2012-07-01T00:10:27 *** ZC|Mobile has quit IRC (Ping timeout: 265 seconds) 2012-07-01T00:19:29 *** rangerz has joined #se2600 2012-07-01T00:25:57 *** ZC|Mobile has joined #se2600 2012-07-01T00:31:43 *** rangerz has quit IRC (Quit: Leaving.) 2012-07-01T00:50:52 Dagmar: apparently java is coughing up hairballs in some circumstances with the leap second addition. 2012-07-01T00:51:06 we were discussing that earlier 2012-07-01T00:51:21 that's why my minecraftbot joined and left a few times 2012-07-01T00:51:32 Of course... oracraple cluster services is randomly rebooting cluster nodes from reports as well. 2012-07-01T00:51:43 Sorry, but I find that quite amusing. 2012-07-01T00:56:04 *** MinecraftRelay has quit IRC () 2012-07-01T00:57:07 I'll tell ya what's awesome is that I haven't gotten a page yet 2012-07-01T00:58:26 Damn I'm going to have to wait until NotLarry's kids show up playing TF2 somewhere 2012-07-01T00:58:55 Three servers I've been on now, none of them with me having a *clue* what was going on 2012-07-01T00:59:49 https://bugzilla.mozilla.org/show_bug.cgi?id=769972 2012-07-01T00:59:51 Title: 769972 Java is choking on leap second. (at bugzilla.mozilla.org) 2012-07-01T00:59:58 I find this amusing, I really do. 2012-07-01T01:00:34 It's kinda hilarious, yes 2012-07-01T01:00:49 There've been two responses to this so far... 2012-07-01T01:01:01 1. Reboot the shit out of everything, get on with day. 2012-07-01T01:01:03 2. Panic. 2012-07-01T01:01:21 updating mods now 2012-07-01T01:01:53 *** MinecraftRelay has joined #se2600 2012-07-01T01:02:16 1. Could be written as "Pretend it's windows, reboot the shit out of everything, get on with day." 2012-07-01T01:02:28 Well, I mean, it's a _Saturday_ 2012-07-01T01:03:01 Most operational admins boil it down to "Do we know why it happened? Yes." "Can we make the problem go away? Yes." 2012-07-01T01:03:03 I just pulled a random check of my internal infra and I have 2 boxes that didn't insert the leap second. Additional checks show 1 remote host that didn't either. I'm afraid to check serentos infra. 2012-07-01T01:03:12 "Is it going to happen again? No." 2012-07-01T01:03:15 REboot that shiot. 2012-07-01T01:04:13 ...and at this point anything that was going to fall over and die hasn't so... 2012-07-01T01:04:55 Yep. I'm just curious why these particular boxes didn't act on the leapadd. 2012-07-01T01:06:57 Well, it may actually have been a race condition of some kind 2012-07-01T01:07:06 I mean, this wasn't _supposed_ to cause a problem 2012-07-01T01:07:35 Contiguous time is something the unix world has had a handle on for a long goddamn time now 2012-07-01T01:07:48 *NORMALLY* when the clock changes, ntpd just fixes it 2012-07-01T01:08:17 We acutally *could* have just ignored the shit out of it and our NTP servers would have made everything adjust 2012-07-01T01:09:44 Indeed. 2012-07-01T01:10:38 I've seen reports with ntpd running fulltilt pulling 100% cpu (locked to a single cpu, obviously), reports of oracle retardary, reports of java retardary, and I stopped caring right about then. 2012-07-01T01:10:56 It *might* have been something exciting with NTPd kicking off in the middle of the leap second 2012-07-01T01:11:19 And, _somehow_, a spot check of serentos boxen shows ntpd not installed anywhere. 2012-07-01T01:11:28 * Bahhumbug sighs and sees if he has any beer around. 2012-07-01T01:14:12 dagmar do you remember what the dmarker command was exactly that wasn't working? 2012-07-01T01:15:12 Like, none of htem 2012-07-01T01:15:35 i'm trying to set one but keep getting the fucking help pages like I am not giving it the right comm 2012-07-01T01:15:52 I thought it was /dmarker add test 2012-07-01T01:16:33 oh jesus christ 2012-07-01T01:16:40 * Catonic runs to check corp email 2012-07-01T01:16:56 00:16:28 [INFO] /dmarker add