Missing AZEL column? #134
Comments
|
There error must be occurring in line 69 or 71 of https://github.com/lofar-astron/prefactor/blob/master/scripts/InitSubtract_sort_and_compute.py (which is strange as this script is also used in the normal Intitial-Subtract pipeline). You should be able to get more info about the error if you comment out the try/except block in this script that begins on line 60 (so that the RuntimeError is not caught). |
|
@darafferty Here's the result from commenting out the try/except: I thought that having a newer standalone casa in my path might be the issue, but I got the same error when I took that out. It makes sense that all the .pre-cal.ms files would have the column that's added in line 61 since they've already been partway through the 'original' Init-Subtract pipeline. I also checked for the missing 'AZEL1' column and the .pre-cal.ms files are indeed missing that column. I'm not sure why that is, since they were produced by the target pipeline, so should theoretically have the right columns for Init-Subtract? The columns they do have are:
|
|
@mknapp55 AFAIK the It looks like there is something screwed up with your installation of casacore (or python-casacore). I believe this could happen if you have the libraries of a standalone casa(py) in your |
|
@AHorneffer There's nothing in my I'm using Kern for casacore, pyrap, lofar, etc. It's worked fine in the past. I'm using the PiLL pipeline in this environment with no issues. Any ideas about what conflicts/bad paths I should look for? |
|
If you run the pipeline with debug output, then it will print the paths that are actually used to the logfile. Can you put the full logfile somewhere where we can have a look at it? See also: FAQ: My pipeline crashes and I cannot find the problem here |
|
See attached. |
|
O.K. Can you check which (shared-)libraries You can see that by running P.S. As David already mentioned: the [...] Well, another possibility: do you use dysco compression on the MSs? If so, is the dysco library found when you run the pipeline? |
|
I apparently don't have I don't know why I don't have _tables.so - I'm guessing it's something to do with the Kern installation? I do have Dysco installed, but I don't think I used it in any of the previous steps... |
|
Could the problem be that the |
|
What should that path be set to? Normally, I would check what it is on
CEP3, but that isn't an option right now.
…On Thu, May 18, 2017 at 10:21 AM, David Rafferty ***@***.***> wrote:
Could the problem be that the LD_LIBRARY_PATH is not set? I have the path
to the casacore libraries in mine, but I don't know for sure that it's
needed.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#134 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AEgM5jeDgs3w5hOzIXA7IhoMAlphO5T2ks5r7FPjgaJpZM4Nb9Fh>
.
|
|
@darafferty Good point, that may be a problem. @mknapp55 Do you have |
|
No, those are not set explicitly. I modified the pipeline.cfg that came with my lofar installation (via Kern) ( |
|
@mknapp55 Well, did you try setting |
|
Same result. I've attached the log and the top block of pipeline.cfg is below. |
|
My |
|
I set |
|
@mknapp55 Try setting the If that doesn't help, then search for the |
|
@AHorneffer That library ( Here's the result from |
|
@mknapp55 Well, when you do tl;dr I don't believe that you have a working pyrap but don't have a |
|
@AHorneffer Yes, |
|
@mknapp55 Is that directly loaded from the python scripts? |
|
@AHorneffer I'm not sure what you mean? The script that's throwing the error calls |
|
@mknapp55 Here I get that: That piece of python code tells the python interpreter to load the dynamic library |
|
This directory |
|
bleeping Kern! Do you have something like: |
|
@AHorneffer I do! |
|
I looks like the paths for the libraries that casacore.tables uses are in |
|
@darafferty I added The error is back to the I'm wondering if perhaps the problem is that these MSs went part way through an Init-Subtract run? Maybe I should rerun the target pipeline to get 'fresh' MSs for Init-Subtract?? Update: I tried rerunning the target pipeline, which worked fine. I tried to run the Initial-Subtract-Fast on the output of the target pipeline and it failed with the same error, so the problem does not appear to be with the MSs generated by the target pipeline... |
|
It still looks to me like a problem with your casacore / python-casacore installation. Can you try to run the relevant pieces of code by hand? E.g. something like this: If the |
|
I get the same Runtime Error as in the logs. It's complaining about |
|
Still looks like a problem with the software installation. (Somewhere at the boundary between python code and the compiled C(++) code some module / class isn't found.) I have no more idea how to fix this, but it doesn't look like a bug in prefactor to me. P.S. That below was a copy&paste error of mine. It should be
|
|
I'll open an issue with Kern. |
|
Just encountered this myself. In case it's still an issue for you, you need the libcasa-derivedmscal2 package installed. |
I'm trying to run the Initial-Subtract-Fast parset on data prepared by the calibration and target pipelines. I'm getting the following error:
2017-05-16 03:28:19 ERROR node.ip-100-64-109-221.python_plugin: TableProxy::getColumn: column AZEL1 does not existI previously ran the 'normal' Initial-Subtract on this data without issues (other than the very long runtime).
The text was updated successfully, but these errors were encountered: