelmoumen.bace...@sonatrach.dz
bacetti.abdelmou...@gmail.com
Linkedin   &
Researchgate

-- Forwarded message --
From: Alf Birger Rustad >
Date: Sun, Apr 15, 2018 at 10:03 PM
Subject: Re: [Opm] (no subject)
To: Atgeirr Rasmussen 
>, 
"opm@opm-project.org" 
>


I meant to say, Red Hat packages do *not* have MPI support :-)

Fra: Opm > på 
vegne av Alf Birger Rustad >
Sendt: 15. april 2018 22:53:02
Til: Atgeirr Rasmussen; opm@opm-project.org

Emne: Re: [Opm] (no subject)


Please note that the Red Hat packages do have MPI support. You will need to 
compile from source, and to get load balancing, you will need to also compile 
and link in zoltan.


Cheers,

Alf



Fra: Opm > på 
vegne av Atgeirr Rasmussen 
>
Sendt: 15. april 2018 21:39
Til: opm@opm-project.org
Emne: Re: [Opm] (no subject)


Dear Abdelmoumen,


Installing and running the RHEL binary packages could work, but there could be 
things that make your HPC system different from a standard RHEL system, causing 
incompatibilities. You could try, and report how it goes?


Atgeirr



Frå: Opm > på 
vegne av bacetti abdelmoumen 
>
Sendt: fredag 13. april 2018 16:20
Til: opm@opm-project.org
Emne: Re: [Opm] (no subject)

Dear Mr. Atgeirr,

Regarding what you have said on Ubuntu binary, is it possible to the same on 
Redhat (which the system used on the HPC)?

Abdelmoumen



Regards/Cordialement

Abdelmoumen Bacetti

abdelmoumen.bace...@sonatrach.dz
bacetti.abdelmou...@gmail.com
Linkedin   &
Researchgate

On Fri, Apr 13, 2018 at 8:24 AM, Atgeirr Rasmussen 
> wrote:

Dear Mr. Abdelmoumen,


I recommend that you try running OPM Flow on HPC systems using the Singularity 
container system. You should ask your HPC maintainer if Singularity is 
supported. If not, Docker containers are also available, but will probably not 
be able to exploit the special hardware typically present in an HPC system.


Also, if you can run a Ubuntu binary on the system, you can install the binary 
packages for Ubuntu 16.04 as explained on the OPM web site:


https://opm-project.org/?page_id=36


All of the preceding options are compiled with MPI, meaning that the Flow 
executable can be run with mpirun in parallel.


Finally, there is a new release underway, before the end of the month a new 
version is available.


Atgeirr



Frå: Opm > på 
vegne av bacetti abdelmoumen 
>
Sendt: 12. april 2018 17:07
Til: opm@opm-project.org
Emne: [Opm] (no subject)

Willing to run OPM on the university HPC, is there any parallel version of 
OPM's binary or does the HPC admin have to compile it from source for that.

If it is required to compile the source to get a parallel version, what are the 
compiler options?

___
Opm mailing list
Opm@opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm




---
The information contained in this 

[Opm] jenkins down time

[Arne Morten Kvarving]

the jenkins server will be taken down for maintenance around 1CET today. we do 
not expect a long down-time.


arnem
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Problem with installing python-ert.ecl

[Arne Morten Kvarving]

happy new year,


i started with cleaning up old sins. package rebuilt.


i didn't recommend force-depends as apt nags on every invocation afterwards 
which is a pain.


Fra: Opm <opm-boun...@opm-project.org> på vegne av Roland Kaufmann 
<roland.kaufm...@uni.no>
Sendt: 23. desember 2017 12:07:28
Til: opm@opm-project.org
Emne: Re: [Opm] Problem with installing python-ert.ecl

On 22. Dec. 2017 at 09:33, Carl Fredrik Berg wrote:
>> I can't install python-ecl again, and get the following
>> error  messages:
>> The following packages have unmet dependencies:

On 22. Dec. 2017 at 11:14, Arne Morten Kvarving wrote:
> for now, all i can offer is to tell you that you can extract
> the deb manually

How about using `dpkg -i --force-depends python-ecl.deb`? At
least, you'll be able to remove the files afterwards using the
package system, when a new fixed package is released?

--
Roland.
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] jenkins downtime

[Arne Morten Kvarving]

jenkins is back better than ever.


- we now have 12 procs and 32GB ram. this translates to 3 concurrent jobs with 
4 build threads.

- a dedicated slave has been added for the post-build jobs. this means that the 
post-builder jobs only run one at at time, and the other 2 slots are reserved 
for pull request builders.


ie. we have higher throughput and higher responsiveness for the PR builder.


yell if anything breaks.


Fra: Opm <opm-boun...@opm-project.org> på vegne av Arne Morten Kvarving 
<arne.morten.kvarv...@sintef.no>
Sendt: 15. desember 2017 10:06:50
Til: opm@opm-project.org
Emne: [Opm] jenkins downtime


hi,


jenkins will be taken down at 12 cet for upgrades. shouldn't take too long.


arnem

___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] jenkins downtime

[Arne Morten Kvarving]

hi,


jenkins will be taken down at 12 cet for upgrades. shouldn't take too long.


arnem

___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Problem with installing python-ert.ecl

[Arne Morten Kvarving]

outdated instructions. replace python-ert.ecl with python-ecl


Fra: Opm  på vegne av Чащин Артём 

Sendt: 23. november 2017 16:55:22
Til: opm@opm-project.org
Emne: [Opm] Problem with installing python-ert.ecl

Hello,

I've installed Opm on Ubuntu 16.04 from a binary package and try to run the 
spe1 tutorial from http://opm-project.org/?page_id=197. Everything worked fine 
until I got to the installation of python-ert.ecl on page 6 of the tutorial. 
The command 'sudo apt-get install python-ert.ecl python-numpy 
python-matplotlib' gives me the following output (I omitted several lines):

...
Selecting previously unselected package libert.ecl1.
Preparing to unpack .../libert.ecl1_2017.04-rfinal-1~xenial_amd64.deb ...
Unpacking libert.ecl1 (2017.04-rfinal-1~xenial) ...
dpkg: error processing archive 
/var/cache/apt/archives/libert.ecl1_2017.04-rfinal-1~xenial_amd64.deb 
(--unpack):
trying to overwrite '/usr/share/man/man1/ecl_summary.1.gz', which is also in 
package libecl1 2017.10-rfinal-1~xenial
dpkg-deb: error: subprocess paste was killed by signal (Broken pipe)
...
Selecting previously unselected package python-ert.ecl.
Preparing to unpack .../python-ert.ecl_2017.04-rfinal-1~xenial_amd64.deb ...
Unpacking python-ert.ecl (2017.04-rfinal-1~xenial) ...
...
Errors were encountered while processing:
/var/cache/apt/archives/libert.ecl1_2017.04-rfinal-1~xenial_amd64.deb
E: Sub-process /usr/bin/dpkg returned an error code (1)

After that, any 'apt-get install' command that I run gives me the message:

You might want to run 'apt-get -f install' to correct these:
The following packages have unmet dependencies:
 python-ert.ecl : Depends: libert.ecl1 but it is not going to be installed
E: Unmet dependencies. Try 'apt-get -f install' with no packages (or specify a 
solution).

But doing 'apt-get -f install' leads to the same 'subprocess paste was killed 
by signal (Broken pipe)' error. As a result of all of this, Python can't find 
the module ecl.ecl and I can't run the summaryplot file. Does anyone know 
solution to this problem?

Regards,
Artyom
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] nightly ubuntu packages

[Arne Morten Kvarving]

nightlies for debian stretch has been added. same story as for xenial, just 
replace xenial with stretch. currently there is no docker support, i will look 
into that eventually.


Fra: Arne Morten Kvarving
Sendt: 27. september 2017 08:36:01
Til: opm@opm-project.org
Emne: nightly ubuntu packages


hi,


nightly packages for ubuntu xenial is now operational. note that they 
temporarily run against my fork of libecl while waiting for my PRs to be 
processed.


install the apt repo using


1) install key: wget http://opm-project.org/package/nightly-xenial/repokey.gpg 
| sudo apt-key add -

2) add repo sudo apt-add-repository 
http://opm-project.org/package/nightly-xenial/

3) sudo apt-get update

4) apt-cache show libopm-simulators1-bin


the repo contains the nightly build for the last seven days, the last four 
sundays and the first of the last five months. obviously not that populated 
yet, but that's the scheme.


in case you do not know how to operate apt when there are several versions of a 
package and you want to choose a particular one, i point you to 
https://github.com/OPM/opm-utilities/blob/master/docker_opm_user/Dockerfile#L42

[https://avatars3.githubusercontent.com/u/1827197?v=4=400]<https://github.com/OPM/opm-utilities/blob/master/docker_opm_user/Dockerfile#L42>

OPM/opm-utilities<https://github.com/OPM/opm-utilities/blob/master/docker_opm_user/Dockerfile#L42>
github.com
opm-utilities - A collection of utilities of interest to the opm community

here opm_version is a date, e.g. 2017-09-25


sadly you have to specify for all packages, not just the top one, as apt tries 
to pull the newest ofupstreams, and that naturally lead to errors.


arnem

___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] 2017.10 rc1

[Arne Morten Kvarving]

Packages for ubuntu xenial (16.04 ) are now available in the testing PPA.


Branches and tags have been published. Have a small backport PR for 
opm-material (packaging files), will get to that tmrw.


arnem

___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Prerequisites for building OPM page ( http://opm-project.org/?page_id=239 )

[Arne Morten Kvarving]

neither. libblas-c-dev is gone in newer ubuntus, it's enough with libblas-dev.

liblas-c-dev is something completely different.


Fra: Opm  på vegne av Rodrigo Piccinini 

Sendt: 10. oktober 2017 12:10:19
Til: opm@opm-project.org
Emne: [Opm] Prerequisites for building OPM page ( 
http://opm-project.org/?page_id=239 )


Hello,

In the Ubuntu section of prerequisites page for building OPM ( 
http://opm-project.org/?page_id=239 ), there is mention to a package named
libblas-c-dev.
It probably should read
liblas-c-dev
(without doubling the letter b), as it is seen in Ubuntu package list.

https://packages.ubuntu.com/search?suite=zesty=all=any=liblas-c-dev=names

Thanks

--
Rodrigo
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] CI system downtime

[Arne Morten Kvarving]

murphy was surprisingly kind.


ci systems now run on debian 9.1, most noteworthy changes are


- jenkins 1.651 -> 2.73.1

- gcc 4.9.2 -> 6.3

- boost 1.55 -> 1.62

- openmpi 1.8 -> 2.0

- cmake 3.6.1 -> 3.7.2

- dune 2.4.1 -> 2.5.0. this is a little unfortunate since 2.4.1 is considered 
the minimum version and we do not get to test things. so i might have to do a 
backport here.


plus obviously a bunch of other updates, but these should be user-facing ones.


happy to report that there is not a lot of fallout.

- there was some trivial build errors in ResInsight (PR submitted).

- some issues in opm-upscaling and opm-simulators wrt umfpack which i took the 
liberty to take care of directly

- an issue in opm-upscaling where gcc6 didn't like implicit casts (taken care 
of)

- opm-grid's 'grid' test fails, likely due to dune-grid 2.5.

- autodiffmatrix test in opm-simulators fails

- some regression test fails in opm-simulators. fairly certain these are 
harmless, will investigate tmrw


i have not tested the PR builders so that could be where murphy decided to 
bite. we will know soon enough i guess, but weekend here i come!


arnem



Fra: Opm <opm-boun...@opm-project.org> på vegne av Arne Morten Kvarving 
<arne.morten.kvarv...@sintef.no>
Sendt: 29. september 2017 10:25:01
Til: opm@opm-project.org
Emne: [Opm] CI system downtime


hi,


at 16:00 CET today i will take down jenkins and to do a debian distribution 
update. depending on how grumpy mr murphy is, it will be down anywhere from 3h 
to 3 days. hopefully closer to the first.


arnem
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] CI system downtime

[Arne Morten Kvarving]

hi,


at 16:00 CET today i will take down jenkins and to do a debian distribution 
update. depending on how grumpy mr murphy is, it will be down anywhere from 3h 
to 3 days. hopefully closer to the first.


arnem
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] jenkins downtime

[Arne Morten Kvarving]

jenkins is back and open for business.


all builds are now dune 2.4.1 based. it fixed the tests that somebody broke 
during my absence, ref


https://ci.opm-project.org/job/opm-simulators/366/


and


https://ci.opm-project.org/job/opm-simulators/367


i suspect the breakage was just a timing thing though, as i could never 
reproduce it locally, neither with 2.3.1 (or 2.4.1).


the norne well curves seems fine to me, but i'm not very experienced in reading 
these graphs - e.g. how far from the reference are we these days etc. so if 
somebody that has looked at more of those than me could give it a looksie i 
would appreciate that.


strictly speaking the ewoms builder is not completely done but i'm 99.9% 
certain (famous last words) it will pass fine, and in case of breakage it will 
only be some test application which do not affect other modules.


arnem


Fra: Opm <opm-boun...@opm-project.org> på vegne av Arne Morten Kvarving 
<arne.morten.kvarv...@sintef.no>
Sendt: 10. juli 2017 13:54:32
Til: opm@opm-project.org
Emne: [Opm] jenkins downtime


hi,


seems there is not a lot of activity, but in any case:


i will have to shut jenkins down for upgrades (dune 2.4.1). i do not anticipate 
a long downtime. this will take place at 14:30 cet. please do not flood it with 
jobs as i will have to wait for those to finish before the shutdown commences.


arnem

___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] jenkins downtime

[Arne Morten Kvarving]

hi,


seems there is not a lot of activity, but in any case:


i will have to shut jenkins down for upgrades (dune 2.4.1). i do not anticipate 
a long downtime. this will take place at 14:30 cet. please do not flood it with 
jobs as i will have to wait for those to finish before the shutdown commences.


arnem

___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] ResInsight available in ppa

[Arne Morten Kvarving]

ResInsight 2016.11 is now available in the opm ppa.


It's basically packaged straight out of the resinsight sources, with some minor 
patchery to properly treat the internal libraries as such (static links so the 
unix linker doesn't get confused) as well as necessary changes to the packaging 
files.


sudo apt-get install resinsight


and optionally


sudo apt-get install octave-resinsight for the octave plugins.


arnem
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] The 2017.04 release branches have been created

[Arne Morten Kvarving]

The testing ppa is now populated with rc1 packages for ubuntu xenial. rc1 has 
been tagged in all repos, and backports of updates have been submitted.


i didn't get explicit instructions on whether to build for trusty or not, so i 
took the "lazy" decision. if you want those packages as well, just say so.


TLDR;lock on release branches lifted. commence backport activities.


Fra: Opm <opm-boun...@opm-project.org> på vegne av Andreas Lauser 
<a...@poware.org>
Sendt: 13. april 2017 13:15:13
Til: opm@opm-project.org
Emne: [Opm] The 2017.04 release branches have been created

Hi,

I have just created and uploaded the stable branches of OPM 2017.04 for all
modules featured by this release. They are called "release/2017.04" and are
fairly close to what the final release will look like, i.e., these branches are
supposed to only get fixes for serious bugs and packaging updates. For this
reason, Arne Morten Kvarving is the person in charge of them. That said,
maintainers are responsible to apply all bugfixes from the master branch to the
release branch and vice versa until the final version have been made; i.e., if
maintainers decide to merge disruptive changes to master, that's okay with me,
but I'll grill them if they do not fix corresponding bugs in the release branch
via minimal patches, so I advice to be a bit conservative with pressing The
Green Button (TM) during the next two weeks or so.

cheers
  Andreas

--
Just because a patch doesn't solve world hunger isn't really a good
reason to reject it.
-- Daniel Vetter
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Are build tests supposed to work again?

[Arne Morten Kvarving]

CI is back in working order.


on jenkins PR's are always built merged into the master of the repo, so as long 
as they merge fine, you do not have to update.


Fra: Opm  på vegne av Markus Blatt 

Sendt: 3. april 2017 13:44:37
Til: opm@opm-project.org
Emne: [Opm] Are build tests supposed to work again?

Hi,

Quick question to not waste CPU cycles for no use:

I have some PRs hanging around which had weired failures (no bc,
regression when comparing results). Is the automated test system
supposed to work again? If it is, do I need to rebase the PRs onto the
current master branch?

Markus
--
Dr. Markus Blatt - HPC-Simulation-Software & Services http://www.dr-blatt.de
Hans-Bunte-Str. 8-10, 69123 Heidelberg, Germany,  USt-Id: DE279960836
Tel.: +49 (0) 160 97590858
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] Travis trouble

[Arne Morten Kvarving]

On 14. nov. 2016 16:01, Joakim Hove wrote:


Hello,

at the moment all Travis builds fail due to timeout. According to 
google Travis has a 50 minutes time limit, it is my impression that 
they have "suddenly?" started to enforce that limit.


as a short-term fix i suggest not passing -DBUILD_TESTING to ewoms. it 
will cut down the build time considerably (at the obvious cost of not 
executing ewoms tests).


arnem
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] jenkins downtime

[Arne Morten Kvarving]

jenkins is back. all previously queued builds are now executing.

arnem
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [Opm] jenkins downtime

[Arne Morten Kvarving]

On 27. juni 2016 15:05, Bård Skaflestad wrote:

Just out of curiosity: If there are any builds running at that time, will they 
be allowed to complete or terminated?



running jobs will be completed.

(sorry for the dupe bård, i forgot to reply to the list..)
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

[Opm] jenkins downtime

[Arne Morten Kvarving]

hi,

i will have to take jenkins down for a while at the end of the day 
(1630-ish CET).


i need to build trilinos (for zoltan) and memory is exhausted with jobs 
running in parallel.

hopefully the downtime won't last long.

arnem
___
Opm mailing list
Opm@opm-project.org
http://opm-project.org/cgi-bin/mailman/listinfo/opm

Re: [OPM] Refactoring of the fully implicit solver class

[Arne Morten Kvarving]

On 22/05/15 15:43, Jørgen Kvalsvik wrote:

On 05/22/2015 01:09 PM, Atgeirr Rasmussen wrote:

Dear OPM community,

We have been considering how to best refactor the (huge) class 
FullyImplicitBlackoilSolver
in such a manner that it can be more easily be extended with new options and 
functionality
without copying the whole class (as is currently done to implement the 
flow_polymer variant
in opm-polymer).

The goal is simple enough: avoid copying of code between the black-oil and 
black-oil-with-polymer
cases, which also will make the code a better base for further extensions. A 
first step has been taken:
separating the NewtonSolver functionality into its own class, leaving the rest 
of the functionality in the
BlackoilModel class.

Further refactoring of this is faced with some nontrivial technical challenges, 
centering on the
State classes. For example, many functions take a State object argument, which 
(in the current approach)
are different between models (black-oil and b.o. with polymer for example). 
Then that function
cannot be virtual (it would need to take the same argument).

There seems to be two main alternatives:


1. Applying the curiously recurring template pattern (it is indeed called 
that!). This means
essentially using templates to get what we want, including the ability to 
extend models,
reusing most functions and only modifying the few that we want to be different. 
This will
all be checked at compile time, so illegal combinations of objects (such as 
accidentally
using a State class without a certain member together with a model that 
requires that member)
will not even compile. The drawback is that it involves a little more 
boilerplate code than regular
inheritance: assuming that we want to an extensible black-oil model and a 
polymer model that
extends it we'll have three classes, like this:

template class Implementation
class BlackoilModelBase
{
// This is where almost all the functions and data go.
};

class BlackoilModel : public BlackoilModelBaseBlackoilModel
{
// This contains only type definitions and a constructor.
};

class BlackoilPolymerModel: public BlackoilModelBaseBlackoilPolymerModel
{
// Contains type definitions and a constructor.
// Contains new functions, as well as new versions of functions from 
BlackoilModelBase.
};

I do think that the boilerplate is manageable, and that people will get by just 
fine by starting their own
extension by copying an existing extension. (The resulting copied code will not 
be that many lines).
This pattern is a fairly common C++ idiom and is used a lot in Dune for example.


2. Using regular inheritance and a generic State class. This means we have a 
single, flexible State
class that can be used for all purposes. It should function a bit like a mapstring, 
vectordouble,
but not be implemented like that. This solution is most similar to how things 
would be done in Matlab,
or other weakly typed languages: any usage errors would not be caught until 
runtime. Each access
to the State will be a little more annoying than before, instead of 
state.saturation() you would need to
call state.get(saturation) and it could possibly fail.

In summary, none of these solutions are perfect, but both are workable. I have 
discussed this with
Andreas, and we agreed to propose alternative 1 as the one we'd like to try out 
first. I would like to
hear your opinions!

Atgeirr

The third option, which is somewhat similar to the first option (which
by the way is much better than the broken common State class idiom) is
to properly split things up into values and transactions and abandon the
idea of a full class that is initialised, then called solve() upon.

This means more flexibility in terms of replacing, plugging in and
extending individual parts, as functions could be either arguments,
static (aka hard coded) or compile-time arguments (hello templates),
with the added benefit that you can replace sub components as containers
with relative ease.

Additionally, this actually opens for concurrency, something we should
strive for in design considering the state we're in now.

And if someone should still prefer having a class manage everything with
init = solve, then this class would now be trivial to write.
this is a good idea, although i dunno what would be involved here in 
achieving it.


as an example, somewhat on a sidetrack but to show how such an approach 
can be used in codes. it's a bit hard to make this understandable in 
email form, but


consider this template definition for a RANS driver:

/*!
  \brief Driver class for RANS simulators.
  \details A RANS simulator is a coupling between a fluid solver
  and a viscosity solver. The coupling may be segregated or semi-implicit.
*/
templateclass FluidSolver, class ViscSolver,
 templateclass T1, class T2 class Coupling=SIMCoupled
class SIMRANS : public CouplingFluidSolver,ViscSolver

a RANS coupling is a general pattern; you have some fluid 
(navier-stokes) solver which provides a 

Re: [OPM] Iterative solvers for autodiff

[Arne Morten Kvarving]

On 20/05/15 14:02, Markus Blatt wrote:

Like I said I am biased as a DUNE developer (parallel iterative
solvers). But if you do not want to use DUNE, then I would rather
recommend PETSc, Trilinos or you will find others that might be more
suitable from Jack Dongarra's list
http://www.netlib.org/utk/people/JackDongarra/la-sw.html
as somewhat less biased input; i'm in 100% agreement with markus, eigen 
for sparse linear algebra is just for toy problems.


i never tried dune in parallel, but my serial experience has proven it 
to be very efficient. i can also vouch for petsc if you can stand its .. 
old style api. it's very good at what it does and very flexible, but not 
necessarily the fastest (in terms of constants..)


arnem

___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm

[OPM] release candidate packages in PPA

[Arne Morten Kvarving]

index 11c5e49..84462d0 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,9 @@
+opm-core (2015.04-rc1-2~precise) precise; urgency=medium
+
+  * Release new version
+
+ -- Arne Morten Kvarving arne.morten.kvarv...@sintef.no  Mon, 13 Apr 2015 13:36:51 +0200
+
 opm-core (2013.10-2~precise) precise; urgency=low
 
   * New release
diff --git a/CMakeLists_files.cmake b/CMakeLists_files.cmake
index 4b70ca1..c37fa18 100644
--- a/CMakeLists_files.cmake
+++ b/CMakeLists_files.cmake
@@ -55,12 +55,12 @@ list (APPEND ATTIC_FILES
 # originally generated with the command:
 # find tests -name '*.cpp' -a ! -wholename '*/not-unit/*' -printf '\t%p\n' | sort
 list (APPEND TEST_SOURCE_FILES
-  tests/cpgrid/distribution_test.cpp
+	# tests/cpgrid/distribution_test.cpp
 	tests/cpgrid/entityrep_test.cpp
-	tests/cpgrid/entity_test.cpp
+	#	tests/cpgrid/entity_test.cpp
 	tests/cpgrid/geometry_test.cpp
 	tests/cpgrid/orientedentitytable_test.cpp
-	tests/cpgrid/partition_iterator_test.cpp
+	#	tests/cpgrid/partition_iterator_test.cpp
 	)
 
 # originally generated with the command:
diff --git a/CMakeLists_files.cmake b/CMakeLists_files.cmake
index a2b91a7..20fc341 100644
--- a/CMakeLists_files.cmake
+++ b/CMakeLists_files.cmake
@@ -30,12 +30,12 @@
 # originally generated with the command:
 # find tests -name '*.cpp' -a ! -wholename '*/not-unit/*' -printf '\t%p\n' | sort
 list (APPEND TEST_SOURCE_FILES
-	tests/test_fluidmatrixinteractions.cpp
+	#	tests/test_fluidmatrixinteractions.cpp
 	tests/test_pengrobinson.cpp
 	tests/test_ncpflash.cpp
 	tests/test_tabulation.cpp
 	tests/test_2dtables.cpp
-	tests/test_fluidsystems.cpp
+	#	tests/test_fluidsystems.cpp
 	tests/test_immiscibleflash.cpp
 	)
 
diff --git a/debian/changelog b/debian/changelog
index 1827228..556583d 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,9 @@
+opm-material (2015.04-rc1-2~trusty) trusty; urgency=medium
+
+  * Release new version
+
+ -- Arne Morten Kvarving arne.morten.kvarv...@sintef.no  Mon, 13 Apr 2015 19:55:49 +0200
+
 opm-material (2013.10-2~precise) precise; urgency=low
 
   * Initial release
diff --git a/debian/control b/debian/control
index 5a2ca42..948baa8 100644
--- a/debian/control
+++ b/debian/control
@@ -7,7 +7,7 @@ Build-Depends: build-essential, debhelper (= 9), libboost-filesystem-dev,
libdune-common-dev, libdune-istl-dev, cmake, libtinyxml-dev, bc,
libert.ecl-dev, git, zlib1g-dev, libtool, doxygen, libopm-core-dev,
texlive-latex-extra, texlive-latex-recommended, ghostscript,
-   libopm-parser-dev, libboost-iostreams-dev
+   libopm-parser-dev, libboost-iostreams-dev, mpi-default-bin, mpi-default-dev, petsc-dev
 Standards-Version: 3.9.2
 Section: libs
 Homepage: http://opm-project.org
diff --git a/debian/rules b/debian/rules
index 6a1d1fa..611928e 100755
--- a/debian/rules
+++ b/debian/rules
@@ -20,7 +20,7 @@ override_dh_auto_build:
 
 # consider using -DUSE_VERSIONED_DIR=ON if backporting
 override_dh_auto_configure:
-	dh_auto_configure --buildsystem=cmake -- -DCMAKE_BUILD_TYPE=RelWithDebInfo -DSTRIP_DEBUGGING_SYMBOLS=ON -DBUILD_SHARED_LIBS=1 -DCMAKE_INSTALL_DOCDIR=share/doc/libopm-material1 -DUSE_RUNPATH=OFF
+	dh_auto_configure --buildsystem=cmake -- -DCMAKE_BUILD_TYPE=RelWithDebInfo -DSTRIP_DEBUGGING_SYMBOLS=ON -DBUILD_SHARED_LIBS=1 -DCMAKE_INSTALL_DOCDIR=share/doc/libopm-material1 -DUSE_RUNPATH=OFF -DUSE_MPI=1
 
 override_dh_auto_install:
 	dh_auto_install -- install-html
diff --git a/CMakeLists.txt b/CMakeLists.txt
index d5a7b2f..54d7429 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -138,4 +138,6 @@ add_custom_target(check-commits
 	-P ${PROJECT_SOURCE_DIR}/cmake/Scripts/CheckCommits.cmake)
 
 # setup extra tests (using helper binaries)
-include (${CMAKE_CURRENT_SOURCE_DIR}/TestUpscalingBinaries.cmake)
+if(NOT USE_MPI)
+  include (${CMAKE_CURRENT_SOURCE_DIR}/TestUpscalingBinaries.cmake)
+endif()
diff --git a/CMakeLists_files.cmake b/CMakeLists_files.cmake
index 990d773..0d5f771 100644
--- a/CMakeLists_files.cmake
+++ b/CMakeLists_files.cmake
@@ -50,7 +50,7 @@ list (APPEND TEST_SOURCE_FILES
 	tests/test_span.cpp
 	tests/test_syntax.cpp
 	tests/test_scalar_mult.cpp
-	tests/test_transmissibilitymultipliers.cpp
+	#	tests/test_transmissibilitymultipliers.cpp
 	tests/test_welldensitysegmented.cpp
 	)
 
___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm

Re: [OPM] Support for more linear solvers: status

[Arne Morten Kvarving]

did you knoll? does istl start out with knoll as well?


I'm afraid I don't know what knoll is, so I wouldn't know.


sorry for being cryptic, figured you'd run into it by now :)

http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/KSP/KSPSetInitialGuessKnoll.html 
- it's one way of obtaining an initial guess that is nonzero.


___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm

Re: [OPM] Making OPM modular: support for more linear solvers

[Arne Morten Kvarving]

On 28/10/14 13:19, Atgeirr Rasmussen wrote:

hi,

my .02c;


However, I am afraid that it
may end up as Yet Another Sparse Matrix and only add to any complexity or 
confusion there already is. Abandoning the Eigen
SparseMatrix for opm-autodiff will require a substantial effort, that I do not 
think we'll make (at this point at least), so I think that will remain.
this. no matter which is chosen, conversions will be the consequence and 
these are nontrivial for large systems. having the flexibility to use 
the native type of whatever underlying solver library used should be the 
goal. that is, a thin wrapper on top of native matrix types. istl got 
this right with its LinearOperators (sadly, but obviously, only as long 
as no algebraic preconditioners are used).


arnem

___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm

Re: [OPM] FW: Boost.Regex and libstc++ debug mode

[Arne Morten Kvarving]

On 08/07/14 10:46, Andreas Lauser wrote:

Hi,

On Tuesday 08 July 2014 09:57:42 Roland Kaufmann wrote:

On 2014-07-07 16:24, Joakim Hove wrote:

libstdc++ comes in a DEBUG version which is linked in when
–DGLIBCXX_DEBUG is used.

On 2014-07-07 16:33, Andreas Lauser wrote:

as a work-around for newish (= GCC 4.9 ?) compilers, we could modify
the build system to check if std::regex is supported and if yes, use
that.

On 2014-07-07 17:14, Bård Skaflestad wrote:

Just for the record, std::regex appears to be available on GCC 4.7
too. I don't know how complete that support is, though.

std::regex is in GCC 4.6.3, too, sort of.

sort of, right ;)

We now get a test failure in opm-parser because of this on Arne's machine,
which I suspect uses GCC 4.7 (right, Arne?):

nope; gcc (Ubuntu 4.8.2-19ubuntu1) 4.8.2

will look into it.

___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm

Re: [OPM] issue with compiling opm-core

[Arne Morten Kvarving]

On 27/06/14 10:20, Andreas Lauser wrote:

Hi Paolo,

On Friday 27 June 2014 09:11:28 Paolo Orsini wrote:

If this is the case, wouldn't it make sense to copy the config.h to the
system include folder during the installation?

not really. That's because external projects may need their own configuration.
What it boils down to is that in the end external projects either need to
either use the same build system as OPM or put a lot of effort into making its
own one compatible. (N.B. that the OPM build system is in the same situation
when it comes to its DUNE requisites. We decided to put a lot of effort into
our build system so that it does not screw up Dune.)

cheers
   Andreas


if this is the case, opm is one sick puppy.

___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm

Re: [OPM] issue with compiling opm-core

[Arne Morten Kvarving]

On 26/06/14 14:55, Paolo Orsini wrote:

Hi Arne,

I looked at and tested the CmakeList you added in the tutorial folder 
of opm-core.


It looks a great example on how to call the opm library from external 
software.


The only limitation seems that one still needs to copy the config.h 
file to the local build folder.
What would be the a work around that? Installing the config.h file as 
well during the opm-core build/install?

Not sure...

if an application uses config.h unconditionally, that is a bug. all 
datas from config.h needed in app space should be in 
opm-core_DEFINITIONS. ie. config.h should only be used in the source tree.

i fixed this up for the tutorials at the same time i added the build system.

arnem
___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm

Re: [OPM] issue with compiling opm-core

[Arne Morten Kvarving]

On 26/06/14 15:39, Paolo Orsini wrote:

Sorry for the half finished email


I was using a tutorial1.cpp which was outdated, a flag was added to 
skip the include config.h


I suppose I extend the CmakeList you written for opm-core to add 
opm-parser ..., I mean I can use:

find_package(opm-parser REQUIRED), etc...

since opm-parser is a dependency of opm-core, it's already there. but 
yeah, you can do the same if you want to use opm-parser only.


arnem
___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm

Re: [OPM] issue with compiling opm-core

[Arne Morten Kvarving]

On 24/06/14 17:41, Paolo Orsini wrote:

Thanks for sll your help,

Adding explicitly the path to the libraries as suggested by Andreas I 
was able to compile tutorial 1 externally.


This the build command I used

g++ -std=c++11 -o mytest tutorial1.cpp 
/home/paolo/OPM/opm-core/build/lib/libopmcore.a 
/home/paolo/OPM/opm-parser/build/lib/x86_64-linux-gnu/libParser.a 
/home/paolo/OPM/ert/ert/devel/lib/x86_64-linux-gnu/libecl.so


However when running mytest I got the following error:

 ./mytest: error while loading shared libraries: libecl.so.1.0: cannot 
open shared object file: No such file or directory


Still some issue with linking...

yes, that's what happens when you take bad advice : (joke, andreas!)

you can make it work this way by doing

LD_LIBRARY_PATH=/home/paolo/OPM/ert/ert/devel/lib/x86_64-linux-gnu ./mytest

you are linking to a shared library outside system paths. 
LD_LIBRARY_PATH adds a path for the linker to look in.


never use shared libs from a build tree. either recompile your ert 
static, or install it on the system level and link with -lecl


(sorry for the dupe paolo, forgot to cc the list..)
___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm

Re: [OPM] issue with compiling opm-core

[Arne Morten Kvarving]

On 23/06/14 09:06, Alf Birger Rustad wrote:


I think we all agree that OPM is (and in my opinions should remain) 
more on libraries and core functionality, and less on end user 
applications.


I hope we can cater to both, and today we really do. I believe 
Resinsight and the upscaling codes are firmly positioned in the 
end-user application category. For the reservoir simulation part it is 
different, and I do believe you are right, we should design for making 
it easy to use as libraries and core functionality. At the same time, 
through the file format, we currently work hard at making an end-user 
reservoir simulator. Actually, I believe the latter is necessary to 
enable the former. That is, I am not sure we can rely on others 
providing the end user application part.


To be honest I think I would have had to resort to Andreas suggestion 
myself -- is this the time when I *really* should understand how 
pkg-config works?



you don't have to use pkgconfig if you use cmake. all opm libraries 
installs cmake config files. all you have to do is add a 
'findpackage(xxx)' in your cmakelists, and you're off. no need for a 
find rule, nothing.


pkg-config is only necessary for autotools users and/or when compiling 
manually (nobody should do that..)


it's not hard to use.

g++ -c -o foo.o foo.cpp `pkg-config --cflags dep`
g++ -o foo foo.o `pkg-config --libs dep`

or, one shot usage for the tutorial;

g++ -std=c++11 -o test `pkg-config --cflags opm-core` tutorial1.cpp 
`pkg-config --libs opm-core`


note that the opm tutorials are semi-broken. they unconditionally 
include config.h, a file which does not exist for installations (nor 
should it..)


arnem
___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm

Re: [OPM] issue with compiling opm-core

[Arne Morten Kvarving]

hi paolo,

if you refetch opm-core we have now added a cmake build system for the 
tutorials.
it's independent of the main build system, so you have to do the make 
install cycle on opm-core with dependencies.


you can then just cmake up opm-core/tutorials, build and play.

cheers

arnem

___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm

Re: [OPM] Umfpack runtime dependendy is problematic

[Arne Morten Kvarving]

If you pass -Dumfpack_LIBRARY=/foo/UMFPACK/Lib/libumfpack.a I think it should 
work as intended; otherwise it is a regular bug issue.

That will make our build script slightly more involved, but should work. It 
will however not work for the next user who wants to kick the tires of OPM (in 
the sense that the binaries will fail by default until the user figures out a 
solution like this one).


I think Arne Morten needs it to do multi-threading (or rather, that UMFPACK is 
the alternative that can co-existing with reasonable multi-threading).

Arne Morten, maybe you can chime in here. Is this a stop gap solution, or a 
more permanent situation? Thanks clearing up my misunderstanding of the AMD lib 
BTW, I just assumed it was related to CPU optimizations. Please consider being 
more verbose too ;)
i consider anything that does not follow standardized procedures to be a 
stop gap solution, cause it will almost certainly break sooner or later. 
you refuse (are unable to) follow standardized procedures, and thus 
you will have to deal with breakage. this sounds like a decent workable 
hack.


arnem

___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm

Re: [OPM] Build opm-porsol with opm-core installed somewhere

[Arne Morten Kvarving]

On 07/25/2013 01:05 PM, Joakim Hove wrote:


Hello;

A bit naïve -- but anyway: I have done some modifications to opm-core, 
built and installed the changes[1]. Then I want to build an updated 
version of opm-porsol which makes use of my updated opm-core. Now I 
wonder how to tell opm-porsol where to find the custom version of 
opm-core?




easiest is through sibling directories.

e.g.

/foo

clone as
/foo/opm-core
/foo/omp-porsol

if you configure on top (no build sub dirs - bad roland! ;D) opm-porsol 
will pick up the opm-core, if you have built it first.


arnem
___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm

Re: [OPM] OPM build

[Arne Morten Kvarving]

On 07/25/2013 02:56 PM, Júlio Hoffimann wrote:
To give you more details, this is the linker command line I get by 
running make in verbose mode:


Linking CXX executable bin/sim_poly2p_incomp_reorder
/usr/bin/cmake -E cmake_link_script 
CMakeFiles/sim_poly2p_incomp_reorder.dir/link.txt --verbose=1
/usr/bin/c++   -pipe -Wall -std=c++11 -ggdb3 -O3 -DNDEBUG -flto 
-mtune=native  -Wl,--as-needed 
CMakeFiles/sim_poly2p_incomp_reorder.dir/examples/sim_poly2p_incomp_reorder.cpp.o 
 -o bin/sim_poly2p_incomp_reorder -rdynamic lib/libopmpolymer.a 
/home/julio/Desktop/opm2/opm-core-build/lib/libopmcore.a 
-lboost_date_time-mt -lboost_filesystem-mt -lboost_system-mt -lumfpack 
-lamd -lm -ldunecommon -llapack -lsuperlu -lblas



this is caused by the decision to enable lto by default.

use cmake -DWHOLE_PROG_OPTIM=0 and off you go.

arnem
___
Opm mailing list
Opm@opm-project.org
http://www.opm-project.org/mailman/listinfo/opm