[02:51] ppa.lp.net going down in 10 mins for quick maintenance [03:00] ppa.lp.net donw for maintenance [03:07] done [07:24] cjwatson: wgrant. (others?): the builders again are soemwhat stuck [07:24] cpaelzer: Looking [07:24] actually more than I've seen every before so maybe this is not one of the usual "please bump" [07:25] but I'll leave that to you [07:25] cpaelzer: Looks okay to me, what's the issue you're seeing? [07:25] really, let me make sure I refreshed that ... [07:25] https://launchpad.net/builders [07:25] well I assumed that 6 day queues are the same usual issue [07:25] but maybe ther just is that much load? [07:26] oh I see [07:26] "ubuntu/test-rebuild-20190614" [07:26] cpaelzer: It's a test rebuild. [07:26] But normal builds will be scheduled in front of it. [07:26] If you have a particular stuck build we can investigate. [07:27] I'll give it a few minutes for the prios to kick in on the next finish [07:27] should be fine [07:27] and I'll not which builders are in "cleaning" as sometimes they seem to not recover [07:27] note to myself atm 4-7, 9, 11, 13, 17-19 [07:28] no I think we are good wgrant [07:28] sorry for the noise [07:28] We track the cleanings fairly closely during this sort of event [07:28] I assumed on the long queues we might again be stuck on cleanup, but we arent [07:28] Heh that would be pretty impressive [08:45] livefs build seems 'stuck'? https://launchpad.net/~ubuntu-cdimage/+livefs/ubuntu/eoan/kubuntu/+build/170391 [08:45] same for ubuntu https://launchpad.net/~ubuntu-cdimage/+livefs/ubuntu/eoan/ubuntu/+build/170405 [08:46] buildd-manager seems to be working happily enough [08:47] Though the job apparently finished (according to its logs) quite a few hours ago and it hasn't finished fetching it, so might be more subtly stuck [08:48] Anyway, I'm off today so can't look [08:59] cjwatson: thanks. I just looked at yesterdays livefs, and the log says it took about 15mins, but the LP page says 'took 6 hours, 15 minutes'. the full cd-build-logs also show there was that ~6hr hang [09:00] so something is amiss, but hopefully will complete after a delay === apw_ is now known as apw [15:19] How do I reopen an "Solved" answer? [15:58] hi, need help with an account issue [15:59] my uploads are getting associated with an incorrect account - 'brmce' instead of 'bryce' [15:59] If you look at https://launchpad.net/ubuntu/+source/exim4/4.92-7ubuntu1 you can see Uploaded by is linking to https://launchpad.net/~brmce instead of https://launchpad.net/~bryce [16:00] 'brmce' doesn't appear anywhere in my local configs nor in the .changes or .dsc I used to upload, so I can't spot where this misconnection occurred [16:01] it is also present on a recent dovecot upload and I believe on logwatch as well, so is not something package-specific === zbr is now known as zbr|ruck [21:40] bryce: The link occurs by email address. See the link on https://launchpad.net/~brmce to merge the account into your main one and add the email address to your account. [21:46] wgrant, alright thanks [21:47] bryce: The issue is you don't have canonical.com addresses on your main LP account [21:47] So a new placeholder was created when LP saw you upload with it again [21:48] huh, interesting. I take it it just randomly changes a char in the firstname or something? [21:49] bryce: It normally tries to append innocuous characters first, but those must have collided too [21:49] Permuting a random character is close to a last resort [21:49] iirc [21:50] I should go ahead and add bryce.harrington@canonical.com while I'm at it [21:51] Indeed. [21:51] I see the merge has worked though [21:51] You should also remove any obsolete OpenPGP or SSH keys [21:51] e.g. there's a couple of 1024Ds there [21:51] ah yes, will do [21:52] bryce: Also revoke any old clients from https://dogfood.paddev.net/%7Ebryce/+oauth-tokens [21:52] er [21:52] https://launchpad.net/%7Ebryce/+oauth-tokens [21:53] 'dogfood' - fun names for instances heh. :) [21:54] wgrant, ok that's quite a list, but think most all of those are obsolete. I'll cull it down. Thanks! [22:00] "Permuting a random character is close to a last resort" [22:00] Wow! [22:00] I was convinced it was caused by a typo because of that permutation :) [22:06] Nope [22:06] * wgrant finds the code [22:09] https://git.launchpad.net/launchpad/tree/lib/lp/registry/model/person.py#n4298 [22:11] pity attempts isn't logged :) [22:12] I like the exception. "some twonk" [22:12] sarnold: So funny story [22:13] bryce: hah, I overlooked that :) [22:13] LOL xD [22:13] must read that exception xD [22:13] just* [22:14] sarnold: Originally generate_nick was run on the whole email address so you ended up with something like foo-gmail. But that revealed too much of the email address, so it was changed to just use the local part (so foo). [22:14] But of course many local parts collide [22:14] And it seeded the RNG from only the localpart [22:14] So we ended up with hundreds of deterministic mutations of "lp" [22:14] ohhhhhh nooo [22:14] We had usernames that were more than 200 characters long as a result of it [22:14] "why does it take ten minutes to create a new user account?" [22:14] hah [22:15] They all look very random [22:15] "mine just created in seconds" [22:15] 78luphr0rnk2nuqimstywepozxn9kl19tqh0tx66b5dki1xx-launchpad-a811i2i3ytqlsztthjth0svbccw8inm65tmkqp9sarr553jq [22:15] 78luphr0rnk2nuqimstywepozxn9kl19tqh0tx66b5dki1xxs-phn5hho65-a811i2i3ytqlsztthjth0svbccw8inm65tmkqp9sarr553jq5 [22:15] random, but suspiciously similar [22:15] 78luphr0rnk2nuqimstywepozxn9kl19tqh0tx66b5dki1xxsh5mkz9gl21a5rlwfnr-launchpad-a811i2i3ytqlsztthjth0svbccw8inm65tmkqp9sarr553jq53in4xm1m8wn3o4rlwa [22:15] Was very confusing before we worked it out :) [22:15] yeah, I can imagine it wasn't real obvious :) [22:16] did you wind up swapping out all those user names? those urls would be pretty unwieldy to use, but they aren't needed often.. [22:16] how many needed changing? [22:16] We didn't go back and fix them. [22:17] There are maybe a thousand past 100 characters, not many. [22:18] no complaints? :) [22:18] People could always change their username [22:18] So anyone who cared had already fixed it [22:19] oh! I never noticed that :) [22:20] (or maybe I did, is that the one where you can't change your name if you've got PPAs? [22:20] It's locked while you have live PPAs [22:20] Yeah [23:04] wgrant: is there a bug about longer-than-4digits CVE codes? [23:04] wrt bugs and CVE linking [23:06] teward: They work fine [23:06] AFAIK [23:07] wgrant: they don't [23:07] 2019-12816 <-- "Invalid format" - unless it needs CVE- before it? [23:07] teward: you can paste long cves into a comment and launchpad will scrape them out of it [23:07] or is that a "CVE Not Listed" issue? [23:07] 2019-12816 is not a known CVE sequence number. <-- this is what i hit [23:08] sarnold: oh good you're alive [23:08] you can answer the question in -hardened nwo :) [23:08] now* [23:09] That sounds like it wasn't in the last dump we imported from Mitre [23:09] LP's supported up to 7 digits since 2014 from what I can see [23:11] ah. [23:11] wgrant: when was the last import? [23:12] 2019-06-17 08:48:40 ERROR Unable to connect for CVE database https://cve.mitre.org/data/downloads/allitems.xml.gz [23:12] I wonder why [23:14] curious indeed. I can scrape it from my home machine [23:15] Yeah, probably an internal proxy issue