[00:49:38] --- kaj has become available [01:30:19] --- deason has left [01:39:25] --- Born Fool has left [03:46:23] --- Jeffrey Altman has left: Replaced by new connection [03:46:24] --- Jeffrey Altman has become available [05:23:41] --- Simon Wilkinson has become available [05:24:28] I don't think Booker has finished with 1521 - I think he's planning on submitting a further set of changes, so it may be premature to review it in its current form. [06:42:46] --- Russ has left: Disconnected [07:05:39] Booker is listed as a reviewer on 1521. I wish he would say that himself. [07:15:59] --- jaltman has left: Disconnected [08:04:14] --- kaj has left [08:27:09] --- jaltman has become available [08:28:06] --- jaltman has left: Replaced by new connection [08:28:07] --- jaltman has become available [11:03:40] --- kaj has become available [11:26:02] --- haba has left [12:48:19] --- deason has become available [13:06:07] --- jaltman has left: Replaced by new connection [13:06:07] --- jaltman has become available [13:38:01] I should probably go read some code, but am I right in thinking that the fact that the bos general restart is scheduled for the same time on all machines, mean that all of the DB servers will stop responding at once? [13:54:50] --- haba has become available [13:58:45] --- Russ has become available [14:40:20] --- haba has left [14:41:17] --- kaj has left [14:41:25] --- kaj has become available [14:41:28] --- haba has become available [14:50:32] --- haba has left [14:50:32] --- haba has become available [14:53:28] --- kaj has left [15:03:00] --- haba has left [15:05:50] --- kaj has become available [15:19:27] --- sxw has become available [15:19:57] Okay, gerrit is about to shutdown for an upgrade to 2.1.2 [15:23:50] Russ: If you have a spare moment, could you give me root access to /etc/init.d/gerrit ? [15:26:47] Sure thing. [15:27:06] --- haba has become available [15:32:47] woo [15:33:02] Done. [15:33:10] Thanks. [15:35:18] > mean that all of the DB servers will stop responding at once? yes [15:54:41] --- jaltman has left: Replaced by new connection [15:54:42] --- jaltman has become available [15:57:21] --- kaj has left [15:58:28] Well that took longer than it should have. [15:58:45] Gerrit is now running 2.1.2. I'll get the hooks set up to link it with RT later this week. [16:02:43] cool [16:13:19] I like 2.1.2 [16:14:49] --- jaltman has left: Disconnected [16:14:55] --- jaltman has become available [16:15:20] I should write some notes for the openafs-devel group. jaltman: What do you particularly like about it? [16:16:33] aside from documentation, what difference do you see? [16:16:59] i guess the new download choices. [16:17:09] The All tab is better [16:17:22] "better" [16:19:12] I won't say that all of these are new but they are things I like in this release. 1. anonymous viewers of the dashboard see the state of verified and reviewed [16:20:15] That's new across the whole dashboard for "all" changes. Which is very nice. [16:24:46] I suspect the Forge Identity change may cause us pain - you can no longer push changes on behalf of other users, it would appear. [16:24:51] 2. the Download choices. In particular, the cherry-pick option since that is how we work. I think cherry-pick should be our default choice if it is not already. 3. when reviewing you can "Download" the source file which makes reviewing .xml and .html files so much easier since I can loaded them from the browser to view them. 4. The tab width and columns options in the review source screen. 5. Search by Change-Id 6. Intraline Difference highlighting 7. Show Tabs 8. Whitespace Errors I'm not sure how I feel about the renaming of "Publish Comments" button to "Review" [16:26:02] for verifying, a download cherry pick is not needed [16:26:12] 10. Documentation is always nice to have. Perhaps we should have a link to the pages you wrote from here as well. [16:26:25] The intraline difference highlighting is very nice, but some (internal) Google sites have reported that it floors their gerrit server. [16:26:47] We'll need to keep an eye on it. I suspect some of our files may break it. [16:33:30] I wonder if I can configure it so there's no way of ignoring whitespace errors ... [16:37:29] do not want. [16:37:42] ? [16:37:44] unless if it's moving existing code, it's not an error [16:38:21] It's just the syntax highlighting thing - whether you get shown them in red, or not. [16:38:27] It won't stop you from pushing. [16:38:44] ah, ok [16:39:16] Now, buildbot could be set up to stop you from pushing, but I've no intention of doing that at present. [16:40:32] i have direct access to the repo, so there's always a way... [16:40:35] Re the Documentation thing. To include our local stuff there, we'd need to patch the WAR that comes with each gerrit. I'd rather not get into doing that, because it just makes upgrades harder. [17:27:21] There's a post from Jeff Blaine in the openafs-info moderation queue - it's held because it's 99k in size. [17:31:37] reject it [17:34:40] Done, with friendly message. [17:36:15] thank you [17:37:02] Oh wonderful. Our Linux build host has a kernel panic. [17:39:04] woohoo! [17:39:08] what's it running? [17:39:16] master [17:40:07] i have fsevents working but it's leaking vnode references. i don't see how. i wonder if the kernel leaks it because we don't do something we're supposed to. [17:41:11] > it's held because it's 99k in size i hear you can give people URLs for things that are huge [17:45:04] panic is in RT as 126755 [17:45:16] Haven't had a chance to look at it as yet. Now off to sleep. [17:46:16] sleep well [17:46:32] Blaine was sending a JPG of a partial view of a edited AS_REP as viewed in wireshark. Nothing remotely useful [18:26:43] I wish I had something like growl at the moment. it would make my life a bit easier [18:27:24] I guess what I really want is log levels for the afsd_service logging. [18:27:36] but that is such a huge project [18:28:06] growl is dead simple [18:28:55] the windows implementation is kind of ugly though. open tcp ports that any machine can connect to [18:29:44] keyed? [18:30:09] the only protection is if you assign "passwords" to each application [18:30:29] but then we need a mechanism to set the password for OpenAFS [19:53:13] --- Born Fool has become available [20:04:50] Simon: I pushed s/bzero/memset to gerrit/1560 a couple days ago [20:50:59] --- Born Fool has left [22:12:01] --- deason has left [22:44:46] --- reuteras has become available