The IRC channel is #sakai on the freenode network, you can join using a web client at http://webchat.freenode.net/?channels=sakai and you can view the complete logs at http://sakai.iriscouch.com/irc/_design/viewer/index.html
13:53 <stuartf> nicolaas_: ping 13:53 <nicolaas_> pong 13:53 <stuartf> I'm sort of stuck on SAKIII-6144, I can't get it to fail the way it does in the screenshots 13:54 <nicolaas_> hm, yes, I have the same problem 13:54 <nicolaas_> I can try to assign it to Dean 13:54 <nicolaas_> he is able to reproduce this consistently 13:54 <stuartf> that might be for the best 13:55 <stuartf> the only thing I can figure is that it might be only certain versions of ie8 13:55 <stuartf> I'm on 8.0.6001.18702 13:55 <nicolaas_> that's definitely possible 13:56 <stuartf> I don't know how anyone can stand to use that browser on a daily basis 13:56 <nicolaas_> It boggles my mind as well 13:56 <nicolaas_> It manages to screw up every single website 14:12 <stuartf> mrvisser: ping 14:12 <mrvisser> stuartf pong 14:13 <stuartf> I think I've got a puppet config for cassandra 14:13 <mrvisser> stuartf: sweet 14:14 <mrvisser> stuartf: was it pretty straight forward? are you using the yum distro? 14:14 <stuartf> I'm just not completely up to speed on how froese arranged his scripts 14:14 <stuartf> yeah, I'm using the datastax yum repo 14:14 <stuartf> basically just configures the repo, installes the package, and puts cassandra.yaml in place so far 14:15 <mrvisser> stuartf: I can give some pointers on the script arrangement if you'd like 14:15 <mrvisser> will just have to quickly refresh my memory in the repo a bit 14:15 <mrvisser> or if you have some specific questions, I can try my best to answer 14:16 <stuartf> basically I just need to know where you specify that EC2_OAE_PREVIEW should have the cassandra class applied 14:16 <stuartf> for instance 14:17 <stuartf> also, one note, I've hard coded the IPs of the 3 servers as the seeds in cassandra.yaml for now, we'll probably want to make that a bit more flexible in the future 14:19 <mrvisser> stuartf: the nodes.pp in the environment folder you're executing should have a node for preview 14:19 <mrvisser> stuartf: then when you apply the config, you specify what node instance you're applying to the machine 14:21 <mrvisser> how it actually determines which node config to apply would be a good question, actually.. looks like it might look up some domain info of the machine it's being run on. 14:28 <stuartf> mrvisser: so, something like this: https://github.com/stuartf/puppet-oae-example/commit/6f6c5759dca6f893edfd6a81d9579c722c3ce1de 14:30 <mrvisser> stuartf that looks right to me 14:30 <stuartf> I think I just noticed one small issue 14:31 <stuartf> just the template path is off 14:32 <stuartf> ok, so if I wanted to try this on the preview node, I'd check out puppet-oae-example in ec2-user's home directory and run `puppet apply ???` 14:33 <stuartf> or is it already checked out somewhere on all these machines 14:34 <mrvisser> stuartf: might not be checked out on the preview machine yet 14:34 <mrvisser> stuartf: err, actually, it should be 14:34 <mrvisser> stuartf: sudo bin/apply.sh --certname "oae-app1.localdomain" --verbose 14:34 <mrvisser> from the root of the puppet-oae-example directory 14:34 <mrvisser> stuartf: but first, create a file '.environment' that specifies the environment you're applying 14:35 <mrvisser> —certname "…" should be set to the name of the node you're applying, as per nodes.pp 14:35 <stuartf> doesn't look like it's in ec2-user's home directory on preview 14:35 <mrvisser> stuartf: it might be in my user's home directory 14:36 <stuartf> should I move it to ec2-user? 14:37 <mrvisser> stuartf: if you want, you could just create a new one too :) 14:37 <mrvisser> stuartf: which repo are you using? efroese? 14:37 <stuartf> mrvisser: I'm going to use my repo since it has my cassandra branch 14:38 <stuartf> I just didn't want us to have dueling configs 14:39 <mrvisser> stuartf: right. might want to rebase to my master branch first, actually 14:39 <mrvisser> the puppet configs that has your login hasn't gone into efroese's repo yet 14:59 <stuartf> mrvisser: looks like it works 14:59 <mrvisser> stuartf: nice 15:05 <GitHub97> [nakamura] thecarlhall pushed 2 new commits to master: http://git.io/F-h9uw 15:05 <GitHub97> [nakamura/master] KERN-3160 set javax.jcr version to 2.0 - Anthony Whyte 15:05 <GitHub97> [nakamura/master] Merge pull request #1003 from arwhyte/KERN-3160 - Carl Hall 15:07 <GitHub51> [nakamura] thecarlhall pushed 1 new commit to 1.4.2: http://git.io/3n4VgQ 15:07 <GitHub51> [nakamura/1.4.2] KERN-3160 set javax.jcr version to 2.0 - Anthony Whyte 16:53 <dean440> ping kentfitz 16:54 <kentfitz> pong dean440 16:55 <dean440> So I'm trying to debug SAKIII-6144, which I can only reproduce on qa20. Bert suggests I upload a non-production jar of 3akai-ux to qa20 for this. 16:56 <dean440> But I need to know how to restore the hashed files when I'm done 16:57 <kentfitz> dean440: Ah, give me just a second to backup the jar, I think I should be able to just flip back to it when you are done. One moment 16:57 <dean440> Probably wouldn't hurt to know what the 3akai-ux bundle is called in the console either. :) 16:57 <dean440> Sure 16:58 <kentfitz> dean440: Backed up. I believe that bundle is Sakai OAE UX Loaderorg.sakaiproject.nakamura.org.sakaiproject.nakamura.uxloader 16:59 <stuartf> ^ I was just about to say that 16:59 <dean440> karma for both :) 16:59 <dean440> kentfitz++ 16:59 <dean440> @stuartf++ 16:59 <dean440> Tis the season 17:02 <kentfitz> dean440: Just let me know when you are all done and I can flip back to the hashed version. No rush at all. 17:02 <dean440> kentfitz Might be a silly question, but I've never uploaded a replace jar for a bundle in the Felix console and I don't want to screw anything up. How do I do it? 17:04 <kentfitz> dean440: As I recall, just click install/update, find your jar, click Start Bundle and then click the Install or Update button 17:04 <dean440> Okay 17:04 <kentfitz> dean440: If it doesn't look like it's picking it up for some reason, I can always restart the server if needed 17:06 <dean440> kentfitz Sorry I haven't uploaded anything yet 17:06 <kentfitz> dean440: No worries. I'm around, just ping me if you need anything 17:23 <dean440> kentfitz I uploaded a regular 3akai-ux jar, and the problem seems to stop 17:24 <dean440> kentfitz It kind of seems like something about our production build is causing this. 17:24 <kentfitz> dean440: Well, it would not be the first time a problem only manifested on the release build. So hashed files are no longer present? 17:25 <dean440> kentfitz Correct. 17:25 <kentfitz> want me to flip it back to the release build so you can try again? 17:26 <dean440> kentfitz Yes, I was just going to ask you to switch it back 17:26 <kentfitz> dean440: Be just a few, let you know when it is back 17:42 <kentfitz> dean440: Sorry for the delay, give it a whirl now 17:44 <dean440> kentfitz No worries, thanks 17:47 <dean440> kentfitz I was able to reproduce it once. But after that, following the normal steps to reproduce, it's not coming back. 17:48 <dean440> kentfitz Oops, I had my browser set to IE9. 17:48 <dean440> Trying again 17:49 <kentfitz> dean440: Ok, let me know 17:50 <dean440> kentfitz Hmm, it's not failing anymore. Would it make sense to try a server restart to be sure? 17:51 <kentfitz> dean440: I actually restarted the server in the process of getting it back to a hashed build 17:54 <dean440> kentfitz I can't reproduce this issue on qa20 any longer 17:54 <kentfitz> dean440: hrumph. :) 18:00 <dean440> kentfitz All's well that ends well I suppose. :p 18:02 <dean440> kentfitz Have any ideas as to what's going on here, or if this might be safe to call resolved? 18:02 <nicolaas> we've seen this on oae-demo/community as well 18:02 <nicolaas> the people in Australia were/are seeing it all the time 18:03 <kentfitz> dean440: No idea what's going on with it, but it doesn't 'feel' resolved 18:03 <dean440> Well, 18:03 <kentfitz> dean440, nicolaas: One thing I could do is start qa20 back up with 1.4 code 18:04 <kentfitz> see if it's reproducable then 18:04 <kentfitz> and then flip back to 1.4.2 18:04 <dean440> kentfitz PhysX nicolaas I was just able to reproduce it a couple of times on qa20 actually, albeit intermittently. Maybe a racing issue? 18:04 <nicolaas> It almost definitely is 18:05 <nicolaas> in the widget loader 18:05 <nicolaas> or it's a caching issue 18:05 <dean440> nicolaas kentfitz It does not happen very often, which is why I mistakenly reported that it no longer occurs. 18:06 <kentfitz> dean440, nicolaas: Might it help to manifest the problem if you throttled your connection speed back? 18:06 <nicolaas> it's possible, although stuart didn't have any luck 18:09 <dean440> kentfitz on qa20? 18:09 <maximka> ping arwhyte1 _______________________________________________ oae-dev mailing list oae-dev@collab.sakaiproject.org http://collab.sakaiproject.org/mailman/listinfo/oae-dev
