In the middle of toolshed installations with Ephemeris I get in a particular HPC installation the following error:
Server side:
galaxy.tool_util.deps DEBUG 2020-05-11 05:27:10,561 [p:32473,w:4,m:0] [uWSGIWorker4Core0] Using dependency monocle3-cli version 0.0.7 of type conda
10.25.100.188 - - [11/May/2020:05:18:38 +0100] "POST /api/tool_shed_repositories/new/install_repository_revision?key=ff1fadcdba102650fb96c7f486b45d6b HTTP/1.1" 200 - "-" "python-requests/2.23.0"
Mon May 11 05:27:10 2020 - SIGPIPE: writing to a closed pipe/socket/fd (probably the client disconnected) on request /api/tool_shed_repositories/new/install_repository_revision?key=ff1fadcdba102650fb96c7f486b45d6b (ip 10.25.100.188) !!!
Mon May 11 05:27:10 2020 - uwsgi_response_writev_headers_and_body_do(): Broken pipe [core/writer.c line 306] during POST /api/tool_shed_repositories/new/install_repository_revision?key=ff1fadcdba102650fb96c7f486b45d6b (10.25.100.188)
OSError: write error
[pid: 32473|app: 0|req: 291/1167] 10.25.100.188 () {42 vars in 751 bytes} [Mon May 11 05:18:38 2020] POST /api/tool_shed_repositories/new/install_repository_revision?key=ff1fadcdba102650fb96c7f486b45d6b => generated 0 bytes in 511584 msecs (HTTP/1.1 200) 3 headers in 0 bytes (0 switches on core 0)
at the client side you see:
05:23:39 Traceback (most recent call last):
05:23:39 File "/var/lib/jenkins/miniconda3/envs/galaxy-ephemeris/lib/python3.6/site-packages/ephemeris/shed_tools.py", line 609, in install_repositories
05:23:39 install_repository_revision(repository, self.tsc)
05:23:39 File "/var/lib/jenkins/miniconda3/envs/galaxy-ephemeris/lib/python3.6/site-packages/ephemeris/shed_tools.py", line 428, in install_repository_revision
05:23:39 response = tool_shed_client.install_repository_revision(**repository)
05:23:39 File "/var/lib/jenkins/miniconda3/envs/galaxy-ephemeris/lib/python3.6/site-packages/bioblend/galaxy/toolshed/__init__.py", line 146, in install_repository_revision
05:23:39 return self._post(url=url, payload=payload)
05:23:39 File "/var/lib/jenkins/miniconda3/envs/galaxy-ephemeris/lib/python3.6/site-packages/bioblend/galaxy/client.py", line 152, in _post
05:23:39 files_attached=files_attached)
05:23:39 File "/var/lib/jenkins/miniconda3/envs/galaxy-ephemeris/lib/python3.6/site-packages/bioblend/galaxyclient.py", line 159, in make_post_request
05:23:39 body=r.text, status_code=r.status_code)
05:23:39 bioblend.ConnectionError: Unexpected HTTP status code: 504: <!DOCTYPE html>
Currently my socket setup is as follows:
uwsgi:
buffer-size: 16384
die-on-term: true
enable-threads: true
hook-master-start: unix_signal:2 gracefully_kill_them_all
hook-master-start: unix_signal:15 gracefully_kill_them_all
manage-script-name: false
master: true
module: galaxy.webapps.galaxy.buildapp:uwsgi_app()
offload-threads: 1
processes: 4
py-call-osafterfork: true
pythonpath: lib
socket: /tmp/galaxy_gxa_uwsgi.sock
static-map: /static/style=static/style/blue
static-map: /static=static
static-map: /favicon.ico=static/favicon.ico
threads: 1
thunder-lock: true
virtualenv: .venv
the call for the web handler is:
/*****/server/venv_20.01/bin/uwsgi --virtualenv /*****/server/venv_20.01 --yaml /*****/server/config_20.01/galaxy.yml --logdate --thunder-lock --master --processes 4 --threads 1 --logto /*****/server/galaxy-20.01/galaxy_web_%(host_node_name)s.log --socket /tmp/galaxy_gxa_uwsgi.sock --pythonpath lib --pidfile2 /*****/server/galaxy-20.01/galaxy_web_%(host_node_name)s.pid --stats 127.0.0.1:9191 -b 16384
any suggestions on how I should change the comm between UWSGI and Galaxy? Any other config bit that I should show for helping in diagnosis? nginx doesn't show any errors. Thanks!
In the middle of toolshed installations with Ephemeris I get in a particular HPC installation the following error:
Server side:
at the client side you see:
Currently my socket setup is as follows:
the call for the web handler is:
any suggestions on how I should change the comm between UWSGI and Galaxy? Any other config bit that I should show for helping in diagnosis? nginx doesn't show any errors. Thanks!