Dictionary changed size during iteration

[Muzry]

when I use DLRN to process rpms with concurrent, occur following problem:
Traceback (most recent call last):
File "/usr/lib64/python2.7/threading.py", line 811, in __bootstrap_inner
self.run()
File "/usr/lib64/python2.7/threading.py", line 764, in run
self.__target(*self.__args, **self.__kwargs)
File "/usr/lib64/python2.7/multiprocessing/pool.py", line 342, in _handle_tasks
put(task)
RuntimeError: dictionary changed size during iteration

How to deal with this problem?

[javierpena]

I have tried to reproduce this issue locally, setting workers=3 in projects.ini, but it didn't fail for me. Could you share some details about your environment:

• Comnand-line used to build the packages
• Contents of projects.ini
• Which version of DLRN you are using (commit hash if downloaded from source)

Thanks!

[Muzry]

Thank you for replying this issue. This problem occurred occasionally, our environment flavor is 16 CPU 32G RAM.

This is my environment configs:

• Command-line

dlrn --config projects.ini --info-repo xxxxx --head-only --local

• Contents of projects.ini

[DEFAULT]
...
tags=mitaka
pkginfo_driver=dlrn.drivers.rdoinfo.RdoInfoDriver
rsyncdest=
rsyncport=22
workers=10
gerrit_topic=rdo-FTBFS
...

• version of DLRN

commit is 7a24a33

[javierpena]

I think I have finally found the reason. https://bugs.python.org/issue23051 matches this issue, so the fix should come via a python update.

I'll investigate into this, a fix for CentOS could require some time.

[javierpena]

The Python bug is fixed, so closing.