Re: ceph / boost broken in koji Rawhide

2017-02-13 Thread Richard W.M. Jones
On Sun, Feb 12, 2017 at 01:45:53PM +0100, Kevin Kofler wrote: > Ken Dreyer wrote: > > My money's on ceph-debuginfo's horrible size. We have to bump HTTP > > server resource limits upstream to handle this, eg. > > https://github.com/ceph/chacra/issues/130 . > > WTF, how does ceph-debuginfo manage t

Re: ceph / boost broken in koji Rawhide

2017-02-12 Thread Kevin Kofler
Kevin Fenzi wrote: > I dont think it's size... we have texlive to keep us on our toes about > that (the texlive src.rpm is ~2.1GB. texlive-debuginfo is only 28 434 646 bytes. Most of the huge SRPM is noarch texmf content that does not produce any debugging information. qt5-qtwebengine-debuginfo

Re: ceph / boost broken in koji Rawhide

2017-02-12 Thread Kevin Fenzi
On Sat, 11 Feb 2017 13:41:33 -0700 Ken Dreyer wrote: > On Fri, Feb 10, 2017 at 8:49 AM, Kevin Fenzi wrote: > > So, today I just manually signed it and retagged it into f26. > > So, it should be in there now... sorry for the hassles. > > > > We will look at logs on the autosigner and see if we ca

Re: ceph / boost broken in koji Rawhide

2017-02-12 Thread Kevin Kofler
Ken Dreyer wrote: > My money's on ceph-debuginfo's horrible size. We have to bump HTTP > server resource limits upstream to handle this, eg. > https://github.com/ceph/chacra/issues/130 . WTF, how does ceph-debuginfo manage to be twice as large as even qt5- qtwebengine-debuginfo (which I thought wa

Re: ceph / boost broken in koji Rawhide

2017-02-11 Thread Ken Dreyer
On Fri, Feb 10, 2017 at 8:49 AM, Kevin Fenzi wrote: > So, today I just manually signed it and retagged it into f26. > So, it should be in there now... sorry for the hassles. > > We will look at logs on the autosigner and see if we can figure out > what happened. My money's on ceph-debuginfo's hor

Re: ceph / boost broken in koji Rawhide

2017-02-10 Thread Richard W.M. Jones
On Fri, Feb 10, 2017 at 08:49:40AM -0700, Kevin Fenzi wrote: > Something went wrong with signing this package and we aren't sure what > yet. When I looked at it yesterday it looked like it was signing part > of it, but then it didn't seem to progress. I also tried retagging > things so the autosign

Re: ceph / boost broken in koji Rawhide

2017-02-10 Thread Jonathan Wakely
On 10/02/17 12:04 +, Richard W.M. Jones wrote: On Fri, Feb 10, 2017 at 11:49:10AM +, Jonathan Wakely wrote: On 10/02/17 10:27 +, Richard W.M. Jones wrote: >On Fri, Feb 10, 2017 at 10:30:51AM +0100, Vít Ondruch wrote: >>Because all packages for Rawhide are signed now? Not sure where/h

Re: ceph / boost broken in koji Rawhide

2017-02-10 Thread Kevin Fenzi
On Fri, 10 Feb 2017 12:04:50 + "Richard W.M. Jones" wrote: > On Fri, Feb 10, 2017 at 11:49:10AM +, Jonathan Wakely wrote: > > On 10/02/17 10:27 +, Richard W.M. Jones wrote: > > >On Fri, Feb 10, 2017 at 10:30:51AM +0100, Vít Ondruch wrote: > > >>Because all packages for Rawhide are

Re: ceph / boost broken in koji Rawhide

2017-02-10 Thread Richard W.M. Jones
On Fri, Feb 10, 2017 at 11:49:10AM +, Jonathan Wakely wrote: > On 10/02/17 10:27 +, Richard W.M. Jones wrote: > >On Fri, Feb 10, 2017 at 10:30:51AM +0100, Vít Ondruch wrote: > >>Because all packages for Rawhide are signed now? Not sure where/how this > >>was communicated ... > > > >Can this

Re: ceph / boost broken in koji Rawhide

2017-02-10 Thread Jonathan Wakely
On 10/02/17 10:27 +, Richard W.M. Jones wrote: On Fri, Feb 10, 2017 at 10:30:51AM +0100, Vít Ondruch wrote: Because all packages for Rawhide are signed now? Not sure where/how this was communicated ... Can this be short-circuited for packages going into Koji? It seems like this will great

Re: ceph / boost broken in koji Rawhide

2017-02-10 Thread Pierre-Yves Chibon
On Fri, Feb 10, 2017 at 10:27:53AM +, Richard W.M. Jones wrote: > On Fri, Feb 10, 2017 at 10:30:51AM +0100, Vít Ondruch wrote: > > > > > > Dne 10.2.2017 v 10:06 Richard W.M. Jones napsal(a): > > > On Fri, Feb 10, 2017 at 10:02:11AM +0100, Vít Ondruch wrote: > > >> Yesterday around 4pm UTC +-

Re: ceph / boost broken in koji Rawhide

2017-02-10 Thread Richard W.M. Jones
On Fri, Feb 10, 2017 at 10:30:51AM +0100, Vít Ondruch wrote: > > > Dne 10.2.2017 v 10:06 Richard W.M. Jones napsal(a): > > On Fri, Feb 10, 2017 at 10:02:11AM +0100, Vít Ondruch wrote: > >> Yesterday around 4pm UTC +- I was told: > >> > >>> nirikit's currently trying to sign the 1.1GB ceph-deb

Re: ceph / boost broken in koji Rawhide

2017-02-10 Thread Vít Ondruch
Dne 10.2.2017 v 10:06 Richard W.M. Jones napsal(a): > On Fri, Feb 10, 2017 at 10:02:11AM +0100, Vít Ondruch wrote: >> Yesterday around 4pm UTC +- I was told: >> >>> nirikit's currently trying to sign the 1.1GB ceph-debuginfo for >> aarch64 >> >> but I'd say it is stuck somewhere > But th

Re: ceph / boost broken in koji Rawhide

2017-02-10 Thread Richard W.M. Jones
On Fri, Feb 10, 2017 at 10:02:11AM +0100, Vít Ondruch wrote: > Yesterday around 4pm UTC +- I was told: > > > nirikit's currently trying to sign the 1.1GB ceph-debuginfo for > aarch64 > > but I'd say it is stuck somewhere But this is for Koji builds. Why do the packages need to be signe

Re: ceph / boost broken in koji Rawhide

2017-02-10 Thread Vít Ondruch
Yesterday around 4pm UTC +- I was told: > nirikit's currently trying to sign the 1.1GB ceph-debuginfo for aarch64 but I'd say it is stuck somewhere V. Dne 10.2.2017 v 09:41 Richard W.M. Jones napsal(a): > https://koji.fedoraproject.org/koji/taskinfo?taskID=17707485 > > The error is

ceph / boost broken in koji Rawhide

2017-02-10 Thread Richard W.M. Jones
https://koji.fedoraproject.org/koji/taskinfo?taskID=17707485 The error is: DEBUG util.py:435: Error: nothing provides libboost_system.so.1.60.0()(64bit) needed by librbd1-1:10.2.5-1.fc26.x86_64 This is fixed by a new ceph build which was completed yesterday morning: https://koji.fedora