I am assuming this is autosigning related

More information... may be specific to this package (or some subset of packages). Builds of other packages are making it to f27 tag, just not this one, even a new rebuild today:
https://koji.fedoraproject.org/koji/buildinfo?buildID=873832

For full disclosure, in case it matters (I doubt, but...), both of these builds were submitted via 'fedpkg chain-build ...' rather than just 'fedpkg build'

I triggered autosign again on the -7 build (by untagging it from f27-pending and then re-tagging it), and it processed just fine.

I think we need @puiterwijk to look at the robosign logs and see if he can see why it failed the first time.

Signing appears to be stuck or delayed again (over 25 minutes since build completed, still in f27-pending):

PyQt4 : https://koji.fedoraproject.org/koji/buildinfo?buildID=874206

qt5-qtbase : https://koji.fedoraproject.org/koji/buildinfo?buildID=874077

yes. I am trying to get it to sign texlive (reported on devel list), so it's doing that and not (yet) processing other things.

Sorry for the delay, but I am trying to find out why it didn't complete texlive correctly...

rex makes kirk-wrath-of-kahn-face: "TexLive!!!"

yes. I am trying to get it to sign texlive (reported on devel list), so it's doing that and not (yet) processing other things.

I believe it's a disk space issue on the signing bridge (maybe vault) because robosignatory signs a NVR in a single batch and the likes of texlive are large and might not fit on the local disk.

Yeah, but there's no sign of that in the logs. So, not sure.

Patrick is going to implement a backoff where if it cannot sign the entire package at once it will try a small batch size. Also, after freeze we are going to redo the bridge with a larger disk.

In the mean time I am manually signing texlive... and other stuff is being signed again (I'm using a batch size of 1 so other builds can be signed).

closing as there has been no updates in a long time