=== maclin1 is now known as maclin === micahg_ is now known as micahg === cpaelzer_ is now known as cpaelzer === cpaelzer is now known as cpaelzer_ === cpaelzer_ is now known as cpaelzer === cpaelzer is now known as cpaelzer_ === cpaelzer_ is now known as cpaelzer === jamesh_ is now known as jamesh === awayney is now known as Laney [10:27] jamespage: alembic is stuck in -proposed due to component mismatch [10:27] doko: ah - yes - noted last week on my list to raise a mir/review that for a delta [10:28] doko: just had some pxc security updates to clear first (now ready for security team review) [10:33] doko: managed to drop a bit of required delta - fixing that right now === niedbalski_ is now known as niedbalski [14:04] when run in autopkgtest I see a set of "tr: write error: Broken pipe" but when logging into the very same guest and runnign again it does not show up [14:05] does the autopkgtest execution have soemthing that comes to your mind that could cause this? [14:05] my login is ssh as reported by autopkgtest on --shel--fail [14:07] I'm trying to debug the tests atm, but lacking a way to reproduce this is hard, so if this might be a known issue to anybody let me know [15:52] if anybody wondered about my question above - I have the solution - shell fragments like the following are the reason [15:52] tr -dc "[:alpha:]" < /dev/urandom | head -c 6 [15:53] common at least on search engine hits (https://gist.github.com/paulomcnally/7209bad52c724d19bdd029b4a6631265) and were used in a packages test [15:53] but the tl;dr the first read is more or less unlimited, which might eventually cause the issue [15:53] something like "head -c 100 /dev/urandom | tr -dc '[:alpha:]' | head -c 6" works better [17:34] cpaelzer: the downside to the second one is that there's a (very small) chance that it won't succeed once in a while [17:35] sarnold: it might be shorter [17:35] right [17:35] sarnold: but I doubt it will be zero [17:35] shorter is fine for me [17:35] oh okay [17:35] sometimes shorter is catastrophic :) [17:35] and one can increase the first number to be "more safe" [17:35] hehe [19:02] !dmb-ping [19:02] bdmurray, BenC, cyphermox, jbicha, micahg, rbasak, sil2100: DMB ping. [20:53] doko: thanks; I've merged your debian/control fixes upstream === led_ir23 is now known as led_ir22