[tripleo] missing centos-8 rpms for kolla builds
Greetings, I know the ceph repo is in progress. TripleO / RDO is not releasing opendaylight Can the RDO team comment on the rest of the missing packages here please? Thank you!! https://review.opendev.org/#/c/699414/9/kolla/image/build.py NOTE(mgoddard): Mark images with missing dependencies as unbuildable for # CentOS 8. 'centos8': { "barbican-api", # Missing uwsgi-plugin-python3 "ceph-base", # Missing Ceph repo "cinder-base", # Missing Ceph repo "collectd", # Missing collectd-ping and # collectd-sensubility packages "elasticsearch", # Missing elasticsearch repo "etcd", # Missing etcd package "fluentd", # Missing td-agent repo "glance-base", # Missing Ceph repo "gnocchi-base", # Missing Ceph repo "hacluster-base", # Missing hacluster repo "ironic-conductor", # Missing shellinabox package "kibana", # Missing elasticsearch repo "manila-share", # Missing Ceph repo "mongodb", # Missing mongodb and mongodb-server packages "monasca-grafana", # Using python2 "nova-compute", # Missing Ceph repo "nova-libvirt", # Missing Ceph repo "nova-spicehtml5proxy", # Missing spicehtml5 package "opendaylight", # Missing opendaylight repo "ovsdpdk", # Not supported on CentOS "sensu-base", # Missing sensu package "tgtd", # Not supported on CentOS 8 }, 'centos8+source': { "barbican-base", # Missing uwsgi-plugin-python3 "bifrost-base", # Bifrost does not support CentOS 8 "cyborg-agent", # opae-sdk does not support CentOS 8 "freezer-base", # Missing package trickle "masakari-monitors", # Missing hacluster repo "zun-compute", # Missing Ceph repo
On 24/01/2020 19:20, Wesley Hayutin wrote:
Greetings,
I know the ceph repo is in progress. TripleO / RDO is not releasing opendaylight
Can the RDO team comment on the rest of the missing packages here please?
Thank you!!
https://review.opendev.org/#/c/699414/9/kolla/image/build.py
NOTE(mgoddard): Mark images with missing dependencies as unbuildable for # CentOS 8. 'centos8': { "barbican-api", # Missing uwsgi-plugin-python3 "ceph-base", # Missing Ceph repo "cinder-base", # Missing Ceph repo "collectd", # Missing collectd-ping and # collectd-sensubility packages
I can comment on collectd here. It should be possible to build packages on the CentOS build system for CentOS 8 right now or the near future. I am planning especially to do the collectd packages in CentOS opstools again, which would also bring back the amqp1 plugin (and a few others). We used to build fluentd in the past for CentOS 7, but I do have no plans for this on CentOS 8. Matthias
Hi, We were given access to CBS to build centos8 dependencies a couple of days ago and we are still in the process of re-bootstraping it. I hope we'll have all that is missing in the next days. See my comments below. Best regards, Alfredo On Fri, Jan 24, 2020 at 7:21 PM Wesley Hayutin <whayutin@redhat.com> wrote:
Greetings,
I know the ceph repo is in progress. TripleO / RDO is not releasing opendaylight
Can the RDO team comment on the rest of the missing packages here please?
Thank you!!
https://review.opendev.org/#/c/699414/9/kolla/image/build.py
NOTE(mgoddard): Mark images with missing dependencies as unbuildable for # CentOS 8. 'centos8': { "barbican-api", # Missing uwsgi-plugin-python3
We'll take care of uwsgi.
"ceph-base", # Missing Ceph repo "cinder-base", # Missing Ceph repo "collectd", # Missing collectd-ping and # collectd-sensubility packages
About collectd and sensu, Matthias already replied from OpsTools side
"elasticsearch", # Missing elasticsearch repo "etcd", # Missing etcd package
Given that etcd is not longer in CentOS base (it was in 7), I guess we'll take care of etcd unless some other sig is building it as part of k8s family.
"fluentd", # Missing td-agent repo
See Matthias reply.
"glance-base", # Missing Ceph repo "gnocchi-base", # Missing Ceph repo "hacluster-base", # Missing hacluster repo
That's an alternative repo for HA related packages for CentOS: http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen... Which still does not provide packages for centos8. Note that centos8.1 includes pacemaker, corosync and pcs in HighAvailability repo. Maybe it could be used instead of the current one.
"ironic-conductor", # Missing shellinabox package
shellinabox is epel. It was never used in tripleo containers, it's really required?
"kibana", # Missing elasticsearch repo
We never provided elasticsearch in the past, is consumed from elasticsearch repo iirc
"manila-share", # Missing Ceph repo "mongodb", # Missing mongodb and mongodb-server packages
Mongodb was retired from RDO time ago as it was not longer the recommended backend for any service. In CentOS7 is pulled from EPEL.
"monasca-grafana", # Using python2 "nova-compute", # Missing Ceph repo "nova-libvirt", # Missing Ceph repo "nova-spicehtml5proxy", # Missing spicehtml5 package
spice-html5 is pulled from epel7 was never part of RDO. Not used in TripleO.
"opendaylight", # Missing opendaylight repo "ovsdpdk", # Not supported on CentOS "sensu-base", # Missing sensu package
See Matthias reply.
"tgtd", # Not supported on CentOS 8
tgtd was replace by scsi-target-utils. It's was never provided in RDO, in kolla was pulled from epel for 7
},
'centos8+source': { "barbican-base", # Missing uwsgi-plugin-python3 "bifrost-base", # Bifrost does not support CentOS 8 "cyborg-agent", # opae-sdk does not support CentOS 8 "freezer-base", # Missing package trickle "masakari-monitors", # Missing hacluster repo "zun-compute", # Missing Ceph repo _______________________________________________ dev mailing list dev@lists.rdoproject.org http://lists.rdoproject.org/mailman/listinfo/dev
To unsubscribe: dev-unsubscribe@lists.rdoproject.org
W dniu 24.01.2020 o 20:20, Alfredo Moralejo Alonso pisze:
We were given access to CBS to build centos8 dependencies a couple of days ago and we are still in the process of re-bootstraping it. I hope we'll have all that is missing in the next days.
Good to hear that.
https://review.opendev.org/#/c/699414/9/kolla/image/build.py
NOTE(mgoddard): Mark images with missing dependencies as unbuildable for CentOS 8. 'centos8': { "elasticsearch", # Missing elasticsearch
Out of CentOS repo.
"hacluster-base", # Missing hacluster repo
Note that centos8.1 includes pacemaker, corosync and pcs in HighAvailability repo. Maybe it could be used instead of the > current one.
https://review.opendev.org/702706 enables this repo. One image is still disabled due to lack of 'crmsh'.
"mongodb", # Missing mongodb and mongodb-server packages
Mongodb was retired from RDO time ago as it was not longer the recommended backend for any service. In CentOS7 is pulled from EPEL.
I think we need to deprecate it in Ussuri and remove in V cycle.
"monasca-grafana", # Using python2
This is nightmare image.
"tgtd", # Not supported on CentOS 8
tgtd was replace by scsi-target-utils. It's was never provided in RDO, in kolla was pulled from epel for 7
https://review.opendev.org/#/c/613815/15 (merged) took care of it.
On Fri, Jan 24, 2020 at 10:48 PM Marcin Juszkiewicz < marcin.juszkiewicz@linaro.org> wrote:
W dniu 24.01.2020 o 20:20, Alfredo Moralejo Alonso pisze:
We were given access to CBS to build centos8 dependencies a couple of days ago and we are still in the process of re-bootstraping it. I hope we'll have all that is missing in the next days.
Good to hear that.
https://review.opendev.org/#/c/699414/9/kolla/image/build.py
NOTE(mgoddard): Mark images with missing dependencies as unbuildable for CentOS 8. 'centos8': { "elasticsearch", # Missing elasticsearch
Out of CentOS repo.
"hacluster-base", # Missing hacluster repo
Note that centos8.1 includes pacemaker, corosync and pcs in HighAvailability repo. Maybe it could be used instead of the > current one.
https://review.opendev.org/702706 enables this repo. One image is still disabled due to lack of 'crmsh'.
How is crmsh used in these images?, ha packages included in HighAvailability repo in CentOS includes pcs and some crm_* commands in pcs and pacemaker-cli packages. IMO, tt'd be good to switch to those commands to manage the cluster.
"mongodb", # Missing mongodb and mongodb-server packages
Mongodb was retired from RDO time ago as it was not longer the recommended backend for any service. In CentOS7 is pulled from EPEL.
I think we need to deprecate it in Ussuri and remove in V cycle.
"monasca-grafana", # Using python2
This is nightmare image.
"tgtd", # Not supported on CentOS 8
tgtd was replace by scsi-target-utils. It's was never provided in RDO, in kolla was pulled from epel for 7
https://review.opendev.org/#/c/613815/15 (merged) took care of it.
W dniu 27.01.2020 o 09:48, Alfredo Moralejo Alonso pisze:
How is crmsh used in these images?, ha packages included in HighAvailability repo in CentOS includes pcs and some crm_* commands in pcs and pacemaker-cli packages. IMO, tt'd be good to switch to those commands to manage the cluster.
No idea. Gaëtan Trellu may know - he created those images.
I know it was for masakari. Gaëtan had to grab crmsh from opensuse: http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen... -yoctozepto pon., 27 sty 2020 o 10:13 Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org> napisał(a):
W dniu 27.01.2020 o 09:48, Alfredo Moralejo Alonso pisze:
How is crmsh used in these images?, ha packages included in HighAvailability repo in CentOS includes pcs and some crm_* commands in pcs and pacemaker-cli packages. IMO, tt'd be good to switch to those commands to manage the cluster.
No idea. Gaëtan Trellu may know - he created those images.
On Mon, 27 Jan 2020 at 09:18, Radosław Piliszek <radoslaw.piliszek@gmail.com> wrote:
I know it was for masakari. Gaëtan had to grab crmsh from opensuse: http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen...
-yoctozepto
Thanks Wes for getting this discussion going. I've been looking at CentOS 8 today and trying to assess where we are. I created an Etherpad to track status: https://etherpad.openstack.org/p/kolla-centos8
pon., 27 sty 2020 o 10:13 Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org> napisał(a):
W dniu 27.01.2020 o 09:48, Alfredo Moralejo Alonso pisze:
How is crmsh used in these images?, ha packages included in HighAvailability repo in CentOS includes pcs and some crm_* commands in pcs and pacemaker-cli packages. IMO, tt'd be good to switch to those commands to manage the cluster.
No idea. Gaëtan Trellu may know - he created those images.
On Tue, 28 Jan 2020 at 15:18, Mark Goddard <mark@stackhpc.com> wrote:
On Mon, 27 Jan 2020 at 09:18, Radosław Piliszek <radoslaw.piliszek@gmail.com> wrote:
I know it was for masakari. Gaëtan had to grab crmsh from opensuse: http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen...
-yoctozepto
Thanks Wes for getting this discussion going. I've been looking at CentOS 8 today and trying to assess where we are. I created an Etherpad to track status: https://etherpad.openstack.org/p/kolla-centos8
We are seeing an odd DNF error sometimes. DNF exits 141 with no error code when installing packages. It often happens on the rabbitmq and grafana images. There is a prompt about importing GPG keys prior to the error. Example: https://4eff4bb69c321960be39-770d619687de1bce0976465c40e4e9ca.ssl.cf2.rackcd... Related bug report? https://github.com/containers/libpod/issues/4431 Anyone familiar with it?
pon., 27 sty 2020 o 10:13 Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org> napisał(a):
W dniu 27.01.2020 o 09:48, Alfredo Moralejo Alonso pisze:
How is crmsh used in these images?, ha packages included in HighAvailability repo in CentOS includes pcs and some crm_* commands in pcs and pacemaker-cli packages. IMO, tt'd be good to switch to those commands to manage the cluster.
No idea. Gaëtan Trellu may know - he created those images.
On Tue, Jan 28, 2020 at 5:53 PM Mark Goddard <mark@stackhpc.com> wrote:
On Tue, 28 Jan 2020 at 15:18, Mark Goddard <mark@stackhpc.com> wrote:
On Mon, 27 Jan 2020 at 09:18, Radosław Piliszek <radoslaw.piliszek@gmail.com> wrote:
I know it was for masakari. Gaëtan had to grab crmsh from opensuse:
http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen...
-yoctozepto
Thanks Wes for getting this discussion going. I've been looking at CentOS 8 today and trying to assess where we are. I created an Etherpad to track status: https://etherpad.openstack.org/p/kolla-centos8
uwsgi and etcd are now available in rdo dependencies repo. Let me know if you find some issue with it.
We are seeing an odd DNF error sometimes. DNF exits 141 with no error code when installing packages. It often happens on the rabbitmq and grafana images. There is a prompt about importing GPG keys prior to the error.
Example: https://4eff4bb69c321960be39-770d619687de1bce0976465c40e4e9ca.ssl.cf2.rackcd...
Related bug report? https://github.com/containers/libpod/issues/4431
Anyone familiar with it?
Didn't know about this issue. BTW, there is rabbitmq-server in RDO dependencies repo if you are interested in using it from there instead of rabbit repo.
pon., 27 sty 2020 o 10:13 Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org> napisał(a):
W dniu 27.01.2020 o 09:48, Alfredo Moralejo Alonso pisze:
How is crmsh used in these images?, ha packages included in HighAvailability repo in CentOS includes pcs and some crm_*
commands in pcs
and pacemaker-cli packages. IMO, tt'd be good to switch to those commands to manage the cluster.
No idea. Gaëtan Trellu may know - he created those images.
On Wed, 29 Jan 2020 at 11:31, Alfredo Moralejo Alonso <amoralej@redhat.com> wrote:
On Tue, Jan 28, 2020 at 5:53 PM Mark Goddard <mark@stackhpc.com> wrote:
On Tue, 28 Jan 2020 at 15:18, Mark Goddard <mark@stackhpc.com> wrote:
On Mon, 27 Jan 2020 at 09:18, Radosław Piliszek <radoslaw.piliszek@gmail.com> wrote:
I know it was for masakari. Gaëtan had to grab crmsh from opensuse: http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen...
-yoctozepto
Thanks Wes for getting this discussion going. I've been looking at CentOS 8 today and trying to assess where we are. I created an Etherpad to track status: https://etherpad.openstack.org/p/kolla-centos8
uwsgi and etcd are now available in rdo dependencies repo. Let me know if you find some issue with it.
I found them, thanks.
We are seeing an odd DNF error sometimes. DNF exits 141 with no error code when installing packages. It often happens on the rabbitmq and grafana images. There is a prompt about importing GPG keys prior to the error.
Example: https://4eff4bb69c321960be39-770d619687de1bce0976465c40e4e9ca.ssl.cf2.rackcd...
Related bug report? https://github.com/containers/libpod/issues/4431
Anyone familiar with it?
Didn't know about this issue.
BTW, there is rabbitmq-server in RDO dependencies repo if you are interested in using it from there instead of rabbit repo.
It seems to be due to the use of a GPG check on the repo (as opposed to packages). DNF doesn't use keys imported via rpm --import for this (I'm not sure what it uses), and prompts to add the key. This breaks without a terminal. More explanation here: https://review.opendev.org/#/c/704782.
pon., 27 sty 2020 o 10:13 Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org> napisał(a):
W dniu 27.01.2020 o 09:48, Alfredo Moralejo Alonso pisze:
How is crmsh used in these images?, ha packages included in HighAvailability repo in CentOS includes pcs and some crm_* commands in pcs and pacemaker-cli packages. IMO, tt'd be good to switch to those commands to manage the cluster.
No idea. Gaëtan Trellu may know - he created those images.
On Wed, Jan 29, 2020 at 9:37 AM Mark Goddard <mark@stackhpc.com> wrote:
On Wed, 29 Jan 2020 at 11:31, Alfredo Moralejo Alonso <amoralej@redhat.com> wrote:
On Tue, Jan 28, 2020 at 5:53 PM Mark Goddard <mark@stackhpc.com> wrote:
On Tue, 28 Jan 2020 at 15:18, Mark Goddard <mark@stackhpc.com> wrote:
On Mon, 27 Jan 2020 at 09:18, Radosław Piliszek <radoslaw.piliszek@gmail.com> wrote:
I know it was for masakari. Gaëtan had to grab crmsh from opensuse: http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen...
-yoctozepto
Thanks Wes for getting this discussion going. I've been looking at CentOS 8 today and trying to assess where we are. I created an Etherpad to track status: https://etherpad.openstack.org/p/kolla-centos8
uwsgi and etcd are now available in rdo dependencies repo. Let me know if you find some issue with it.
I found them, thanks.
We are seeing an odd DNF error sometimes. DNF exits 141 with no error code when installing packages. It often happens on the rabbitmq and grafana images. There is a prompt about importing GPG keys prior to the error.
Example: https://4eff4bb69c321960be39-770d619687de1bce0976465c40e4e9ca.ssl.cf2.rackcd...
Related bug report? https://github.com/containers/libpod/issues/4431
Anyone familiar with it?
Didn't know about this issue.
BTW, there is rabbitmq-server in RDO dependencies repo if you are interested in using it from there instead of rabbit repo.
It seems to be due to the use of a GPG check on the repo (as opposed to packages). DNF doesn't use keys imported via rpm --import for this (I'm not sure what it uses), and prompts to add the key. This breaks without a terminal. More explanation here: https://review.opendev.org/#/c/704782.
librepo has its own keyring for repo signature verification. -- 真実はいつも一つ！/ Always, there's only one truth!
On Sun, 2 Feb 2020 at 21:06, Neal Gompa <ngompa13@gmail.com> wrote:
On Wed, Jan 29, 2020 at 9:37 AM Mark Goddard <mark@stackhpc.com> wrote:
On Wed, 29 Jan 2020 at 11:31, Alfredo Moralejo Alonso <amoralej@redhat.com> wrote:
On Tue, Jan 28, 2020 at 5:53 PM Mark Goddard <mark@stackhpc.com> wrote:
On Tue, 28 Jan 2020 at 15:18, Mark Goddard <mark@stackhpc.com> wrote:
On Mon, 27 Jan 2020 at 09:18, Radosław Piliszek <radoslaw.piliszek@gmail.com> wrote:
I know it was for masakari. Gaëtan had to grab crmsh from opensuse: http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen...
-yoctozepto
Thanks Wes for getting this discussion going. I've been looking at CentOS 8 today and trying to assess where we are. I created an Etherpad to track status: https://etherpad.openstack.org/p/kolla-centos8
uwsgi and etcd are now available in rdo dependencies repo. Let me know if you find some issue with it.
I found them, thanks.
We are seeing an odd DNF error sometimes. DNF exits 141 with no error code when installing packages. It often happens on the rabbitmq and grafana images. There is a prompt about importing GPG keys prior to the error.
Example: https://4eff4bb69c321960be39-770d619687de1bce0976465c40e4e9ca.ssl.cf2.rackcd...
Related bug report? https://github.com/containers/libpod/issues/4431
Anyone familiar with it?
Didn't know about this issue.
BTW, there is rabbitmq-server in RDO dependencies repo if you are interested in using it from there instead of rabbit repo.
It seems to be due to the use of a GPG check on the repo (as opposed to packages). DNF doesn't use keys imported via rpm --import for this (I'm not sure what it uses), and prompts to add the key. This breaks without a terminal. More explanation here: https://review.opendev.org/#/c/704782.
librepo has its own keyring for repo signature verification.
Thanks Neal. Any pointers on how to add keys to it?
-- 真実はいつも一つ！/ Always, there's only one truth!
On Wed, Feb 5, 2020 at 5:36 AM Mark Goddard <mark@stackhpc.com> wrote:
On Sun, 2 Feb 2020 at 21:06, Neal Gompa <ngompa13@gmail.com> wrote:
On Wed, Jan 29, 2020 at 9:37 AM Mark Goddard <mark@stackhpc.com> wrote:
On Wed, 29 Jan 2020 at 11:31, Alfredo Moralejo Alonso <amoralej@redhat.com> wrote:
On Tue, Jan 28, 2020 at 5:53 PM Mark Goddard <mark@stackhpc.com> wrote:
On Tue, 28 Jan 2020 at 15:18, Mark Goddard <mark@stackhpc.com> wrote:
On Mon, 27 Jan 2020 at 09:18, Radosław Piliszek <radoslaw.piliszek@gmail.com> wrote: > > I know it was for masakari. > Gaëtan had to grab crmsh from opensuse: > http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen... > > -yoctozepto
Thanks Wes for getting this discussion going. I've been looking at CentOS 8 today and trying to assess where we are. I created an Etherpad to track status: https://etherpad.openstack.org/p/kolla-centos8
uwsgi and etcd are now available in rdo dependencies repo. Let me know if you find some issue with it.
I found them, thanks.
We are seeing an odd DNF error sometimes. DNF exits 141 with no error code when installing packages. It often happens on the rabbitmq and grafana images. There is a prompt about importing GPG keys prior to the error.
Example: https://4eff4bb69c321960be39-770d619687de1bce0976465c40e4e9ca.ssl.cf2.rackcd...
Related bug report? https://github.com/containers/libpod/issues/4431
Anyone familiar with it?
Didn't know about this issue.
BTW, there is rabbitmq-server in RDO dependencies repo if you are interested in using it from there instead of rabbit repo.
It seems to be due to the use of a GPG check on the repo (as opposed to packages). DNF doesn't use keys imported via rpm --import for this (I'm not sure what it uses), and prompts to add the key. This breaks without a terminal. More explanation here: https://review.opendev.org/#/c/704782.
librepo has its own keyring for repo signature verification.
Thanks Neal. Any pointers on how to add keys to it?
There's no direct way other than to make sure that your repo files include the GPG public key in the gpgkey= entry, and that repo_gpgcheck=1 is enabled. DNF will automatically tell librepo to do the right thing here. Ideally, in the future, all of it will use the rpm keyring, but it doesn't for now... -- 真実はいつも一つ！/ Always, there's only one truth!
On Wed, 29 Jan 2020 at 11:31, Alfredo Moralejo Alonso <amoralej@redhat.com> wrote:
On Tue, Jan 28, 2020 at 5:53 PM Mark Goddard <mark@stackhpc.com> wrote:
On Tue, 28 Jan 2020 at 15:18, Mark Goddard <mark@stackhpc.com> wrote:
On Mon, 27 Jan 2020 at 09:18, Radosław Piliszek <radoslaw.piliszek@gmail.com> wrote:
I know it was for masakari. Gaëtan had to grab crmsh from opensuse: http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen...
-yoctozepto
Thanks Wes for getting this discussion going. I've been looking at CentOS 8 today and trying to assess where we are. I created an Etherpad to track status: https://etherpad.openstack.org/p/kolla-centos8
uwsgi and etcd are now available in rdo dependencies repo. Let me know if you find some issue with it.
I've been working on the backport of kolla CentOS 8 patches to the stable/train branch. It looks like these packages which were added to master are not present in Train.
We are seeing an odd DNF error sometimes. DNF exits 141 with no error code when installing packages. It often happens on the rabbitmq and grafana images. There is a prompt about importing GPG keys prior to the error.
Example: https://4eff4bb69c321960be39-770d619687de1bce0976465c40e4e9ca.ssl.cf2.rackcd...
Related bug report? https://github.com/containers/libpod/issues/4431
Anyone familiar with it?
Didn't know about this issue.
BTW, there is rabbitmq-server in RDO dependencies repo if you are interested in using it from there instead of rabbit repo.
pon., 27 sty 2020 o 10:13 Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org> napisał(a):
W dniu 27.01.2020 o 09:48, Alfredo Moralejo Alonso pisze:
How is crmsh used in these images?, ha packages included in HighAvailability repo in CentOS includes pcs and some crm_* commands in pcs and pacemaker-cli packages. IMO, tt'd be good to switch to those commands to manage the cluster.
No idea. Gaëtan Trellu may know - he created those images.
On Mon, Feb 24, 2020 at 8:55 AM Mark Goddard <mark@stackhpc.com> wrote:
On Wed, 29 Jan 2020 at 11:31, Alfredo Moralejo Alonso <amoralej@redhat.com> wrote:
On Tue, Jan 28, 2020 at 5:53 PM Mark Goddard <mark@stackhpc.com> wrote:
On Tue, 28 Jan 2020 at 15:18, Mark Goddard <mark@stackhpc.com> wrote:
On Mon, 27 Jan 2020 at 09:18, Radosław Piliszek <radoslaw.piliszek@gmail.com> wrote:
I know it was for masakari. Gaëtan had to grab crmsh from opensuse:
http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen...
-yoctozepto
Thanks Wes for getting this discussion going. I've been looking at CentOS 8 today and trying to assess where we are. I created an Etherpad to track status: https://etherpad.openstack.org/p/kolla-centos8
uwsgi and etcd are now available in rdo dependencies repo. Let me know if you find some issue with it.
I've been working on the backport of kolla CentOS 8 patches to the stable/train branch. It looks like these packages which were added to master are not present in Train.
I'll help check in on that for you Mark. Thank you!!
We are seeing an odd DNF error sometimes. DNF exits 141 with no error code when installing packages. It often happens on the rabbitmq and grafana images. There is a prompt about importing GPG keys prior to the error.
Example:
Related bug report? https://github.com/containers/libpod/issues/4431
Anyone familiar with it?
Didn't know about this issue.
BTW, there is rabbitmq-server in RDO dependencies repo if you are interested in using it from there instead of rabbit repo.
pon., 27 sty 2020 o 10:13 Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org> napisał(a):
W dniu 27.01.2020 o 09:48, Alfredo Moralejo Alonso pisze: > How is crmsh used in these images?, ha packages included in > HighAvailability repo in CentOS includes pcs and some crm_*
commands in pcs
> and pacemaker-cli packages. IMO, tt'd be good to switch to
https://4eff4bb69c321960be39-770d619687de1bce0976465c40e4e9ca.ssl.cf2.rackcd... those commands
> to manage the cluster.
No idea. Gaëtan Trellu may know - he created those images.
On Mon, Feb 24, 2020 at 8:08 PM Wesley Hayutin <whayutin@redhat.com> wrote:
On Mon, Feb 24, 2020 at 8:55 AM Mark Goddard <mark@stackhpc.com> wrote:
On Wed, 29 Jan 2020 at 11:31, Alfredo Moralejo Alonso <amoralej@redhat.com> wrote:
On Tue, Jan 28, 2020 at 5:53 PM Mark Goddard <mark@stackhpc.com> wrote:
On Tue, 28 Jan 2020 at 15:18, Mark Goddard <mark@stackhpc.com> wrote:
On Mon, 27 Jan 2020 at 09:18, Radosław Piliszek <radoslaw.piliszek@gmail.com> wrote:
I know it was for masakari. Gaëtan had to grab crmsh from opensuse:
http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen...
-yoctozepto
Thanks Wes for getting this discussion going. I've been looking at CentOS 8 today and trying to assess where we are. I created an Etherpad to track status: https://etherpad.openstack.org/p/kolla-centos8
uwsgi and etcd are now available in rdo dependencies repo. Let me know if you find some issue with it.
I've been working on the backport of kolla CentOS 8 patches to the stable/train branch. It looks like these packages which were added to master are not present in Train.
I'll help check in on that for you Mark. Thank you!!
I've just synced all deps in centos8 master to train, could you check again?
We are seeing an odd DNF error sometimes. DNF exits 141 with no error code when installing packages. It often happens on the rabbitmq and grafana images. There is a prompt about importing GPG keys prior to the error.
Example:
Related bug report? https://github.com/containers/libpod/issues/4431
Anyone familiar with it?
Didn't know about this issue.
BTW, there is rabbitmq-server in RDO dependencies repo if you are interested in using it from there instead of rabbit repo.
pon., 27 sty 2020 o 10:13 Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org> napisał(a): > > W dniu 27.01.2020 o 09:48, Alfredo Moralejo Alonso pisze: > > How is crmsh used in these images?, ha packages included in > > HighAvailability repo in CentOS includes pcs and some crm_*
commands in pcs
> > and pacemaker-cli packages. IMO, tt'd be good to switch to
https://4eff4bb69c321960be39-770d619687de1bce0976465c40e4e9ca.ssl.cf2.rackcd... those commands
> > to manage the cluster. > > No idea. Gaëtan Trellu may know - he created those images. >
On Tue, 25 Feb 2020 at 17:06, Alfredo Moralejo Alonso <amoralej@redhat.com> wrote:
On Mon, Feb 24, 2020 at 8:08 PM Wesley Hayutin <whayutin@redhat.com> wrote:
On Mon, Feb 24, 2020 at 8:55 AM Mark Goddard <mark@stackhpc.com> wrote:
On Wed, 29 Jan 2020 at 11:31, Alfredo Moralejo Alonso <amoralej@redhat.com> wrote:
On Tue, Jan 28, 2020 at 5:53 PM Mark Goddard <mark@stackhpc.com> wrote:
On Tue, 28 Jan 2020 at 15:18, Mark Goddard <mark@stackhpc.com> wrote:
On Mon, 27 Jan 2020 at 09:18, Radosław Piliszek <radoslaw.piliszek@gmail.com> wrote: > > I know it was for masakari. > Gaëtan had to grab crmsh from opensuse: > http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen... > > -yoctozepto
Thanks Wes for getting this discussion going. I've been looking at CentOS 8 today and trying to assess where we are. I created an Etherpad to track status: https://etherpad.openstack.org/p/kolla-centos8
uwsgi and etcd are now available in rdo dependencies repo. Let me know if you find some issue with it.
I've been working on the backport of kolla CentOS 8 patches to the stable/train branch. It looks like these packages which were added to master are not present in Train.
I'll help check in on that for you Mark. Thank you!!
I've just synced all deps in centos8 master to train, could you check again?
Looking good now, thanks.
We are seeing an odd DNF error sometimes. DNF exits 141 with no error code when installing packages. It often happens on the rabbitmq and grafana images. There is a prompt about importing GPG keys prior to the error.
Example: https://4eff4bb69c321960be39-770d619687de1bce0976465c40e4e9ca.ssl.cf2.rackcd...
Related bug report? https://github.com/containers/libpod/issues/4431
Anyone familiar with it?
Didn't know about this issue.
BTW, there is rabbitmq-server in RDO dependencies repo if you are interested in using it from there instead of rabbit repo.
> > pon., 27 sty 2020 o 10:13 Marcin Juszkiewicz > <marcin.juszkiewicz@linaro.org> napisał(a): > > > > W dniu 27.01.2020 o 09:48, Alfredo Moralejo Alonso pisze: > > > How is crmsh used in these images?, ha packages included in > > > HighAvailability repo in CentOS includes pcs and some crm_* commands in pcs > > > and pacemaker-cli packages. IMO, tt'd be good to switch to those commands > > > to manage the cluster. > > > > No idea. Gaëtan Trellu may know - he created those images. > > >
One comment inline, On Fri, Jan 24, 2020 at 8:21 PM Alfredo Moralejo Alonso <amoralej@redhat.com> wrote:
Hi,
We were given access to CBS to build centos8 dependencies a couple of days ago and we are still in the process of re-bootstraping it. I hope we'll have all that is missing in the next days.
See my comments below.
Best regards,
Alfredo
On Fri, Jan 24, 2020 at 7:21 PM Wesley Hayutin <whayutin@redhat.com> wrote:
Greetings,
I know the ceph repo is in progress. TripleO / RDO is not releasing opendaylight
Can the RDO team comment on the rest of the missing packages here please?
Thank you!!
https://review.opendev.org/#/c/699414/9/kolla/image/build.py
NOTE(mgoddard): Mark images with missing dependencies as unbuildable for # CentOS 8. 'centos8': { "barbican-api", # Missing uwsgi-plugin-python3
We'll take care of uwsgi.
"ceph-base", # Missing Ceph repo "cinder-base", # Missing Ceph repo "collectd", # Missing collectd-ping and # collectd-sensubility packages
About collectd and sensu, Matthias already replied from OpsTools side
"elasticsearch", # Missing elasticsearch repo "etcd", # Missing etcd package
Given that etcd is not longer in CentOS base (it was in 7), I guess we'll take care of etcd unless some other sig is building it as part of k8s family.
"fluentd", # Missing td-agent repo
See Matthias reply.
"glance-base", # Missing Ceph repo "gnocchi-base", # Missing Ceph repo "hacluster-base", # Missing hacluster repo
That's an alternative repo for HA related packages for CentOS:
http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/Cen...
Which still does not provide packages for centos8.
Note that centos8.1 includes pacemaker, corosync and pcs in HighAvailability repo. Maybe it could be used instead of the current one.
"ironic-conductor", # Missing shellinabox package
shellinabox is epel. It was never used in tripleo containers, it's really required?
It's a part of an optional ironic feature. TripleO doesn't use it by default [1] so probably fine to remove. However, there may be people using it outside of RH products. [1] https://opendev.org/openstack/tripleo-heat-templates/src/branch/master/deplo...
"kibana", # Missing elasticsearch repo
We never provided elasticsearch in the past, is consumed from elasticsearch repo iirc
"manila-share", # Missing Ceph repo "mongodb", # Missing mongodb and mongodb-server packages
Mongodb was retired from RDO time ago as it was not longer the recommended backend for any service. In CentOS7 is pulled from EPEL.
"monasca-grafana", # Using python2 "nova-compute", # Missing Ceph repo "nova-libvirt", # Missing Ceph repo "nova-spicehtml5proxy", # Missing spicehtml5 package
spice-html5 is pulled from epel7 was never part of RDO. Not used in TripleO.
"opendaylight", # Missing opendaylight repo "ovsdpdk", # Not supported on CentOS "sensu-base", # Missing sensu package
See Matthias reply.
"tgtd", # Not supported on CentOS 8
tgtd was replace by scsi-target-utils. It's was never provided in RDO, in kolla was pulled from epel for 7
},
'centos8+source': { "barbican-base", # Missing uwsgi-plugin-python3 "bifrost-base", # Bifrost does not support CentOS 8 "cyborg-agent", # opae-sdk does not support CentOS 8 "freezer-base", # Missing package trickle "masakari-monitors", # Missing hacluster repo "zun-compute", # Missing Ceph repo _______________________________________________ dev mailing list dev@lists.rdoproject.org http://lists.rdoproject.org/mailman/listinfo/dev
To unsubscribe: dev-unsubscribe@lists.rdoproject.org
_______________________________________________ dev mailing list dev@lists.rdoproject.org http://lists.rdoproject.org/mailman/listinfo/dev
To unsubscribe: dev-unsubscribe@lists.rdoproject.org
Greetings,
I know the ceph repo is in progress.
On 1/24/20 7:20 PM, Wesley Hayutin wrote: the ceph [1] package and storage8-centos-nautilus-* repo [2] are now done not installable yet because we can't build the centos-release-ceph-nautilus package for el8 yet, but we're trying to solve that in centos-devel [3] 1. https://cbs.centos.org/koji/buildinfo?buildID=28564 2. https://cbs.centos.org/koji/builds?tagID=1891 3. https://lists.centos.org/pipermail/centos-devel/2020-February/036544.html -- Giulio Fidente GPG KEY: 08D733BA
participants (9)
-
Alfredo Moralejo Alonso
-
Dmitry Tantsur
-
Giulio Fidente
-
Marcin Juszkiewicz
-
Mark Goddard
-
Matthias Runge
-
Neal Gompa
-
Radosław Piliszek
-
Wesley Hayutin