[libvirt] [PATCH libvirt-cim 0/2] Small improvements
The first patch will make it possible to have a nicer CI setup. The second one is just something that I couldn't help fixing while preparing the first one :) Andrea Bolognani (2): maint: Rename autoconfiscate.sh to autogen.sh README: Point to git repository README | 9 ++++++--- autoconfiscate.sh => autogen.sh | 0 2 files changed, 6 insertions(+), 3 deletions(-) rename autoconfiscate.sh => autogen.sh (100%) -- 2.13.5
While the "autoconfiscate" name is very clever and cute, the de-facto standard name for this kind of script is "autogen", and deviating from it means having to special-case the libvirt-cim project when, for example, setting up a CI environment. Signed-off-by: Andrea Bolognani <abologna@redhat.com> --- README | 2 +- autoconfiscate.sh => autogen.sh | 0 2 files changed, 1 insertion(+), 1 deletion(-) rename autoconfiscate.sh => autogen.sh (100%) diff --git a/README b/README index 8b6909c..d3e9172 100644 --- a/README +++ b/README @@ -30,7 +30,7 @@ Requirements: Procedure: -If building a tree cloned from the hg repository, run this step first: - $ ./autoconfiscate.sh + $ ./autogen.sh $ ./configure $ make diff --git a/autoconfiscate.sh b/autogen.sh similarity index 100% rename from autoconfiscate.sh rename to autogen.sh -- 2.13.5
On 09/22/2017 09:28 AM, Andrea Bolognani wrote:
While the "autoconfiscate" name is very clever and cute, the de-facto standard name for this kind of script is "autogen", and deviating from it means having to special-case the libvirt-cim project when, for example, setting up a CI environment.
Signed-off-by: Andrea Bolognani <abologna@redhat.com> --- README | 2 +- autoconfiscate.sh => autogen.sh | 0 2 files changed, 1 insertion(+), 1 deletion(-) rename autoconfiscate.sh => autogen.sh (100%)
libvirt-cim has it's own mailing list (libvirt-cim@redhat.com) and achives: https://www.redhat.com/archives/libvirt-cim/index.html. The last email there (april 2015) and last patch (aug 2014) - I venture to say perhaps the "doesn't anyone really care" starts to factor in. Then of course, is there a way (need/desire) to kill it completely? John
diff --git a/README b/README index 8b6909c..d3e9172 100644 --- a/README +++ b/README @@ -30,7 +30,7 @@ Requirements: Procedure:
-If building a tree cloned from the hg repository, run this step first: - $ ./autoconfiscate.sh + $ ./autogen.sh
$ ./configure $ make diff --git a/autoconfiscate.sh b/autogen.sh similarity index 100% rename from autoconfiscate.sh rename to autogen.sh
On Fri, 2017-09-22 at 10:08 -0400, John Ferlan wrote:
libvirt-cim has it's own mailing list (libvirt-cim@redhat.com) and achives: https://www.redhat.com/archives/libvirt-cim/index.html.
Fair enough, my bad for not noticing.
The last email there (april 2015) and last patch (aug 2014) - I venture to say perhaps the "doesn't anyone really care" starts to factor in. Then of course, is there a way (need/desire) to kill it completely?
It's telling that the last patch that made it to libvirt-cim's master was posted on libvir-list[1]. In that same thread, Dan suggested closing the libvirt-cim mailing list altogether[2]. The same day, someone noticed that the README was pointing to a non-existing Mercurial repository[3]: a fact that, incidentally, I address with patch 2/2. ... Not even a mere 29 months later :) [1] https://www.redhat.com/archives/libvir-list/2015-April/msg01446.html [2] https://www.redhat.com/archives/libvir-list/2015-April/msg01456.html [3] https://www.redhat.com/archives/libvir-list/2015-April/msg01453.html -- Andrea Bolognani / Red Hat / Virtualization
On Fri, Sep 22, 2017 at 10:08:44AM -0400, John Ferlan wrote:
On 09/22/2017 09:28 AM, Andrea Bolognani wrote:
While the "autoconfiscate" name is very clever and cute, the de-facto standard name for this kind of script is "autogen", and deviating from it means having to special-case the libvirt-cim project when, for example, setting up a CI environment.
Signed-off-by: Andrea Bolognani <abologna@redhat.com> --- README | 2 +- autoconfiscate.sh => autogen.sh | 0 2 files changed, 1 insertion(+), 1 deletion(-) rename autoconfiscate.sh => autogen.sh (100%)
libvirt-cim has it's own mailing list (libvirt-cim@redhat.com) and achives: https://www.redhat.com/archives/libvirt-cim/index.html.
The last email there (april 2015) and last patch (aug 2014) - I venture to say perhaps the "doesn't anyone really care" starts to factor in. Then of course, is there a way (need/desire) to kill it completely?
IBM was the only company that ever cared about CIM for virtualization. Everyone else wanted an API that was easy to use instead ;-P Even IBM has lost interest now, so it is mostly a historical curiosity at this point. I wouldn't delete it, but also wouldn't spend alot of time in it. If someone does have a desire to update the code in any way, go at it and we can review it in the normal manner - no point waiting for the original maintainers to review IMHO. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
On Fri, 2017-09-22 at 17:17 +0100, Daniel P. Berrange wrote:
libvirt-cim has it's own mailing list (libvirt-cim@redhat.com) and achives: https://www.redhat.com/archives/libvirt-cim/index.html.
The last email there (april 2015) and last patch (aug 2014) - I venture to say perhaps the "doesn't anyone really care" starts to factor in. Then of course, is there a way (need/desire) to kill it completely?
IBM was the only company that ever cared about CIM for virtualization. Everyone else wanted an API that was easy to use instead ;-P Even IBM has lost interest now, so it is mostly a historical curiosity at this point. I wouldn't delete it, but also wouldn't spend alot of time in it. If someone does have a desire to update the code in any way, go at it and we can review it in the normal manner - no point waiting for the original maintainers to review IMHO.
I'm pretty sure John was referring to the mailing list, not the project itself :) -- Andrea Bolognani / Red Hat / Virtualization
On 09/22/2017 12:47 PM, Andrea Bolognani wrote:
On Fri, 2017-09-22 at 17:17 +0100, Daniel P. Berrange wrote:
libvirt-cim has it's own mailing list (libvirt-cim@redhat.com) and achives: https://www.redhat.com/archives/libvirt-cim/index.html.
The last email there (april 2015) and last patch (aug 2014) - I venture to say perhaps the "doesn't anyone really care" starts to factor in. Then of course, is there a way (need/desire) to kill it completely?
IBM was the only company that ever cared about CIM for virtualization. Everyone else wanted an API that was easy to use instead ;-P Even IBM has lost interest now, so it is mostly a historical curiosity at this point. I wouldn't delete it, but also wouldn't spend alot of time in it. If someone does have a desire to update the code in any way, go at it and we can review it in the normal manner - no point waiting for the original maintainers to review IMHO.
I'm pretty sure John was referring to the mailing list, not the project itself :)
I had forgotten about Daniel's last email on about libvirt-cim on libvir-list... By killing it - I was thinking more along the lines of removing it from our CI infrastructure and just letting the repo sit silently without updates, but that ship sailed today. We could still remove it from the CI mix as if something breaks - there's not going to be anyone that really wants to fix it. There's also something called cimtest that should have the same luxurious retirement. No one seems to care any more and it's woefully behind at this point. The main reason I was the stuckee was that I came from HP and agreed that I knew how to spell CIM because HP had a passing interesting in CIM along with IBM. John
On Fri, 2017-09-22 at 12:58 -0400, John Ferlan wrote:
By killing it - I was thinking more along the lines of removing it from our CI infrastructure and just letting the repo sit silently without updates, but that ship sailed today. We could still remove it from the CI mix as if something breaks - there's not going to be anyone that really wants to fix it.
Well, there hasn't been a single release since 2013, and any actual development since 2014... Maybe it's really time to move on and free up some cycles for the CI workers, especially since there are going to be more jobs running on there pretty soon. -- Andrea Bolognani / Red Hat / Virtualization
On Fri, Sep 22, 2017 at 07:27:33PM +0200, Andrea Bolognani wrote:
On Fri, 2017-09-22 at 12:58 -0400, John Ferlan wrote:
By killing it - I was thinking more along the lines of removing it from our CI infrastructure and just letting the repo sit silently without updates, but that ship sailed today. We could still remove it from the CI mix as if something breaks - there's not going to be anyone that really wants to fix it.
Well, there hasn't been a single release since 2013, and any actual development since 2014...
Maybe it's really time to move on and free up some cycles for the CI workers, especially since there are going to be more jobs running on there pretty soon.
Time consumed by the CI job is < 2% of total time required to do a full rebuild when libvirt sees a git change. Almost every other module takes a greater amount of time. This drops still further when you add in the time consumed by changes to non-libvirt core modules which don't trigger the CIM binding. Realisticaly the only thing that will have a measurable improvement on our CI system at this point is obtaining more hardware. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
The Mercurial repositories are no more, so update the pointer. Signed-off-by: Andrea Bolognani <abologna@redhat.com> --- README | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/README b/README index d3e9172..09e0059 100644 --- a/README +++ b/README @@ -9,7 +9,7 @@ platforms with a single provider. The latest version of the library can be found here: - http://libvirt.org/hg/libvirt-cim + https://libvirt.org/git/?p=libvirt-cim.git and the development discussion mailing list is here: @@ -29,8 +29,11 @@ Requirements: Procedure: - -If building a tree cloned from the hg repository, run this step first: - $ ./autogen.sh + # If building a tree cloned from the git repository, run this step first: + + $ ./autogen.sh + + # Then, regardless of how you obtained the source, run: $ ./configure $ make -- 2.13.5
On 09/22/2017 03:28 PM, Andrea Bolognani wrote:
The first patch will make it possible to have a nicer CI setup.
The second one is just something that I couldn't help fixing while preparing the first one :)
Andrea Bolognani (2): maint: Rename autoconfiscate.sh to autogen.sh README: Point to git repository
README | 9 ++++++--- autoconfiscate.sh => autogen.sh | 0 2 files changed, 6 insertions(+), 3 deletions(-) rename autoconfiscate.sh => autogen.sh (100%)
ACK Michal
participants (4)
-
Andrea Bolognani -
Daniel P. Berrange -
John Ferlan -
Michal Privoznik