[00:34:13] --- sxw has become available [01:14:09] --- Russ has left: Disconnected [02:49:15] --- sxw has left [03:24:41] --- sxw has become available [05:07:13] --- sxw has left [05:25:59] --- SecureEndpoints has left: Replaced by new connection [05:53:37] --- sxw has become available [06:08:56] --- sxw has left [06:11:44] --- sxw has become available [06:13:47] --- sxw has left [06:22:21] --- dev-zero@jabber.org has become available [06:22:30] --- dev-zero@jabber.org has left: offline [06:29:10] --- sxw has become available [06:42:29] --- sxw has left [06:46:40] --- sxw has become available [06:47:34] --- sxw has left [07:39:26] --- dev-zero@jabber.org has become available [07:39:32] --- dev-zero@jabber.org has left: offline [09:05:55] --- sxw has become available [09:07:25] --- sxw has left [09:52:31] --- sxw has become available [09:53:59] --- sxw has left: Replaced by new connection [09:53:59] --- sxw has become available [10:13:55] --- sxw has left [13:00:02] --- Russ has become available [15:53:47] --- edgester has become available [17:55:32] hello [17:59:25] I would like a couple of people to review a draft of the openafs newsletter to see if it's good. http://www.secure-endpoints.com/afs/dementia.org/home/edgester/openafs-newsletter.txt [18:02:02] sic: Jeff hopes ot have a beta [18:02:32] you might add a blurb about Felix's latest efforts on getting Rx OSD integrated [18:02:41] ie, the 'translate_rpc' discussions. [18:03:11] I think it would be good to have the delta names in the newsletter, as well as the cvs stats [18:03:33] to me, the actual delta info is far more interesting than how many LOC were changed. [18:04:01] What kind of delta info should be added? [18:04:03] also, do you want to 'pretty up' the doc a bit, or are you thinking plain text is easiest. [18:04:15] delta names would be fine, I think. [18:04:26] I think plain text is the lowest common denominator [18:04:43] hmmm, I'll have to rewrite by script. [18:04:44] depends on your goal. [18:04:58] are you aiming at the lowest common denominator w/this newsletter? [18:05:09] or are you trying to show that OpenAFS is alive and well? [18:05:21] The goal is to give the folks on -info a summary of what's going on and the fact that openafs is alive and well [18:06:19] I'd encourage you to write the newsletter in some kind of markup that at least lets you auto-generate a prettier version [18:06:37] like what? [18:06:55] POD, Textile, etc. [18:08:19] first iteration done in plain text is fine, though. It's better than what's existed previously. [18:08:53] wrt the documentation work.. [18:09:06] I have no clue what's needed at this point; i.e., I thought the man pages were all done. [18:09:21] I know there's been some noise recently about the quick start guides, but I'm not sure what's going on w/that. [18:09:36] I started looking at the quick sart guide [18:09:52] The man pages still need work, but are very usable [18:10:04] some obscre commands aren't documented [18:11:43] ah, ok. I see the list in docs/man-pages/READEM [18:11:45] err, README [18:12:03] does fs getfid even exist? [18:12:13] as a man page? [18:12:31] as a command. and, yes, it does. [18:13:15] the stuff that's left is mostly beyond my knowledge [18:14:03] There's a bunch of mechanical changes that are needed still too. [18:14:13] argh. [18:14:19] I started looking at the quickstart guide because I'm tired of new admins doing the wrong thing by following the guide, setting up kaserver and getting chided on the mailing list [18:14:21] I don't think we have -noresolve documented in every vos command, there's kaserver references, that sort of thing. [18:14:25] fs cscpolicy, fs memdump man pages should be there.. [18:14:40] russ - /me nods. I see that in the README as well. [18:14:51] Yeah, I tried to collect everything in the README. [18:15:02] I have man pages for fs cscpolicy and fs memdump..have had them since July.. [18:15:57] put the file in RT [18:16:00] will do [18:16:20] * Russ hasn't been as fast at processing tickets as I'd like since tons of other stuff going on lately, but I'll get to them. [18:16:38] please include a patch for the README as well, to help Russ [18:16:46] these were never submitted. a former employee wrote them; was supposed to s ubmit them, but I see he never did. [18:17:24] give me just a few mins; I can at least get those two uploaded. [18:17:34] Im' surprised that I didn't do more over the winter, but work has been hectic [18:17:41] no guarantees on quality, though; it wont hurt my feelings if you tell me they are garbage. [18:17:47] * Russ finds the XML manuals intimidating. I can't ever figure out where to start. [18:18:13] * edgester just dove into the XML and started at a random place. [18:18:22] That sounds like a good approach. [18:18:46] I may also have some updates to the quick start guides in xml. yet another former employee did some work. I'll look around and see (or ask him) [18:18:48] The XML doesn't intimidate me [18:19:03] I often feel that I'm not the best person to write some of this stuff [18:19:29] Everyone feels that way except maybe for Derrick and Jeff. :) [18:19:45] * edgester feels better [18:19:52] I have to admit I feel a certain amount of ego/confidence wrt some of the documentation... [18:19:59] ?? [18:20:08] e.g., the quick start guides. [18:20:27] i.e., I've taught so many people that stuff over the years, that I know it relatively well. [18:20:49] Jason has done a great job just diving into something he'd never used and writing a man page based on mailing list comments. It's like the old adage goes: if you want replies, don't ask questions, post incorrect information. :) Once we have the framework of documentation, people come out of the woodwork to correct it, usually. [18:20:49] oh yeah, anyone who has set up openafs from scratch properly is qualified to rewrite most/all of it. [18:21:09] there are lots of cases, though, I've either never touched (e.g., Heimdal setup), or only touched years and years ago (fakeka & friends), that I shouldn't be trusted to write on those w/o some refresher. [18:21:45] --- abo has left [18:21:51] fakeka and heimdal don't scare me, it's the stuff that only Jeff and Derrick know [18:21:58] that scares me [18:22:27] Russ, would you please skim http://www.secure-endpoints.com/afs/dementia.org/home/edgester/openafs-newsletter.txt and let me know what you think? [18:22:49] I'll put an fs_getfid man page together real fast, too. that one is simple. [18:23:30] For rxk5-devel in the CVS statistics, may want to say "none up to April of 2009" since a ton of stuff has changed recently, I think. [18:23:43] Russ: thanks for the compliment on the docs. I still remember feeling honored when I got a mention at the openafs status update at the 2008 workshop [18:24:06] You've been a ton of help. I really appreciate the work you've done. [18:24:45] thanks [18:24:48] edgester: I think the idea for gerrit is more to streamline code review than specifically to allow voting. [18:25:55] Typo (ot for to) in the Native Windows section. [18:26:45] do you want to mention the security issue and add a reference to the advisory? [18:27:53] probably not, that was in the news section on the home page [18:28:22] updated [18:28:36] w/typo fix and OSD blurb [18:28:42] an gerrit stuff [18:29:12] Fixes look good here. [18:29:53] cool. I added a link to one of Felix's threads [18:30:15] Is there anything else that should be added [18:30:16] ? [18:31:18] Not that I know of, but I've also not been paying a lot of close attention lately. [18:31:34] That's one of the reasons for the newsletter ;) [18:32:05] It's a pain to keep up, chats happen in different channels, jabber, zephyr, etc [18:32:31] Russ: do you know how to use zephyr? [18:32:36] * Russ mostly ignores Jabber too since the interface is so crappy and I haven't had a chance to experiment with barnowl. [18:32:41] Yeah, I'm on Zephyr. [18:33:37] * Russ looks in here periodically, but rarely reads backscroll. [18:33:49] I'm using owl on lily, but I don't know how to join the main openafs zephyr channel [18:34:38] Exit owl, add shadow,*,@ANDREW.CMU.EDU to .zephyr.subs, restart owl. There's presumably some way to do it from inside owl, but I never bother. [18:35:15] You can sub to openafs,*,@ANDREW.CMU.EDU too, but mostly everyone just uses instances under Derrick's class. [18:36:07] ok, I added those two channels, how to I enter them? [18:36:13] I'm a total noob on owl [18:36:23] ": load-subs" will work inside owl after you've added them to the subs file. [18:36:31] Strictly speaking, shadow,openafs is all that you need, but people tend to wander all over shadow,*, so it's hard to follow all of the discussion without also getting random updates on Derrick's life. :) [18:36:50] You don't -- Zephyr doesn't have a notion of rooms or channels. If you're subscribed, you see all the messages. [18:37:02] If you reply, you'll reply to the same class and instance as the message to which you're replying. [18:37:41] hmm, I don't see messages [18:37:50] I'm on moro, does that matter? [18:37:53] No one's saying anything at the moment. One second. [18:38:20] * Russ just sent a message. [18:38:39] shold I see it in the home screen? [18:38:53] Yeah, unless you've done something weird with filters that I don't know how to do in owl. [18:39:07] I don't know how to do filters [18:39:36] If you do :listsubs, do you see the shadow sub? [18:40:23] RT 124756 has 3 man pages and an updated README. [18:42:19] :listsubs is not a valid command [18:42:28] "l" shows shadow [18:42:49] It might be :list-subs. If I use it, I have to go get new Kerberos tickets. [18:43:01] Ah, :getsubs [18:43:21] better [18:43:35] I see shadow and openafs [18:43:50] but I only see the welcome message in the home screen that appears when olw starts [18:43:52] They're @ANDREW.CMU.EDU too? [18:44:13] │ │ [18:44:44] Yeah, that's right. [18:45:02] Not sure why it wouldn't be working. [18:47:15] * Russ just leaves owl running in screen all the time. [18:51:23] Can you read zephyr messages that are sent before you logged in? [18:51:31] Nope. [18:51:36] ah [18:51:50] You get a warning if you send a message to someone who isn't logged in. [18:52:23] hmmm [18:55:41] ick. [18:55:49] russ - do you know of anyone who uses copyauth? [18:56:59] No, but I have very little contact at that level with anyone who has multiple cells. [18:57:27] tx; I wonder if anyone would complain if it went away -- it should probably be slated to die when kaserver goes away. [18:57:27] But with a modern aklog, I have a hard time seeing the point. [18:57:45] * stevenjenkins nods. [18:57:50] It's only useful with multiple cells with the same key. [18:57:54] exactly. [18:58:00] that's why I said 'ick'. [18:58:09] Whereas aklog works fine and even supports different keys with the same realm. [18:58:26] do you know the status of the kaserver deprecation/removal? [18:58:40] e.g., is kaserver going to be in 1.6? [18:58:43] It's dependent on native K5 support. [18:58:58] So until either rxk5 or rxgk is in stable, kaserver will be there. [18:59:03] ah, right. [18:59:07] and docs [18:59:11] :( [18:59:23] Error SERVNAK received subscribing to zephyr notifications. [18:59:37] It doesn't *need* to be dependent on that, but that was the general plan. [18:59:39] I got that in owl, any clues how to fix? [19:00:01] Do you still have a K4 ticket? [19:00:18] Or maybe K5 now -- Derrick probably upgraded dementia or something. [19:00:35] yes [19:05:46] oh well, thanks for the help. I'll try zephyr later to see if the error goes away. [19:07:37] hmm, zephyr is working now [19:23:29] --- edgester has left [19:51:21] --- SecureEndpoints has become available [20:18:33] --- sxw has become available [20:19:58] --- sxw has left: Replaced by new connection [20:19:59] --- sxw has become available [20:37:31] last time I tried, I couldn't get zephyr to compile [22:18:31] tho, pidgin appears to have zephyr support, now [23:43:29] --- sxw has left