Open edwardsnj opened 8 years ago
This was only partially fixed by galaxyproject/galaxy#2928 (the import grp
issue)
We also don't see the pulsar.exe file either. We copied over server.ini and app.yml from pulsar\ to pulsar\venv\Lib\site-packages and were able to get puslar-main.exe to run but we still get the same error as original poster when we run pulsar-check.
Also, when we attempted to run the pulsar-main.exe with -d option we got the following error:
Traceback (most recent call last):
File "c:\python27\Lib\runpy.py", line 174, in _run_module_as_main
"main", fname, loader, pkg_name)
File "c:\python27\Lib\runpy.py", line 72, in _run_code
exec code in run_globals
File "C:\Program Files\pulsar\venv\Scripts\pulsar-main.exemain.py", line 9, in
Tried this again with 0.7.3 and am still receiving the same error as above. Is there a way to get this working on windows server?
I have had more success with downloading the latest tagged source release (0.7.3) zipfile and following the install instructions here. I use the virtualenv version of the install, give the explicit path to my python installation's venv command, and execute pulsar using the run.bat script. I was even able to configure https to work (I created the certificate under Linux and it worked without any additional installation, I think, on Windows). So far, so good.
To test, start pulsar with run.bat, then at another command-line, execute the run_client_tests.py using the .venv\bin\python.exe binary.
I guess the open aspect of this ticket is just making sure that all of the installation instructions work for Windows as well as Linux.
Hope this helps...
@jmchilton I'm out of my depth on this one. Any ideas?
This one's been open for a while, but I'm running into the same issues (No connection could be made because the target machine actively refused it
) trying to get Pulsar running on an EC2 Windows Server 2016 instance. I followed the docs using pip
and made it through pulsar-config
. Are there any updated guides or advice for running Pulsar on Windows?
I'm having trouble getting to square one with a Windows deployment of pulsar. I'm working towards a cloud-based (AWS cloud formation, auto-scaling, ideally) windows instance that can act as a worker for a CloudMan galaxy cluster to carry out mass-spec conversion utilities which must run on windows.
I'm following the published installation docs, using virtualenv and pip. I'm using EPD Python corresponding to 2.7.11 on a clean, new (virtual) 64-bit Windows 7 installation.
Install with pip in the virtual environment seems to go fine. Pulsar-config with no arguments runs without any errors, but app.yaml is essentially empty, with just "---" in it. server.ini is created too - this looks correct, although I'm worried that there are so many things in there I don't plan to use. Should I comment out most of it?
There is no runable pulsar.exe, but I do see pulsar-main.exe and pulsar-run.exe. Pulsar-main.exe doesn't seem to accept the --url argument (referenced later in the docs) so I guess pulsar-run.exe is the Windows executable for pulsar.
The galaxy.utils import is problematic, since my Windows/EPD python has no module "grp" (for Linux groups). I wrapped this in a try/except ImportError block with a pass on exception to proceed.
Both pulsar-run and pulsar-check generate the following error when I run them:
File "c:\users\nathan\pulsar\venv\lib\site-packages\pulsar\client\transport\standard.py", line 49, in execute raise PulsarClientTransportError(transport_message=exc.reason) pulsar.client.exceptions.PulsarClientTransportError: Unknown transport error (transport message: [Errno 10061] No connection could be made because the target machine actively refused it)
Thanks,