bug#49781: Successful build on ci.guix not available as substitute

2021-08-10 Thread Ludovic Courtès
Hi, Mathieu Othacehe skribis: > I found out the reason of this failure, the .185 key was not authorized > on Berlin. This issue while really simple, took us a long time to > detect. I think we need to come out with a big warning somewhere when a > substitute is refused because the server is not

bug#49781: Successful build on ci.guix not available as substitute

2021-08-09 Thread Mathieu Othacehe
Hey Ludo! I found out the reason of this failure, the .185 key was not authorized on Berlin. This issue while really simple, took us a long time to detect. I think we need to come out with a big warning somewhere when a substitute is refused because the server is not authorized. In the

bug#49781: Successful build on ci.guix not available as substitute

2021-08-02 Thread Ludovic Courtès
Hello! Mathieu Othacehe skribis: > Your understanding is correct :). However, looking at the > cuirass-remote-server.log file, I found: > > 2021-07-29T22:31:48 fetching > '/gnu/store/hrn9hj7v1img3il6afh3r036aczvwnyd-xnnpack-0.0-1.bbe8824.drv' from > http://141.80.167.185:5558 >

bug#49781: Successful build on ci.guix not available as substitute

2021-07-30 Thread Mathieu Othacehe
Hello Ludo, > It’s the first time I see this. My understanding is that Cuirass makes > a GET request on ‘guix publish’ upon build completion. Could there be a > race condition or something that can explain this? Your understanding is correct :). However, looking at the

bug#49781: Successful build on ci.guix not available as substitute

2021-07-30 Thread Ludovic Courtès
Hi! I stumbled upon the case of a build that was successful on ci.guix but that does not have a corresponding substitute: https://ci.guix.gnu.org/build/806609/details and: --8<---cut here---start->8--- $ wget -O-