Timestamps are in GMT/BST.
[0:20] * fasseg (~ruckus@HSI-KBW-091-089-154-204.hsi2.kabel-badenwuerttemberg.de) Quit (Ping timeout: 252 seconds)
[0:29] * qwebirc637681 (0cf896aa@gateway/web/freenode/ip.188.8.131.52) has joined #duraspace
[0:29] <qwebirc637681> working on migrating dspace from dev to prod. took a snapshot of dev instance. relaunch dev instance image in prod. brought up the server. however, discovery module is missing, was previously enabled and working on dev?
[0:29] <qwebirc637681> any thoughts or suggestions will be greatly appreciated
[0:29] <kshepherd> qwebirc637681: hi, i just responded in #dspace
[0:30] * qwebirc637681 (0cf896aa@gateway/web/freenode/ip.184.108.40.206) Quit (Client Quit)
[0:35] * dspace__discover (0cf896aa@gateway/web/freenode/ip.220.127.116.11) has joined #duraspace
[0:36] <dspace__discover> hi everyone
[0:37] <dspace__discover> working on migrating dspace from dev to prod. took a snapshot of dev instance. relaunch dev instance image in prod. brought up the server. however, discovery module is missing, was previously enabled and working on dev?
[0:37] <dspace__discover> actually, i can't run dspace update-discovery-index
[0:37] <dspace__discover> Caused by: org.apache.solr.common.SolrException: Internal Server Error Internal Server Error request: http://localhost:8080/solr/search/select?q=*:*&fq=NOT(withdrawn:true)&fq=read:(g0)&start=0&rows=0&facet.field=author_filter&facet.field=subject_tax_0_filter&fac at org.apache.solr.client.solrj.impl.CommonsHttpSolrServer.request(CommonsHttpSolrServer.java:432) at org.apache.solr.client.solrj.impl.CommonsHttpSolr
[0:37] <dspace__discover> any help will be greatly appreciated
[0:46] <dspace__discover> kshepherd: any thoughts?
[0:55] <dspace__discover> anyone?
[1:33] <kshepherd> sorry, wasn't at my pc
[1:33] <kshepherd> can you confirm that your solr server is running on your prod instance?
[1:37] <dspace__discover> ksheperd: how would I confirm that solr server is running, i checked the solar config file http://localhost:8080/solr/search
[1:37] <dspace__discover> and port 8080 is listening
[1:38] <dspace__discover> ksheperd: how can we restart solr service
[1:54] <dspace__discover> ksheperd: u there?
[2:48] * lo5an (~textual@unaffiliated/lo5an) has joined #duraspace
[2:55] <kshepherd> dspace__discover: could you try this on the server: "curl http://localhost:8080/solr/search/select?q=*:*" and see if it returns xml results, or an error?
[2:58] * lo5an (~textual@unaffiliated/lo5an) Quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
[3:48] * kshepherd (email@example.com) Quit (Ping timeout: 252 seconds)
[4:14] * fasseg (~ruckus@HSI-KBW-091-089-022-149.hsi2.kabelbw.de) has joined #duraspace
[6:35] -kornbluth.freenode.net- *** Looking up your hostname...
[6:35] -kornbluth.freenode.net- *** Checking Ident
[6:35] -kornbluth.freenode.net- *** Found your hostname
[6:35] -kornbluth.freenode.net- *** No Ident response
[6:35] * DuraLogBot (~PircBot@atlas.duraspace.org) has joined #duraspace
[6:35] * Topic is '[Welcome to DuraSpace - This channel is logged - http://irclogs.duraspace.org/]'
[6:35] * Set by cwilper!ad579d86@gateway/web/freenode/ip.18.104.22.168 on Fri Oct 22 01:19:41 UTC 2010
[13:34] * mhwood (firstname.lastname@example.org) has joined #duraspace
[14:00] * tdonohue (~email@example.com) has joined #duraspace
[16:08] * fasseg (~ruckus@HSI-KBW-091-089-022-149.hsi2.kabelbw.de) Quit (Ping timeout: 252 seconds)
[16:11] * jonathangee (~firstname.lastname@example.org) has left #duraspace
[16:21] * lo5an (~email@example.com) has joined #duraspace
[16:21] * lo5an (~firstname.lastname@example.org) Quit (Changing host)
[16:21] * lo5an (~textual@unaffiliated/lo5an) has joined #duraspace
[19:09] * kompewter (~email@example.com) has joined #duraspace
[19:46] * kstamatis (5b8c1689@gateway/web/freenode/ip.22.214.171.124) has joined #duraspace
[19:50] * dspace__discover (0cf896aa@gateway/web/freenode/ip.126.96.36.199) Quit (Quit: Page closed)
[20:01] <tdonohue> Hi all. It's time for our weekly DSpace Developers Meeting: https://wiki.duraspace.org/display/DSPACE/DevMtg+2014-01-15
[20:01] <kompewter> [ DevMtg 2014-01-15 - DSpace - DuraSpace Wiki ] - https://wiki.duraspace.org/display/DSPACE/DevMtg+2014-01-15
[20:01] <tdonohue> Hmm.. It's looking like our attendance is lower today. :)
[20:02] <helix84> hello
[20:02] <mhwood> Hello.
[20:02] <tdonohue> I'll admit, I didn't really have any great ideas for agenda items today. So, I mostly just added some questions to 4.1. We can choose to go "off the agenda" today if we want
[20:02] <tdonohue> s/to 4.1/to it/
[20:02] <kompewter> tdonohue meant to say: I'll admit, I didn't really have any great ideas for agenda items today. So, I mostly just added some questions to it. We can choose to go "off the agenda" today if we want
[20:03] <tdonohue> The first question was just to bring up the idea of a 4.1 release. I don't know that we're ready *yet*. But, I wanted to put it into everyone's minds...
[20:04] <helix84> so far, master doesn't look so hot
[20:05] <helix84> although we might want to continue in PR reviews and thus get more commits into master, some of which might be backported. Or we might focus solely on bugfixes during reviews.
[20:05] <helix84> can't say we have a lack of bugs...
[20:06] <tdonohue> Yea, it seems like we have some bugs coming in...and some that missed the 4.0 cut. So, we likely will have a 4.1. Just wasn't sure we have anything requiring "immediate release" which would necessitate a 4.1
[20:07] <tdonohue> So, it makes sense to me to also wait a bit, and work at reviewing JIRA tickets / PRs. Once we have more activity on the "dspace-4_x" branch or something requiring immediate release, then we would want a 4.1
[20:08] <tdonohue> Currently, it looks like we have *nothing* on 'dspace-4_x' since 4.0: https://github.com/DSpace/DSpace/commits/dspace-4_x
[20:08] <kompewter> [ Commits · DSpace/DSpace · GitHub ] - https://github.com/DSpace/DSpace/commits/dspace-4_x
[20:08] <helix84> Nothing comes to my mind. I just wanted to say that with 4.1, we'll want to update the manual PDF because the one included in the 4.0 release wrongly instructed to build from [dspace-src]/dspace, which doesn't work anymore
[20:09] * PeterDietz (~firstname.lastname@example.org) has joined #duraspace
[20:09] <mhwood> Doesn't that mean we need to fix the "binary" build, instead of the documentation?
[20:10] * hpottinger (~email@example.com) has joined #duraspace
[20:10] <helix84> mhwood: I don't know why it doesn't work - whether it's fixable or it was changed intentionally
[20:11] <helix84> anyway, the documentation on the wiki now reflects says to build from [dspace-src]. it's just the PDF that has wrong instructions.
[20:11] <tdonohue> I *believe* installation from [dspace-src]/dspace/ should still work fine. It's that *upgrading* from [dspace-src]/dspace/ has issues
[20:12] <mhwood> Actually, if we do edit the manual, one thing we want to do is *stop* putting the PDF in the source kit. In 4.1 onward it should be tucked into the release directory on SourceForge separately, alongside the archives.
[20:12] <helix84> right, we can simply exclude it from 4.1.
[20:13] <mhwood> Also need to 'git rm' it from the source kit.
[20:13] <helix84> so, what about REST API? in case there are any interesting developments, do we want to include them in 4.1? Do we consider REST API beta or stable?
[20:13] <tdonohue> Also, I think the cause of those build problems is the fix we had to implement for UTF-8 & build.properties: DS-1528. I haven't had a chance to really look closely, but it seems like the issue has to do with the encoding of the build.properties for UTF-8
[20:13] <kompewter> [ https://jira.duraspace.org/browse/DS-1528 ] - [DS-1528] build.properties doesn't support UTF-8 encoding - DuraSpace JIRA
[20:15] <helix84> it should be easy to try reverting the patch and seeing if the build starts working again
[20:16] <mhwood> Hmmm, I wonder if there's a more recent resource plugin available so we don't have to recode stuff.
[20:16] <mhwood> Assuming it fixed the problem.
[20:17] <tdonohue> reverting that patch will re-break UTF-8. But, I think the issue has to do with the fact that build.properties is now copied to [src]/target/ to UTF-8 encode it. But, when you run mvn from [src]/dspace/ , the build.properties doesn't always get copied to [src]/target/
[20:17] <helix84> I meant to revert it just to confirm that it's the cause of the problem.
[20:18] <tdonohue> we could look at resource plugin, but I wouldn't get my hopes up.
[20:19] <tdonohue> but, yes, I think we should investigate if 1528 is the cause. If it is, we may need to dig back into our POMs to figure out a better solution to 1528
[20:20] <hpottinger> hmm... so maybe the magic UTF-8 encoding code needs to examine the current working directory?
[20:21] <tdonohue> hpottinger: yes...either that, or we need to ensure the "magic UTF-8 encoding code" is called properly whether you run mvn from [src] or [src]/dspace/
[20:21] <hpottinger> sorry, I get non-specific with my lingo when I don't know the details, it's funnier if you can see my hands and face :-)
[20:21] <tdonohue> It's definitely always called when running mvn from [src] (as that "magic code" is in the [src]/pom.xml). But, it sounds like it is *not* always called from [src]/dspace/
[20:22] <hpottinger> oh, so the "fast build" is having problems
[20:22] <tdonohue> So, I feel like it's a minor POM issue...something is not fully "stitched together" right. I just haven't had any time myself to actually dig in and look at it.
[20:23] <tdonohue> correct, hpottinger. Just the "fast build" (from [src]/dspace/) has issues
[20:24] <tdonohue> we probably should create a JIRA ticket around this issue, so we can track what we *think* and what we actually *figure out*. For the time being, telling people to run 'mvn' from [src] is the workaround.
[20:25] <hpottinger> this feels like a good candidate for a "hackathon"... just a few of us need to dig in and compare notes
[20:26] <tdonohue> yea, it may be. I had thought this was all working when I last touched this code (in 1528), but obviously there must be some scenario(s) when running mvn from [src]/dspace/ that just throw errors.
[20:27] <tdonohue> In any case, I can create a ticket after this mtg. Since I think I know what is going on, even though I'm not 100% sure how to replicate it yet.
[20:27] <hpottinger> we all build from [src], so we're not likely to catch it... I wonder if we could have Bamboo also test the "fast build"?
[20:28] <tdonohue> yea, I think that should be possible. We all (Committers) have access to tweak Bamboo settings. We should be able to create a second project to test the "fast build" process too.
[20:29] <tdonohue> In any case, I'll create a JIRA ticket
[20:30] <tdonohue> So, reading back... I see helix84 had mentioned REST API. Obviously we'd accept bug fixes to REST API in 4.1.
[20:30] <tdonohue> but major changes to the REST API would have to wait for 5.0, just like any other major changes
[20:33] <tdonohue> As I had mentioned at the beginning of this meeting, I really was struggling to come up with an agenda for today. Although I had some other topics listed, we really could discuss whatever we feel is most important/pressing/interesting today. The "floor" is open
[20:34] <helix84> I suggested we should focus with our reviews on bugfixes
[20:34] <helix84> i.e. give priority to reviewing PRs which fox a bug
[20:35] <tdonohue> One thing I will mention: You all should be thinking about submitting some talk/workshop proposals for Open Repositories 2014 (in Helsinki)!
[20:36] <tdonohue> It'd be nice to have some OR2014 talks on 4.0 and major new features (in recent releases): REST API, redesigned JSPUI, DOI work, etc etc.
[20:36] <PeterDietz> Bram has be scheduled to give a webinar on REST some time in February
[20:36] <PeterDietz> be => me
[20:37] <tdonohue> yep, I am aware of those DSpace 4 upcoming webinars. Just noting that OR2014 might also be another place to present essentially the same thing -- may actually be a slightly different audience in Helsinki
[20:38] <mhwood> No notion yet that I'd even be going to OR 2014. But I will keep it in mind.
[20:38] <hpottinger> "Towards Repository Ecosystems" is the theme, if you haven't seen it yet
[20:40] <tdonohue> yea, just thought I'd reminder folks regarding OR2014, especially if it helps you find funding to go to Helsinki. Presentation proposals are due Feb 3, 2014 (both for the main conference and for the DSpace User Group Mtg).
[20:40] <hpottinger> also, note that the conference date moved up a month, to June, which puts it in the same FY as OR13, which is bad news if you spent your entire travel budget in 2013, and good news if you didn't
[20:41] <tdonohue> I will be in Helsinki. I'm also hoping to schedule another face-to-face DSpace Developers Mtg at OR14 (for anyone who can be there), likely on that Monday. But, I'm still trying to work out details with conference planners, etc.
[20:43] <tdonohue> In any case, just wanted to put OR2014 in your minds. Moving along...
[20:43] <tdonohue> do we want to do some PR reviews for the last 15 mins?
[20:44] <mhwood> Bloody JIRA just forgot my login again.
[20:44] <helix84> perhaps the reason for short Jira sessions is to make us remember our password?
[20:44] <helix84> s
[20:45] <tdonohue> We (DuraSpace) are still trying to tackle the horrible "session timeouts" in JIRA/Wiki. We thought we had it fixed yesterday, but it came back this morning. We have a support ticket in with Atlassian, and unfortunately even they are "stumped" so far.
[20:45] <mhwood> Thank you.
[20:45] <tdonohue> So, believe me...we are also frustrated with it, and want it to go away
[20:46] <helix84> I've had an idea for some time now, I never put it into practice though. Since today's meeting is slow, I'll try to pitch it now.
[20:46] * lo5an (~textual@unaffiliated/lo5an) Quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
[20:47] <helix84> I try to answer most question on dspace-tech I can, but of course there are questions that remain unanswered.
[20:47] <helix84> I was wondering if the reason is that they are buried in answered questions and other people simply didn't notice that they are still unanswered.
[20:48] <helix84> If I make a list of unanswered questions for the last week every week and other developers could look at them, would they (you)?
[20:48] <mhwood> In my case, the usual reason I don't respond is that I don't know. :-/
[20:48] <mhwood> I mean, don't know the answer, never used that, no idea where to look.
[20:49] <tdonohue> for me, the reason for not answering can either be: "I unfortunately don't have time today", or "I don't know offhand" or some combo of the two.
[20:50] <hpottinger> gathering questions someplace else would help
[20:50] <PeterDietz> If we had some type of paid support, then it would make sense to have some type of helpdesk queue to ensure that questions don't age, have some type of default assignee. But with this being "best effort" community service, then I typically just peel off ones that pique my interest
[20:51] <helix84> If it pleases you, you can pay me and I'll work on managing the queue ;)
[20:51] <mhwood> And a paid support email would probably be hooked right into the helpdesk, with emails captured as issues and going to a triage team for routing.
[20:51] <mhwood> But we don't have that.
[20:51] <tdonohue> gather questions elsewhere *may* help (and it could actually help track common questions), but we don't have a really good "queue" /helpdesk. Sometimes, for me at least, a simple "ping" on the list (which mhwood has done sometimes) will remind me that oh-yea, I should have answered that ;)
[20:53] <helix84> Some system with a public interface and the possibility of tagging emails would go a long way. Does anyone know of something like that?
[20:54] <PeterDietz> Not to sound like a google spokes-person, but I've been using Google+ Communities very heavily lately, and its helpful to +mention someone, as it goes into their Alerts
[20:54] <helix84> PeterDietz: how does that relate to emails?
[20:55] <mhwood> Let's see, I've been on + once this year....
[20:55] * lo5an (~firstname.lastname@example.org) has joined #duraspace
[20:56] * lo5an is now known as Guest98977
[20:57] <tdonohue> One sidetopic here that is semi-related: DuraSpace is working on establishing a DSpace Steering Committee (of both appointed and eventually elected folks), who we could as such questions to. E.g. we could point out the possible need for a "helpdesk" and ask them if it'd be worth funding from the $$ donated to DSpace from sponsoring institutions. This Steering Committee is not yet formed, but they will be able to allocate funds/r
[20:58] <tdonohue> I'm not saying that we definitely need a fully-fledged "helpdesk" system here... But, this discussion just reminded me to mention the forthcoming Steering Committee who can help fund or find resources for such ventures
[20:58] <PeterDietz> So, I'm not really aware of a tool to give you features beyond email, using an email list. But if you shifted towards a different community support forum. i.e. Google Groups, Google Community, ..., then you could get extra features, maybe sort by questions without answers
[20:59] <mhwood> I'd be less likely to become aware of questions in a timely fashion. The nice thing about email is it comes to me; I don't have to remember to go to it.
[21:00] <tdonohue> Similar to what PeterDietz is mentioning...perhaps a migration to Google Groups from SourceForge mailing lists is worth discussing again? I'll mention that Fedora just recently migrated all their mailing lists from SourceForge to Google Groups
[21:00] <mhwood> Migrating from SourceForge lists to almost anything would be an improvement. The only other thing I dislike nearly as much is...Google Groups.
[21:00] <tdonohue> I don't know that a migration to Google Groups 100% *solves* this problem, but it at least makes old emails/questions easier to locate
[21:02] <mhwood> Although PHPbb is right down there at the bottom of my list too.
[21:02] <hpottinger> I'm OK with moving the lists, *as long as* there's an archive of them somwhere
[21:02] <tdonohue> So, what do you like, mhwood? :)
[21:03] <mhwood> Email works fine. It's the archive interface at SF that's lousy.
[21:03] <hpottinger> nibble does OK searching the archives
[21:03] <tdonohue> I'm not suggesting we go willy-nilly moving the lists without a good reason. I'm just mentioning that we have the option to move our lists away from SourceForge, if we wanted to. But, we need to really want to, as the migration would be time consuming :)
[21:04] <hpottinger> nabble, you silly Mountain Lion
[21:04] <mhwood> I admit to being a dinosaur where this sort of commo is concerned. I use Mutt because I like it better than Thunderbird or Evolution or a dozen others that I've briefly tried over the years.
[21:05] <helix84> although there's Nabble and mail-archive, I'd still strongly prefer if we could export and import the archive to the new service.
[21:07] <tdonohue> I'll admit, I'm often frustrated with SourceForge (in general). I just haven't become frustrated enough to push us to migrate elsewhere. But, if there is a combined frustration here...we really have the option to move things elsewhere in general. No one even is requiring us to use SourceForge at all. We could think about moving lists to Google Groups, and our "pre-packaged downloads" to GitHub "Releases" if we really wanted t
[21:07] <tdonohue> I'm not suggesting we decide this today. Just wanted to throw it out there.
[21:08] <tdonohue> And besides, I'm noticing we're "over time" already
[21:09] <hpottinger> personally, I've noticed a weird "lag" with Google Groups-based mail lists
[21:09] <helix84> SF mailing lists have 2 big drawbacks - slow browsing of archives (I don't remeber I ever searched in them, either it's not possible or slow) and delays in delivery (emails are often delayed up to an hour)
[21:10] <hpottinger> and I know some have noticed the same thing with source forge lists... but it's really consistent for Google lists
[21:10] <mhwood> Considering what we're paying, I suppose we can't complain very loudly.
[21:11] <tdonohue> haha, yea :)
[21:11] <helix84> In case we decide to host something ourselves: http://en.wikipedia.org/wiki/Category:Free_mailing_list_software
[21:11] <kompewter> [ Category:Free mailing list software - Wikipedia, the free encyclopedia ] - http://en.wikipedia.org/wiki/Category:Free_mailing_list_software
[21:12] <tdonohue> and if there's a better free hosted email list service out there, we can look at other options. I just know most other Library open source projects are using either Google Groups or SourceForge
[21:12] <mhwood> Except ImpactStory, who moved from Google Groups to Twitter, so I never hear anything out of them anymore.
[21:13] <hpottinger> is "free" really a requirement?
[21:14] <mhwood> Free or perhaps hosted by a community member with a long-term commitment.
[21:15] <tdonohue> perhaps not "free", but "inexpensive/affordable" would be a requirement. DSpace is funded by community sponsorship / volunteerism. We don't have lots of $$ to throw around :)
[21:15] <hpottinger> almost any web hosting service throws in mail lists for free
[21:15] <tdonohue> or yes, hosted as mhwood said may be workable
[21:16] * tdonohue notes that as we are "over time", I'm not going to call any more topics. But, I'll let this discussion continue as long as we want. Feel free to drop off though if you need to head out.
[21:16] <hpottinger> Mailman is free and not a bear to keep going
[21:18] <tdonohue> I suspect that "hosted" is going to be a requirement. Either a hosted service (like Google Groups/SourceForge), or hosted by a community member with a long-term commitment.
[21:18] <tdonohue> Though, maybe that'd change if the DSpace Steering Committee was able to find some extra funding to support more DuraSpace-sysadmin-types to run a Mailman and other services. Hard to say.
[21:20] <tdonohue> (Just wanting to admit that DuraSpace is trying to not run *too much* internally, unless really necessary, like in the case of the Atlassian toolset. We just don't have the $$ to hire that many sysadmin types)
[21:21] <tdonohue> plus, as-is, DuraSpace is only 10 people ;)
[21:22] <tdonohue> But, if a kind institution wanted to host a mailman on our behalf (which is actually what MIT *used to do* before we migrated to SourceForge), it might be an option to consider
[21:24] <tdonohue> In summary: It sounds like we don't have a solution to the problem of "unanswered questions". We've noted it is a problem. We've noted that it might be worth thinking about migrating our mailing lists elsewhere (to ease locating older/unanswered questions). But, no decisions as of yet.
[21:25] <helix84> I'll head out now. Good night!
[21:25] <mhwood> That is a problem for which tools won't really do a lot. It takes people, and in particular a kind of people who are probably a little rare on our MLs: the ones who like to sit all day watching and classifying messages.
[21:25] <mhwood> Good night.
[21:26] * hpottinger (~email@example.com) has left #duraspace
[21:28] * kstamatis (5b8c1689@gateway/web/freenode/ip.188.8.131.52) Quit (Quit: Page closed)
[21:36] * Tojil (~maya@unaffiliated/tojil) has joined #duraspace
[21:37] <Tojil> Hello
[21:39] <PeterDietz> hello
[21:40] <Tojil> Hello Peter, what's up?
[21:41] <PeterDietz> not too much. We just wrapped a #dspace developers meeting in here a bit ago. I'm just fixing some local bugs for now
[21:43] <Tojil> Nice
[21:44] <Tojil> For how long are you working with DSpace?
[21:45] <PeterDietz> maybe about 4 years. I started gradually, just giving feedback while trying to use some features, and since then I've built a few features myself. (ElasticSearch statistics, REST API)
[21:46] <Tojil> Nice
[21:48] <Tojil> I have like... A Mercury synodic cycle of experience in implementing DSpace in VPS.
[21:48] <Tojil> i.e. ~116 days
[21:48] <Tojil> hehe :)
[21:51] <PeterDietz> fun, hopefully its working for you. Luckily I was able to get a little bit of training when I started, which gave me some initial momentum. And then I was able to contact people online for help
[21:53] <Tojil> Yes, I think i already can install and fix permissions, hehe. And some customization like having your own images, header/footer, CSS, for sure
[21:54] * Guest98977 (~firstname.lastname@example.org) Quit (Quit: Textual IRC Client: www.textualapp.com)
[21:55] * lo5an (~email@example.com) has joined #duraspace
[21:55] * lo5an (~firstname.lastname@example.org) Quit (Changing host)
[21:55] * lo5an (~textual@unaffiliated/lo5an) has joined #duraspace
[21:55] <Tojil> I just had a little part of that time to actually work with DSpace (and the VPS were not ON the whole time), so it seems I'm going well as a beginner
[21:56] * lo5an (~textual@unaffiliated/lo5an) Quit (Client Quit)
[21:56] * lo5an (~lo5an@unaffiliated/lo5an) has joined #duraspace
[22:01] * mhwood (email@example.com) has left #duraspace
[22:28] * Tojil (~maya@unaffiliated/tojil) Quit (Quit: My damn controlling terminal disappeared!)
[23:25] * tdonohue (~firstname.lastname@example.org) has left #duraspace
These logs were automatically created by DuraLogBot on irc.freenode.net using the Java IRC LogBot.