[16:14:44] can someone who knows more about our exim and mail aliases stuff check on: https://gerrit.wikimedia.org/r/c/operations/puppet/+/842499 ? [16:15:03] (new main project domain, we need hostmaster@ and such to work for TLS cert verification) [16:15:19] you probably want jhathaway: ^^^ [16:15:23] that patch was my naive attempt, but I kind of assume there could be more to it :) [16:16:06] bblack: I can take a look [16:16:40] thanks! [17:28:26] https://signal.org/blog/sms-removal-android/ ugh, annoying as it's been my default SMS for a while :P [17:40:59] I finally switched to Google Messages because of it :( based on https://community.signalusers.org/t/signal-blog-removing-sms-support-from-signal-android-soon/47954/57 it seems like all third-party SMS apps are doomed [17:42:36] If Signal was community open source rather than open source for transparency but uninterested in community contributions maybe something could be done... [17:43:27] I get their reasoning, but I feel like there were other ways to handle it. Like, just have an annoying dark red highlight on all insecure SMS conversations or whatever. [17:43:41] (if it's about confusion and/or future breakage) [17:44:09] I haven't had the RCS issue, but I guess if that's the future there's not much other choice [17:44:10] some additional info at https://community.signalusers.org/t/signal-blog-removing-sms-support-from-signal-android-soon/47954/57 [17:44:26] looks like its all a mess [17:44:56] anyways, I don't even have an 'export SMS' option yet, I guess that's coming in a future update [17:46:17] TIL that "RCS" is Google's attempt at an iMessage competitor [17:47:17] yeah basically [17:47:43] I think it was supposed to become a new standard for all, but the iPeople didn't want to play ball (probably for good reasons) [17:47:57] this is why we can't have nice things anymore :P [17:49:02] I would guess reason #1 would be something like "how does this help us?" and google laughed :) [17:51:04] yeah but the universality of SMS was nice. it's a shame we couldn't build new things without fragmenting [19:18:51] jbond: should systemd::override be passing override => true to systemd::unit? [19:23:36] ori: yes it should :) sending a patch [19:25:56] I have a couple rsync servers and somehow they are all set to read_only now. But I don't recall doing that. Did we make a change there by any chance [19:28:00] ori: 842523 merged [19:29:23] Thank you [21:34:12] OSError: [Errno 28] No space left on device: '/srv/jenkins/puppet-compiler/37548' [21:34:16] we were seeing this before, right? [23:15:36] cdanis: we get that every now and again, there are some clean up timers but if pcc is run with lots of hosts (e.g. no Hosts: footer) then the dir can fill up pretty quickly, i can take a look tomorrow but i genrally just run the timer with a lower mtime (on the pcc-worker host)