This was already fix released for ocata but was buried under another
version in the changelog. Manually marking fix released.
** Changed in: cloud-archive/ocata
Status: Fix Committed => Fix Released
** Changed in: cloud-archive/newton
Status: Fix Committed => Fix Released
--
You
This was also already fix released for newton but was buried under
another version in the changelog. Manually marking fix released.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1752838
Title:
This bug was fixed in the package neutron - 2:11.0.3-0ubuntu2~cloud0
---
neutron (2:11.0.3-0ubuntu2~cloud0) xenial-pike; urgency=medium
.
* New update for the Ubuntu Cloud Archive.
.
neutron (2:11.0.3-0ubuntu2) artful; urgency=medium
.
*
Hello,
for pike
I followed below step to verify this.
1. deploy openstack xenial
2. source novarc
3. create instance ( https://pastebin.ubuntu.com/p/gZ3XXy9j4F/ )
- check this instance's host
4. login to host
5. adding dummy ports via ( https://paste.ubuntu.com/p/X35H85XWPh/ )
6. reboot and
Seyeong, Can you test pike-proposed again please? The updated package
wasn't in proposed.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1752838
Title:
[SRU] neutron-ovs-cleanup runs earlier than
Hello Seyeong, or anyone else affected,
Accepted neutron into pike-proposed. The package will build now and be
available in the Ubuntu Cloud Archive in a few hours, and then in the
-proposed repository.
Please help us by testing this new package. To enable the -proposed
repository:
sudo
This bug was fixed in the package neutron - 2:11.0.3-0ubuntu2
---
neutron (2:11.0.3-0ubuntu2) artful; urgency=medium
* d/neutron-openvswitch-agent.service.in,
d/neutron-openvswitch-agent.neutron-ovs-cleanup.service.in:
Ensure neutron-ovs-cleanup runs after
Hello,
for artful, ( without lxd )
I followed below step to verify this.
1. deploy openstack based on artful
2. source novarcv3_project
3. create instance via horizon
- check this instance's host
4. login to host
5. adding dummy ports via ( https://paste.ubuntu.com/p/X35H85XWPh/ )
6. reboot and
Hello Seyeong, or anyone else affected,
Accepted neutron into artful-proposed. The package will build now and be
available at
https://launchpad.net/ubuntu/+source/neutron/2:11.0.3-0ubuntu2 in a few
hours, and then in the -proposed repository.
Please help us by testing this new package. See
This bug was fixed in the package neutron - 2:8.4.0-0ubuntu7.3
---
neutron (2:8.4.0-0ubuntu7.3) xenial; urgency=medium
[ Edward Hope-Morley ]
* d/p/Wait-for-ovsdb_monitor-to-be-active-before-use-it.patch: backport
fix to ensure ovsdb mon blocks on start until it is ready (LP:
Hello,
for xenial
I followed below step to verify this.
1. deploy openstack xenial
2. source novarc
3. create instance ( https://pastebin.ubuntu.com/p/gZ3XXy9j4F/ )
- check this instance's host
4. login to host
5. adding dummy ports via ( https://paste.ubuntu.com/p/X35H85XWPh/ )
6. reboot and
Hello Seyeong, or anyone else affected,
Accepted neutron into xenial-proposed. The package will build now and be
available at
https://launchpad.net/ubuntu/+source/neutron/2:8.4.0-0ubuntu7.3 in a few
hours, and then in the -proposed repository.
Please help us by testing this new package. See
Hello,
for xenial-queens
I followed below step to verify this. ( slightly different )
1. deploy openstack xenial-queens
2. create networks routers etc on horizon
3. create instance on horizon
4. login to host
5. adding dummy ports via ( https://paste.ubuntu.com/p/X35H85XWPh/ )
6. reboot and
Hello,
for xenial-newton
I followed below step to verify this. ( same as above )
1. deploy openstack xenial-ocata
2. source novarc
3. create instance ( https://pastebin.ubuntu.com/p/gZ3XXy9j4F/ )
- check this instance's host
4. login to host
5. adding dummy ports via (
Hello,
for xenial-ocata
I followed below step to verify this.
1. deploy openstack xenial-ocata
2. source novarc
3. create instance ( https://pastebin.ubuntu.com/p/gZ3XXy9j4F/ )
- check this instance's host
4. login to host
5. adding dummy ports via ( https://paste.ubuntu.com/p/X35H85XWPh/ )
6.
This bug was fixed in the package neutron - 2:12.0.0-0ubuntu2
---
neutron (2:12.0.0-0ubuntu2) bionic; urgency=medium
* d/neutron-openvswitch-agent.service.in,
d/neutron-openvswitch-agent.neutron-ovs-cleanup.service.in:
Ensure neutron-ovs-cleanup runs after
Thanks Seyeong. I've uploaded new versions of neutron with your patch
(updated changelog as we discussed) for all affected releases. Artful
and Xenial are awaiting review by the SRU team.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to
** Patch added: "lp1752838_xenial_v2.debdiff"
https://bugs.launchpad.net/ubuntu/+source/neutron/+bug/1752838/+attachment/5093483/+files/lp1752838_xenial_v2.debdiff
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
Hey Corey
yes I meant that yours. ( removing Requires line from regression patch )
1630968 is the other issue I faced.
not related to this issue. ( just mentioned it for log )
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
It seems that adding:
After=openvswitch-switch.service
to neutron-ovs-cleanup.service will be enough to fix this bug.
As long as openvswitch-switch is not already installed at the time of
neutron-openvswitch-agent being installed, then neutron-openvswitch-
agent will cause openvswitch-switch to
Confirmed that removing Require-openvswitch-switch.server line in
neutron-ovs-cleanup.service file solve regression and this order issue.
but as I mentioned just before, if there are a lot of (1000) ports
there, 14 alarm clock error when running neutron-ovs-cleanup
and this[1] is pop up (this
Require=openvswitch-switch.service in neutron-ovs-cleanup.service file
trigger calling neutron-ovs-cleanup when restarting openvswitch-switch
as earlier analysis, neutron-ovs-cleanup shouldn't have called while
running machine.
it is only for booting sequence.
I'm testing it without require
This change has been reverted due to regression. Please see
https://bugs.launchpad.net/cloud-archive/mitaka/+bug/1758411 for more
details.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1752838
Title:
Change Status back to New.
** Changed in: cloud-archive/ocata
Status: Fix Released => Triaged
** Changed in: cloud-archive/newton
Status: Fix Released => Triaged
** Changed in: cloud-archive/queens
Status: Fix Released => Triaged
** Changed in: cloud-archive/pike
reverting in
https://bugs.launchpad.net/ubuntu/+source/neutron/+bug/1758411
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1752838
Title:
[SRU] neutron-ovs-cleanup runs earlier than ovsdb-server
WARNING: this update has introduced a race between ovs-cleanup and
neutron-openvswitch-agent such that when ovs is restarted, neutron-
openvswitch-agent AND neutron-ovs-cleanup are also restarted resulting
in the guest flows never getting re-added to br-tun. Recommend we either
fix or pull this
** Changed in: cloud-archive/queens
Status: Fix Committed => Fix Released
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1752838
Title:
[SRU] neutron-ovs-cleanup runs earlier than
This bug was fixed in the package neutron - 2:9.4.1-0ubuntu1~cloud2.1
---
neutron (2:9.4.1-0ubuntu1~cloud2.1) xenial-newton; urgency=medium
.
* d/neutron-openvswitch-agent.service.in: Ensure
neutron-openvswitch-agent starts after neutron-ovs-cleanup.
*
This bug was fixed in the package neutron - 2:11.0.2-0ubuntu1.3~cloud0
---
neutron (2:11.0.2-0ubuntu1.3~cloud0) xenial-pike; urgency=medium
.
* New update for the Ubuntu Cloud Archive.
.
neutron (2:11.0.2-0ubuntu1.3) artful; urgency=medium
.
*
This bug was fixed in the package neutron - 2:10.0.4-0ubuntu2~cloud1.1
---
neutron (2:10.0.4-0ubuntu2~cloud1.1) xenial-ocata; urgency=medium
.
* d/neutron-openvswitch-agent.service.in: Ensure
neutron-openvswitch-agent starts after neutron-ovs-cleanup.
*
This bug was fixed in the package neutron - 2:8.4.0-0ubuntu7.1
---
neutron (2:8.4.0-0ubuntu7.1) xenial; urgency=medium
* d/neutron-openvswitch-agent.service.in: Ensure
neutron-openvswitch-agent starts after neutron-ovs-cleanup.
*
This bug was fixed in the package neutron - 2:11.0.2-0ubuntu1.3
---
neutron (2:11.0.2-0ubuntu1.3) artful; urgency=medium
* d/neutron-openvswitch-agent.service.in: Ensure
neutron-openvswitch-agent starts after neutron-ovs-cleanup.
*
Hello
I verified this.
There was an issue about deploying lxd on artful machine.
I just remove them from lxd and depoyed with virtual machines each.
and done steps below
1. deploy artful openstack env ( without any lxd instance )
2. run scripts on test case
3. reboot compute node and check if
hey sil2100,
sure, I'm trying to verify this on artful as well
with juju openstack deployment with artful release, there is an issue on
deployment
and now im in sprint, so it is delaying..
I'm going to fix it or find new reproducer for this and verify this ASAP
Thanks.
--
You received this
Could this also be verified for artful? We can't release the xenial
version if the artful counterpart isn't verified as well.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1752838
Title:
[SRU]
for xenial pike
I did same test on pike deployment
1. deploy pike env
2. run script
3. reboot nova-compute node and check if there is order weirdness
4. upgrading neutron pkgs from -proposed
5. reboot nova-compute node and check if it is correct order ( it is fine )
it is working fine.
Thanks.
for xenial
I checked that this patch is working.
1. deploy xenial
2. ran script on test case
3. reboot, and check if there is issue on order or failed service
4. upgrade to -proposed pkg
5. reboot, and check if there is issue(it is gone)
Thanks a lot
dpkg -l | grep neutron
ii neutron-common
Hello Seyeong, or anyone else affected,
Accepted neutron into xenial-proposed. The package will build now and be
available at
https://launchpad.net/ubuntu/+source/neutron/2:8.4.0-0ubuntu7.1 in a few
hours, and then in the -proposed repository.
Please help us by testing this new package. See
Hello Seyeong, or anyone else affected,
Accepted neutron into artful-proposed. The package will build now and be
available at
https://launchpad.net/ubuntu/+source/neutron/2:11.0.2-0ubuntu1.3 in a
few hours, and then in the -proposed repository.
Please help us by testing this new package. See
for xenial newton
this is the same as above result,
in this case, I saw same issue as description(several connection error
to ovsdb-server)
after patching, it has gone.
order is as expected.
Thanks
dpkg -l | grep neutron
ii neutron-common 2:9.4.1-0ubuntu1~cloud2.1
for xenial ocata
ok, service starting order is same as above(queens)
there is no error in neutron or openvswitch services with 1000ports
Thanks
dpkg -l | grep neutron
ii neutron-common 2:10.0.4-0ubuntu2~cloud1.1
all Neutron is a virtual network
for xenial queens
updated pkg after last comment above,
now services run as below order
1. openvswitch-switch
2. neutron-ovs-cleanup
3. neutron-openvswitch-agent
this is expected
Thanks.
dpkg -l | grep neutron
ii neutron-common 2:12.0.0-0ubuntu1.2~cloud0
Based on Seyeong's testing input, I've uploaded new versions of neutron
to all affected releases. These versions add After=neutron-ovs-cleanup
to neutron-openvswitch-agent. Thanks Seyeong!
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to
for xenial-queens
After patching neutron-ovs-cleanup is following after ovsdb-server
properly.
while testing xenial-queens
I noticed that neutron-openvswitch-agent has Requires, but not After for
neutron-ovs-cleanup
and in syslog
neutron-openvswitch-agent started earlier than above
for xenial-newton
I could see above error without patch, after running script and
rebooting
after patch, order is correct and error is gone.
Thanks.
dpkg info
ii neutron-common 2:9.4.1-0ubuntu1~cloud2
all Neutron is a virtual network service
for xenial-ocata,
I ran script as above (changed after some more tests)
before installing patch, neutron openvswitch agent ran earlier than
ovsdb-server(openvswitch-switch service)
after installing patch, openvswitch-switch runs earlier.
the symptom as described above is sometimes happening
** Description changed:
[Impact]
In code, Basically all ubuntu version could be affected.
I'm trying to do test them
neutron-ovs-cleanup service runs earlier than ovsdb-server sometimes.
then, neutron-ovs-cleanup service's status remains failed
with below error msg
** Description changed:
[Impact]
In code, Basically all ubuntu version could be affected.
I'm trying to do test them
Xenial - Tested, Affected
Artful -
Bionic -
- UCA-Mitaka -
+ UCA-Mitaka -
UCA-Newton -
UCA-Ocata -
UCA-Pike -
UCA-Queens -
..?
This bug was fixed in the package neutron - 2:12.0.0-0ubuntu1.1
---
neutron (2:12.0.0-0ubuntu1.1) bionic; urgency=medium
* d/neutron-openvswitch-agent.neutron-ovs-cleanup.service.in:
Adding dependency on openvswitch-switch.service because
neutron-ovs-cleanup will fail if it
Hi Seyeong, I've uploaded patched versions of neutron to xenial, xenial-
newton, xenial-ocata, artful, and bionic. xenial, artful, and bionic are
in the unapproved queue awaiting sru review.
Thanks again for the bug/patch.
Corey
--
You received this bug notification because you are a member of
Adding Requires=neutron-ovs-cleanup.service to neutron-l3-agent.service,
neutron-dhcp-agent.service, and nova-compute.service is turning out to
be perhaps too tightly coupled.
Doing this would also require the binary packages for each of the above
named services to depend on the
@corey,
I inserted service file in debian folder.
Tested with manipulated init.in file but no luck.
is it proper to upload nova debdiff here?
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1752838
** Patch added: "lp1752838_nova_denial.debdiff"
https://bugs.launchpad.net/cloud-archive/+bug/1752838/+attachment/5074235/+files/lp1752838_nova_denial.debdiff
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
** Patch added: "lp1752838_neutron_xenial.debdiff"
https://bugs.launchpad.net/cloud-archive/+bug/1752838/+attachment/5074234/+files/lp1752838_neutron_xenial.debdiff
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
Hello
I've been testing xenial-newton env if there is also same issue.
but final symptom was not same.
it keep restarting on boot and eventually status is settled on active.
(there was error about connecting to ovsdb-server several times but.. )
I was stuck on that to find why it is different.
... it will run it's action without keeping an active process.
RemainAfterExit=yes causes the unit to remain active after the start
action exited successfully.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
I reviewed the systemd unit files that have a dependency on neutron-ovs-
cleanup and I think we could use some updates there as well to tighten
up the dependency on neutron-ovs-cleanup.
We currently have the following and I think the latter 3 should be
updated with a Requires like
Triaged high as this neutron-ovs-cleanup init scripts are not behaving
as expected, which may cause a number of issues.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1752838
Title:
[SRU]
This affects xenial-mitaka and above. My test results above are on
xenial-queens.
** Also affects: neutron (Ubuntu Bionic)
Importance: Low
Status: New
** Also affects: neutron (Ubuntu Xenial)
Importance: Undecided
Status: New
** Also affects: neutron (Ubuntu Artful)
Correction. kilo is trusty only at this point so kilo is not affected.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1752838
Title:
[SRU] neutron-ovs-cleanup runs earlier than ovsdb-server when
Note: this affects back to kilo. It doesn't affect icehouse as we were
fully using upstart at that point.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1752838
Title:
[SRU] neutron-ovs-cleanup runs
Current behavior: starting neutron-ovs-cleanup when openvswitch-switch
is stopped: https://paste.ubuntu.com/p/MxHT5rXJJ2/
New behavior [1]: starting neutron-ovs-cleanup when openvswitch-switch
is stopped: https://paste.ubuntu.com/p/wzJ3kh6CPp/
[1] After updating
Hi Seyeong, I think your patch makes sense. I wonder if we also need to
add an After statement to ensure openvswitch-switch starts before
neutron-ovs-cleanup:
+Requires=openvswitch-switch.service
+After=openvswitch-switch.service
Note that the LSB init info in the beginning of
Hi Mathew Hodson, In the future can you please explain your reasoning
when you set bug importance? Thanks.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1752838
Title:
[SRU] neutron-ovs-cleanup
** Description changed:
[Impact]
+
+ In code, Basically all ubuntu version could be affected.
+ I'm trying to do test them
+
+ Xenial - Tested, Affected
+ Artful -
+ Bionic -
+
+ UCA-Mitaka -
+ UCA-Newton -
+ UCA-Ocata -
+ UCA-Pike -
+ UCA-Queens -
+ ..?
neutron-ovs-cleanup service
** Summary changed:
- [SRU] sometimes neutron-ovs-cleanup runs earlier than ovsdb-server
+ [SRU] neutron-ovs-cleanup runs earlier than ovsdb-server when there are many
port
** Tags added: sts sts-sru-needed
** Description changed:
[Impact]
neutron-ovs-cleanup service runs earlier than
66 matches
Mail list logo
