[Bug 1817376] Re: krb5-admin-server postinst has broken debconf if RUN_KADMIND set in /etc/default/krb5-admin-server
We ran into this doing the upgrade from trusty to xenial. Perhaps address that transition (though trusty is nearing EOL). Maybe just remove the defaults file check all together so that package installs work reliably. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1817376 Title: krb5-admin-server postinst has broken debconf if RUN_KADMIND set in /etc/default/krb5-admin-server To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/krb5/+bug/1817376/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1817376] [NEW] krb5-admin-server postinst has broken debconf if RUN_KADMIND set in /etc/default/krb5-admin-server
Public bug reported: Package installation fails on postinst due to broken debconf when /etc/default/krb5-admin-server sets RUN_KADMIND to a value. This happens because the .config script [0] sets debconf question krb5 -admin-server/kadmind to the defaults file value when set to a value. But the .templates file [1] has no entry for this question resulting in a lookup failure. [0] https://git.launchpad.net/ubuntu/+source/krb5/tree/debian/krb5-admin-server.config#n16 [1] https://git.launchpad.net/ubuntu/+source/krb5/tree/debian/krb5-admin-server.templates This bug seems to go back to at least Xenial (and also affects debian packaging too). We worked around this locally by removing RUN_KADMIND from our local defaults file, but this was an annoying bug to sort out and should probably be fixed. That said I'm not sure if it would be more appropraite to remove the db_set from [0] or add the question back to [1]. ** Affects: krb5 (Ubuntu) Importance: Undecided Status: New ** Summary changed: - krb5-admin-server postinst has broken debconf if RUN_KADMIND set to false in /etc/default/krb5-admin-server + krb5-admin-server postinst has broken debconf if RUN_KADMIND set in /etc/default/krb5-admin-server -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1817376 Title: krb5-admin-server postinst has broken debconf if RUN_KADMIND set in /etc/default/krb5-admin-server To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/krb5/+bug/1817376/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1638982] Re: libvirtd randomly crashes on xenial nodes with "realloc(): invalid next size"
It should be noted that after switching to using Ubuntu Cloud Archive which includes newer libvirt this issue went away in the gate. ** Changed in: openstack-gate Status: Confirmed => Fix Released ** Changed in: openstack-gate Importance: Undecided => Medium ** Changed in: openstack-gate Assignee: (unassigned) => Clark Boylan (cboylan) -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1638982 Title: libvirtd randomly crashes on xenial nodes with "realloc(): invalid next size" To manage notifications about this bug go to: https://bugs.launchpad.net/openstack-gate/+bug/1638982/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1643911] Re: libvirt randomly crashes on xenial nodes with "*** Error in `/usr/sbin/libvirtd': malloc(): memory corruption:"
It should be noted that after switching to using Ubuntu Cloud Archive which includes newer libvirt this issue went away in the gate. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1643911 Title: libvirt randomly crashes on xenial nodes with "*** Error in `/usr/sbin/libvirtd': malloc(): memory corruption:" To manage notifications about this bug go to: https://bugs.launchpad.net/libvirt/+bug/1643911/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1643911] Re: libvirt randomly crashes on xenial nodes with "*** Error in `/usr/sbin/libvirtd': malloc(): memory corruption:"
We've also somewhat recently gotten the OOMkiller problems to go away. And yet these problems remain. I doubt it is related to OOMKiller. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1643911 Title: libvirt randomly crashes on xenial nodes with "*** Error in `/usr/sbin/libvirtd': malloc(): memory corruption:" To manage notifications about this bug go to: https://bugs.launchpad.net/libvirt/+bug/1643911/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1367907] Re: Segfault in gc with cyclic trash
I have confirmed that running `tox -epy34` against oslo.messaging with the old python3.4 packages continues to segfault. After upgrading to the newly built python3.4.3-1ubuntu1~14.04 package the segfaulting stops and the tests pass. This looks good to me. Thank you. Version of python from aptitude: $ aptitude show python3.4 Package: python3.4 State: installed Automatically installed: no Multi-Arch: allowed Version: 3.4.3-1ubuntu1~14.04 Priority: important Section: python Maintainer: Ubuntu Core Developers Architecture: amd64 Uncompressed Size: 353 k Depends: python3.4-minimal (= 3.4.3-1ubuntu1~14.04), libpython3.4-stdlib (= 3.4.3-1ubuntu1~14.04), mime-support Suggests: python3.4-venv, python3.4-doc, binutils Conflicts: python3.4 Provides: python3.4:any Description: Interactive high-level object-oriented language (version 3.4) ** Tags removed: verification-needed ** Tags added: verification-done -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1367907 Title: Segfault in gc with cyclic trash To manage notifications about this bug go to: https://bugs.launchpad.net/oslo.messaging/+bug/1367907/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1382607] Re: Backport python3.4 logging module backward incompatibility fix.
I should also note that https://hg.python.org/cpython/rev/a4c5effb8698 is where the upstream bug was fixed. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1382607 Title: Backport python3.4 logging module backward incompatibility fix. To manage notifications about this bug go to: https://bugs.launchpad.net/python/+bug/1382607/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1382607] [NEW] Backport python3.4 logging module backward incompatibility fix.
Public bug reported: Python 3.4 released with a backward incompatible change in the logging module. This bug has been tracked and fixed upstream in http://bugs.python.org/issue22386. Summary is that logging.getLevelName(lvl) would return an integer value for a string lvl and a string value for an integer lvl. It mapped the two representations of log levels to each other. This behavior existed in python2.6 through python3.3 but was removed in 3.4. The new behavior in 3.4 was to only map integer lvl values to strings. [Impact] Any python code that aims to be compatible with python2.6 and python3.4 must carry its own log level mappings (which may change under it because the loglevels are extensible) or access private data within the python logging module. Both approaches are fragile but thankfully upstream python 3.4 has patched this behavior so a downstream update to Trusty python3.4 would allow code to easily support 2.6 and 3.4. [Test Case] As detailed in the upstream bug you can test this easily via the interactive interpreter. Desired Behavior: >>> logging.getLevelName('INFO') 20 >>> logging.getLevelName(20) 'INFO' >>> Current Trusty Behavior: # This function call should return 20. >>> logging.getLevelName('INFO') 'Level INFO' >>> logging.getLevelName(20) 'INFO' >>> logging.getLevelName(logging.INFO) 'INFO' >>> [Regression Potential] The upstream patch is tiny and comes with a test case to track the desired behavior (upstream already had tests for the other behavior of this function). Beacuse this patch is so small and comes with new tests I think the regression potential is very small. Probably the only thing to consider is that new python3.4 code may have come to rely on this new backward incompatible behavior. But that would only be the case if code specifically looked for the mapping of 'Level %s' % 'stringlevelhere'. ** Affects: python Importance: Unknown Status: Unknown ** Affects: python3.4 (Ubuntu) Importance: Undecided Status: New ** Bug watch added: Python Roundup #22386 http://bugs.python.org/issue22386 ** Also affects: python via http://bugs.python.org/issue22386 Importance: Unknown Status: Unknown -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1382607 Title: Backport python3.4 logging module backward incompatibility fix. To manage notifications about this bug go to: https://bugs.launchpad.net/python/+bug/1382607/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1367907] Re: Segfault in gc with cyclic trash
** Description changed: Trusty's python3.4 package is affected by python bug http://bugs.python.org/issue21435. This bug was fixed in http://hg.python.org/cpython/rev/64ba3f2de99c. Trusty should pull this fix into the python3.4 package. Note this definitely affects some python projects. `git clone https://git.openstack.org/openstack/oslo.messaging && cd oslo.messaging && edit tox.ini file, copy py33 target and rename to py34 && tox -repy34` will segfault due to this bug. With previous versions of python unaffected by this bug there is no segfaulting. - Now for details: + Now for details, broken using: $ lsb_release -rd Description: Ubuntu 14.04.1 LTS Release: 14.04 $ apt-cache policy python3.4 python3.4: Installed: 3.4.0-2ubuntu1 Candidate: 3.4.0-2ubuntu1 Version table: *** 3.4.0-2ubuntu1 0 500 http://us.archive.ubuntu.com/ubuntu/ trusty/main amd64 Packages 100 /var/lib/dpkg/status + + If I install the following python3.4 packages for Unicorn on Trusty this + bug is corrected. The Unicorn packages here on lp for python3.4 are + python3.4.1 which includes the bug fix for this bug. + + libpython3.4_3.4.1-10ubuntu1_amd64.deb + libpython3.4-dev_3.4.1-10ubuntu1_amd64.deb + libpython3.4-stdlib_3.4.1-10ubuntu1_amd64.deb + python3.4-dbg_3.4.1-10ubuntu1_amd64.deb + python3.4-minimal_3.4.1-10ubuntu1_amd64.deb + libpython3.4-dbg_3.4.1-10ubuntu1_amd64.deb + libpython3.4-minimal_3.4.1-10ubuntu1_amd64.deb + python3.4_3.4.1-10ubuntu1_amd64.deb + python3.4-dev_3.4.1-10ubuntu1_amd64.deb + + Note I installed the -dbg packages too just in case I needed to use gdb + but that wasn't necessary as the test case works with these newer + versions of python. + + $ apt-cache policy python3.4 + python3.4: + Installed: 3.4.1-10ubuntu1 + Candidate: 3.4.1-10ubuntu1 + Version table: + *** 3.4.1-10ubuntu1 0 + 100 /var/lib/dpkg/status + 3.4.0-2ubuntu1 0 + 500 http://az2.clouds.archive.ubuntu.com/ubuntu/ trusty/main amd64 Packages + + [Impact] + + This python bug can cause python processes that tickle it to segfault + python. This means there could be a significant number of python + packages that are broken when run under this interpreter. The fix for + this bug should be backported to avoid seemingly random and hard to + debug segfaults from happening when users use python. + + Unicorn's 3.4.1 packages fix this bug by including the patch at + http://hg.python.org/cpython/rev/64ba3f2de99c. + + [Test Case] + + There is a test case detailed in the upstream bug http://bugs.python.org/issue21435, though I have been doing the following instead: + git clone https://git.openstack.org/openstack/oslo.messaging + cd oslo.messaging + git fetch https://review.openstack.org/openstack/oslo.messaging refs/changes/90/118790/2 && git checkout FETCH_HEAD + tox -repy34 + # This should end with segfaulting test runners. Note if you don't have tox installed you will need to install version 1.7.2 or greater. `sudo pip install tox==1.7.2` will do this. + + [Regression Potential] + + The patch in question is small. If we go straight to python 3.4.1 the + diff will be larger but that isn't necessary to fix this particular + issue. The biggest regression potential would be that the garbage + collector is newly broken by the this bug fix. Considering that the + garbage collector is already broken and this patch comes with test cases + to check it is less broken the risk of regression here is less than the + pain of dealing with this bug. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1367907 Title: Segfault in gc with cyclic trash To manage notifications about this bug go to: https://bugs.launchpad.net/oslo.messaging/+bug/1367907/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1367907] [NEW] Segfault in gc with cyclic trash
Public bug reported: Trusty's python3.4 package is affected by python bug http://bugs.python.org/issue21435. This bug was fixed in http://hg.python.org/cpython/rev/64ba3f2de99c. Trusty should pull this fix into the python3.4 package. Note this definitely affects some python projects. `git clone https://git.openstack.org/openstack/oslo.messaging && cd oslo.messaging && edit tox.ini file, copy py33 target and rename to py34 && tox -repy34` will segfault due to this bug. With previous versions of python unaffected by this bug there is no segfaulting. Now for details: $ lsb_release -rd Description:Ubuntu 14.04.1 LTS Release:14.04 $ apt-cache policy python3.4 python3.4: Installed: 3.4.0-2ubuntu1 Candidate: 3.4.0-2ubuntu1 Version table: *** 3.4.0-2ubuntu1 0 500 http://us.archive.ubuntu.com/ubuntu/ trusty/main amd64 Packages 100 /var/lib/dpkg/status ** Affects: oslo.messaging Importance: Undecided Assignee: Clark Boylan (cboylan) Status: New ** Affects: python Importance: Unknown Status: Unknown ** Affects: python3.4 (Ubuntu) Importance: Undecided Status: New ** Bug watch added: Python Roundup #21435 http://bugs.python.org/issue21435 ** Also affects: python via http://bugs.python.org/issue21435 Importance: Unknown Status: Unknown ** Description changed: Trusty's python3.4 package is affected by python bug http://bugs.python.org/issue21435. This bug was fixed in http://hg.python.org/cpython/rev/64ba3f2de99c. Trusty should pull this fix into the python3.4 package. Note this definitely affects some python projects. `git clone https://git.openstack.org/openstack/oslo.messaging && cd oslo.messaging - && tox -repy34` will segfault due to this bug. With previous versions of - python unaffected by this bug there is no segfaulting. + && edit tox.ini file, copy py33 target and rename to py34 && tox + -repy34` will segfault due to this bug. With previous versions of python + unaffected by this bug there is no segfaulting. Now for details: $ lsb_release -rd Description: Ubuntu 14.04.1 LTS Release: 14.04 $ apt-cache policy python3.4 python3.4: - Installed: 3.4.0-2ubuntu1 - Candidate: 3.4.0-2ubuntu1 - Version table: - *** 3.4.0-2ubuntu1 0 - 500 http://us.archive.ubuntu.com/ubuntu/ trusty/main amd64 Packages - 100 /var/lib/dpkg/status + Installed: 3.4.0-2ubuntu1 + Candidate: 3.4.0-2ubuntu1 + Version table: + *** 3.4.0-2ubuntu1 0 + 500 http://us.archive.ubuntu.com/ubuntu/ trusty/main amd64 Packages + 100 /var/lib/dpkg/status ** Also affects: oslo.messaging Importance: Undecided Status: New -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1367907 Title: Segfault in gc with cyclic trash To manage notifications about this bug go to: https://bugs.launchpad.net/oslo.messaging/+bug/1367907/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1242992] Re: Unable to autolaunch a dbus-daemon without a $DISPLAY for X11
This is still a bug... python keystoneclient does not explicitly depend on python keyring. So when you pip install python-keystoneclient it will not pull in the version of keyring that it needs. However if you have keyring installed (say because your distro gives it to you), python- keystoneclient will use that version without checking if it is the correct version. If python-keystone client is going to depend on python keyring it should list that package as a dependency. If it is going to optionally use keyring it should double check the version of keyring that is available. Also, maybe python-keystoneclient should just cache tokens in memory and avoid this keyring business completely. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1242992 Title: Unable to autolaunch a dbus-daemon without a $DISPLAY for X11 To manage notifications about this bug go to: https://bugs.launchpad.net/python-openstackclient/+bug/1242992/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1242992] Re: Unable to autolaunch a dbus-daemon without a $DISPLAY for X11
** Changed in: python-openstackclient Status: Invalid => Confirmed ** Changed in: python-keystoneclient (Ubuntu) Status: Invalid => Confirmed -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1242992 Title: Unable to autolaunch a dbus-daemon without a $DISPLAY for X11 To manage notifications about this bug go to: https://bugs.launchpad.net/python-openstackclient/+bug/1242992/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1273386] Re: Neutron namespace metadata proxy triggers kernel crash on Ubuntu 12.04/3.2 kernel
Couple of things. We get ext3 on these filesystems because that is what Rackspace gives us. I have a hunch that if OpenStack (nova) were to change its default fs to ext4 that in a week and a half we would start seeing that in the nodes Rackspace gives us. But the first iteration of that change got reverted so it will need to be sorted. 3.2 is a bit old but we recently tried 3.11 (using the saucy precise kernel backport) and that didn't fix the problem according to the comments above. I do have a change (https://review.openstack.org/70283) that will mount the ext3 filesystems we are given as ext4. We can see if that will help. But really we have the ability to fix these problems as an upstream, we should take advantage of that. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1273386 Title: Neutron namespace metadata proxy triggers kernel crash on Ubuntu 12.04/3.2 kernel To manage notifications about this bug go to: https://bugs.launchpad.net/neutron/+bug/1273386/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1228977] Re: n-cpu seems to crash when running with libvirt 1.0.6 from ubuntu cloud archive
Daniel Berrange is proposing that nova stop supporting libvirt older than 0.9.11 as libvirt should be able to provide pip installable python bindings for libvirt >= 0.9.11. The vast majority of our test slaves run on precise which has libvirt 0.9.8 in the base install. The only sane way for us to support libvirt >= 0.9.11 is to use the cloud archive. Can someone test this with cloud archive's libvirt 1.1.1 so that we can move to it if the problem is solved? ML thread is http://lists.openstack.org/pipermail/openstack- operators/2013-November/003748.html and there is a sister thread at http://lists.openstack.org/pipermail/openstack- dev/2013-November/020172.html. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1228977 Title: n-cpu seems to crash when running with libvirt 1.0.6 from ubuntu cloud archive To manage notifications about this bug go to: https://bugs.launchpad.net/cloud-archive/+bug/1228977/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs