Re: [galaxy-dev] data collections - workflow - bug?
Sorry again about this mixup (bringing conversation back on channel). Does this happen everytime you run a workflow with collections? Do you know how many job handlers Galaxy is configured with? -John On Mon, Jan 26, 2015 at 11:24 AM, John Chilton <jmchilton@gmail.com> wrote:
Ugh.... misread sqlalchemy as sqlite. I will work on this.
-John
On Mon, Jan 26, 2015 at 11:16 AM, Torsten Houwaart <houwaart@informatik.uni-freiburg.de> wrote:
Hi John,
i ran this on the Galaxy server in Freiburg: galaxy.uni-freiburg.de Björn (who sits immediately opposite of my desk :) ) told me it's a Postres database and that shouldn't be the problem.
Best, Torsten
On 26.01.2015 16:48, John Chilton wrote:
Hey,
Really intensive database operations including datasest collections but other things too like multi-running tools or workflows over many individual datasets for instance - can very easily overwhelm the default sqlite database. This is frustrating and shouldn't happen - but it does unfortunately. I would recommend using a postgres database when testing out dataset collections. The good news is that it is easier than ever to get a fully fledged production-quality server thanks to Bjoern's docker server (https://github.com/bgruening/docker-galaxy-stable) - it comes bundled with Postgres and Slurm so it should be able to handle the collection operations. If you need to run Galaxy on a non-containerized server (for instance because that is where the software is) more information on setting up Galaxy can be found here https://wiki.galaxyproject.org/Admin/Config/Performance/ProductionServer.
Here is a Trello card to track progress on the database optimization efforts if you are interested https://trello.com/c/UPLsMKQI.
Very sorry.
-John
-John
On Mon, Jan 26, 2015 at 9:35 AM, Torsten Houwaart <houwaart@informatik.uni-freiburg.de> wrote:
Hello Galaxy Devs,
I was using data collections (for the first time) for a new workflow of ours and I ran into this problem. There was no complaint by the workflow-editor and I could start the workflow but then <see below> happened. If you need more information about the workflow or otherwise let me know.
Best, Torsten H.
job traceback: Traceback (most recent call last): File "/usr/local/galaxy/galaxy-dist/lib/galaxy/jobs/runners/__init__.py", line 565, in finish_job job_state.job_wrapper.finish( stdout, stderr, exit_code ) File "/usr/local/galaxy/galaxy-dist/lib/galaxy/jobs/__init__.py", line 1250, in finish self.sa_session.flush() File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/scoping.py", line 114, in do return getattr(self.registry(), name)(*args, **kwargs) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/session.py", line 1718, in flush self._flush(objects) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/session.py", line 1789, in _flush flush_context.execute() File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/unitofwork.py", line 331, in execute rec.execute(self) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/unitofwork.py", line 475, in execute uow File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/persistence.py", line 59, in save_obj mapper, table, update) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/persistence.py", line 485, in _emit_update_statements execute(statement, params) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/base.py", line 1449, in execute params) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/base.py", line 1584, in _execute_clauseelement compiled_sql, distilled_params File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/base.py", line 1698, in _execute_context context) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/base.py", line 1691, in _execute_context context) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/default.py", line 331, in do_execute cursor.execute(statement, parameters) DBAPIError: (TransactionRollbackError) deadlock detected DETAIL: Process 3144 waits for ShareLock on transaction 2517124; blocked by process 3143. Process 3143 waits for ShareLock on transaction 2517123; blocked by process 3144. HINT: See server log for query details. 'UPDATE workflow_invocation SET update_time=%(update_time)s WHERE workflow_invocation.id = %(workflow_invocation_id)s' {'update_time': datetime.datetime(2015, 1, 26, 14, 20, 4, 155440), 'workflow_invocation_id': 5454}
___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: https://lists.galaxyproject.org/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/
I will try to build a smaller testcase that fails and talk about the job handlers with Björn tomorrow. I tried this with quite a big workflow which I built from scratch (in the editor). I'll get back to you. Best, Torsten On 27.01.2015 18:25, John Chilton wrote:
Sorry again about this mixup (bringing conversation back on channel). Does this happen everytime you run a workflow with collections? Do you know how many job handlers Galaxy is configured with?
-John
On Mon, Jan 26, 2015 at 11:24 AM, John Chilton <jmchilton@gmail.com> wrote:
Ugh.... misread sqlalchemy as sqlite. I will work on this.
-John
On Mon, Jan 26, 2015 at 11:16 AM, Torsten Houwaart <houwaart@informatik.uni-freiburg.de> wrote:
Hi John,
i ran this on the Galaxy server in Freiburg: galaxy.uni-freiburg.de Björn (who sits immediately opposite of my desk :) ) told me it's a Postres database and that shouldn't be the problem.
Best, Torsten
On 26.01.2015 16:48, John Chilton wrote:
Hey,
Really intensive database operations including datasest collections but other things too like multi-running tools or workflows over many individual datasets for instance - can very easily overwhelm the default sqlite database. This is frustrating and shouldn't happen - but it does unfortunately. I would recommend using a postgres database when testing out dataset collections. The good news is that it is easier than ever to get a fully fledged production-quality server thanks to Bjoern's docker server (https://github.com/bgruening/docker-galaxy-stable) - it comes bundled with Postgres and Slurm so it should be able to handle the collection operations. If you need to run Galaxy on a non-containerized server (for instance because that is where the software is) more information on setting up Galaxy can be found here https://wiki.galaxyproject.org/Admin/Config/Performance/ProductionServer.
Here is a Trello card to track progress on the database optimization efforts if you are interested https://trello.com/c/UPLsMKQI.
Very sorry.
-John
-John
On Mon, Jan 26, 2015 at 9:35 AM, Torsten Houwaart <houwaart@informatik.uni-freiburg.de> wrote:
Hello Galaxy Devs,
I was using data collections (for the first time) for a new workflow of ours and I ran into this problem. There was no complaint by the workflow-editor and I could start the workflow but then <see below> happened. If you need more information about the workflow or otherwise let me know.
Best, Torsten H.
job traceback: Traceback (most recent call last): File "/usr/local/galaxy/galaxy-dist/lib/galaxy/jobs/runners/__init__.py", line 565, in finish_job job_state.job_wrapper.finish( stdout, stderr, exit_code ) File "/usr/local/galaxy/galaxy-dist/lib/galaxy/jobs/__init__.py", line 1250, in finish self.sa_session.flush() File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/scoping.py", line 114, in do return getattr(self.registry(), name)(*args, **kwargs) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/session.py", line 1718, in flush self._flush(objects) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/session.py", line 1789, in _flush flush_context.execute() File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/unitofwork.py", line 331, in execute rec.execute(self) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/unitofwork.py", line 475, in execute uow File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/persistence.py", line 59, in save_obj mapper, table, update) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/persistence.py", line 485, in _emit_update_statements execute(statement, params) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/base.py", line 1449, in execute params) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/base.py", line 1584, in _execute_clauseelement compiled_sql, distilled_params File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/base.py", line 1698, in _execute_context context) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/base.py", line 1691, in _execute_context context) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/default.py", line 331, in do_execute cursor.execute(statement, parameters) DBAPIError: (TransactionRollbackError) deadlock detected DETAIL: Process 3144 waits for ShareLock on transaction 2517124; blocked by process 3143. Process 3143 waits for ShareLock on transaction 2517123; blocked by process 3144. HINT: See server log for query details. 'UPDATE workflow_invocation SET update_time=%(update_time)s WHERE workflow_invocation.id = %(workflow_invocation_id)s' {'update_time': datetime.datetime(2015, 1, 26, 14, 20, 4, 155440), 'workflow_invocation_id': 5454}
___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: https://lists.galaxyproject.org/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/
Torsten - thanks for reporting this. This bug was reported a few more times - and after several false starts I think we have finally found a fix (with a lot of help from Marius van den Beek, Philip Mabon, Aaron Petkau, and Franklin Bristow - big thanks to each of them). Pull Request with Fix: https://github.com/galaxyproject/galaxy/pull/52 Trello Card Tracking Progress: https://trello.com/c/I0n23JEP Thanks for the patience all, -John On Tue, Jan 27, 2015 at 12:35 PM, Torsten Houwaart <houwaart@informatik.uni-freiburg.de> wrote:
I will try to build a smaller testcase that fails and talk about the job handlers with Björn tomorrow. I tried this with quite a big workflow which I built from scratch (in the editor). I'll get back to you.
Best, Torsten
On 27.01.2015 18:25, John Chilton wrote:
Sorry again about this mixup (bringing conversation back on channel). Does this happen everytime you run a workflow with collections? Do you know how many job handlers Galaxy is configured with?
-John
On Mon, Jan 26, 2015 at 11:24 AM, John Chilton <jmchilton@gmail.com> wrote:
Ugh.... misread sqlalchemy as sqlite. I will work on this.
-John
On Mon, Jan 26, 2015 at 11:16 AM, Torsten Houwaart <houwaart@informatik.uni-freiburg.de> wrote:
Hi John,
i ran this on the Galaxy server in Freiburg: galaxy.uni-freiburg.de Björn (who sits immediately opposite of my desk :) ) told me it's a Postres database and that shouldn't be the problem.
Best, Torsten
On 26.01.2015 16:48, John Chilton wrote:
Hey,
Really intensive database operations including datasest collections but other things too like multi-running tools or workflows over many individual datasets for instance - can very easily overwhelm the default sqlite database. This is frustrating and shouldn't happen - but it does unfortunately. I would recommend using a postgres database when testing out dataset collections. The good news is that it is easier than ever to get a fully fledged production-quality server thanks to Bjoern's docker server (https://github.com/bgruening/docker-galaxy-stable) - it comes bundled with Postgres and Slurm so it should be able to handle the collection operations. If you need to run Galaxy on a non-containerized server (for instance because that is where the software is) more information on setting up Galaxy can be found here
https://wiki.galaxyproject.org/Admin/Config/Performance/ProductionServer.
Here is a Trello card to track progress on the database optimization efforts if you are interested https://trello.com/c/UPLsMKQI.
Very sorry.
-John
-John
On Mon, Jan 26, 2015 at 9:35 AM, Torsten Houwaart <houwaart@informatik.uni-freiburg.de> wrote:
Hello Galaxy Devs,
I was using data collections (for the first time) for a new workflow of ours and I ran into this problem. There was no complaint by the workflow-editor and I could start the workflow but then <see below> happened. If you need more information about the workflow or otherwise let me know.
Best, Torsten H.
job traceback: Traceback (most recent call last): File "/usr/local/galaxy/galaxy-dist/lib/galaxy/jobs/runners/__init__.py", line 565, in finish_job job_state.job_wrapper.finish( stdout, stderr, exit_code ) File "/usr/local/galaxy/galaxy-dist/lib/galaxy/jobs/__init__.py", line 1250, in finish self.sa_session.flush() File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/scoping.py", line 114, in do return getattr(self.registry(), name)(*args, **kwargs) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/session.py", line 1718, in flush self._flush(objects) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/session.py", line 1789, in _flush flush_context.execute() File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/unitofwork.py", line 331, in execute rec.execute(self) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/unitofwork.py", line 475, in execute uow File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/persistence.py", line 59, in save_obj mapper, table, update) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/orm/persistence.py", line 485, in _emit_update_statements execute(statement, params) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/base.py", line 1449, in execute params) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/base.py", line 1584, in _execute_clauseelement compiled_sql, distilled_params File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/base.py", line 1698, in _execute_context context) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/base.py", line 1691, in _execute_context context) File
"/usr/local/galaxy/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs2.egg/sqlalchemy/engine/default.py", line 331, in do_execute cursor.execute(statement, parameters) DBAPIError: (TransactionRollbackError) deadlock detected DETAIL: Process 3144 waits for ShareLock on transaction 2517124; blocked by process 3143. Process 3143 waits for ShareLock on transaction 2517123; blocked by process 3144. HINT: See server log for query details. 'UPDATE workflow_invocation SET update_time=%(update_time)s WHERE workflow_invocation.id = %(workflow_invocation_id)s' {'update_time': datetime.datetime(2015, 1, 26, 14, 20, 4, 155440), 'workflow_invocation_id': 5454}
___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: https://lists.galaxyproject.org/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/
participants (2)
-
John Chilton -
Torsten Houwaart