Log of the #duraspace-ff channel on chat.freenode.net

Using timezone: Eastern Standard Time
* ksclarke leaves02:14
* fcrepo-bot joins02:24
* fcrepo-bot leaves03:31
* lasseaagren joins05:06
hello .. with fedora 4.0.0-beta3 there was a fcrepo-webapp-4.0.0-beta-03-auth.war that made deployment for me really easy. With newer versions that doesn’t exist anymore - or am I missing something?05:08
* lasseaagren leaves06:47
* dwilcox joins07:49
* awead joins08:58
* acoburn joins09:21
* jgpawletko joins09:24
* github-ff joins09:45
[fcrepo-camel] awoods closed pull request #39: fixed travis-ci link (master...fix-links) http://git.io/heLAnQ
* github-ff leaves
* github-ff joins09:49
[fcrepo4] escowles created baseurl-system-property (+1 new commit): http://git.io/As-37g
fcrepo4/baseurl-system-property 00ad354 Esmé Cowles: Using System property to pass baseURL instead of ObservationManager (FCREPO-1272)
* github-ff leaves
* travis-ci joins10:02
fcrepo4/fcrepo4#3290 (baseurl-system-property - 00ad354 : Esmé Cowles): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/commit/00ad35402210
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/47253234
* travis-ci leaves
* ksclarke joins10:05
awoods: meant to let you know that despite not being on the next sprint I expect to have some time to put towards the grinder tickets over the next two weeks10:06
<awoods>ksclarke: That is great, thank you. Are you still thinking that packer.io is a good fit?10:07
<ksclarke>I am for what it does, the current issue seems to be with the aws cli tools -- packer can really only build the images but there is the additional work of associated them with each other which has to be done via aws cli10:09
I suspect the lack of a vpc may be what's causing aws cli to fail for you (that my commands are implicitly expecting that to be in place)10:11
think I need to create another test aws account without the vpc to verify
* acoburn leaves10:22
<awoods>ksclarke: I would like to also explore the scenario of: I already have some Grinder/F4 AMIs, I now want to spin up the test framework and run some tests.10:23
<ksclarke>Grinder AMIs generated via packer (which have the associations built in -- which is done via aws cli) or different grinder AMIs?10:24
without the connections, you'd have to go in and spin up grinder on each agent machine, pointing it to the console machine10:25
* mikeAtUVa joins
<awoods>ksclarke: When you spin up and AMI, you can pass in "user data", which could have the connection/console details.10:26
ksclarke: When you spin up an AMI, you can pass in "user data", which could have the connection/console details.
<ksclarke>can that use data be used by scripts on the machine?
<awoods>mikeAtUVa: when you have a chance, could you please select/add a JIRA avatar?
<ksclarke>(an alternate option is to get an elastic ip for the console)10:27
<mikeAtUVa>awoods: ok...
<awoods>mikeAtUVa: thanks
ksclarke: sure... although we may want to dedicate several people in a sprint to be running tests.10:28
<ksclarke>(I was trying to avoid the elastic ip option because it's several steps, assuming you didn't want to have one around unattached to the console when it wasn't up)
sure passed in "user data" can be accessed by scripts on the ec2 instance?
I've never tried that10:29
<awoods>ksclarke: "user data" is available to the ec2 instance. I have a fairly high degree of confidence that it could be used by scripts on that instance.10:30
<ksclarke>okay, I'll make a note to look into that first then... won't return to this until next week though
<awoods>ksclarke: I definitely like the idea of packer.io helping to create AMIs, but it makes less sense to me that a new set of AMIs need to be created every time I want to run some tests.
<ksclarke>right, that's not needed10:31
<awoods>ksclarke: really?
<ksclarke>you can keep the AMIs around...
you also have to keep the console around
<awoods>ksclarke: With the AMIs that have been created by packer.io, I can now spin them up and run tests?10:32
<ksclarke>the agent ec2 instances can be deleted, but the agent AMI keeps a record of the EC2 console instance so it knows how to connect new agents to it
yes, as long as you've kept the console instance around
<awoods>ksclarke: you mean that once I terminate the console instance, I have to rebuild the AMIs?10:33
<ksclarke>so you keep the console around and then you can spin up agents from the agent AMI whenever you need them -- that doesn't have to be automated
yes, as it is currently
<awoods>ksclarke: That is not exactly ideal.
ksclarke: how about if I "stop" the console instance instead of "terminating" it?10:34
<ksclarke>doesn't seem that bad to me to have a stopped console instance
right, stopping is fine... it doesn't have to be running all the time
the reason you can't terminate it though is that the agent AMI remembers the instance id for the console (which would change on termination/recreation but not stopping/starting)10:35
but I can look into passing in the user data and having the startup script use that when it starts the agent
<awoods>ksclarke: yes, that seems like that would do the trick. That would also allow me to use the AMIs that you created (if you shared them with me, for example) to run tests.10:36
<awoods>ksclarke: that would be perfect10:37
* ksclarke nods
<awoods>ksclarke: I would like to have some project AMIs that others can simply spin up and start testing with. It would remove potential variability.10:38
* ksclarke has added note to the ticket for next week10:40
* dwilcox leaves
* travis-ci joins10:53
fcrepo4/fcrepo-camel#93 (master - 675a9ae : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4/fcrepo-camel/compare/ebfa035cfbba...675a9ae5ae8f
Build details : http://travis-ci.org/fcrepo4/fcrepo-camel/builds/47252836
* travis-ci leaves
* dwilcox joins10:55
<awoods>mikeAtUVa: here is the link: https://jira.duraspace.org/secure/ViewProfile.jspa10:59
* dwilcox leaves11:24
* dwilcox joins11:25
ruebot: I'm putting together an OR2015 submission for a panel featuring representatives from institutions migrating to Fedora 4. Are you interested in participating?11:58
<ruebot>dwilcox: yessir!11:59
<dwilcox>ruebot: Cool, thanks. I'll share a draft with you before I submit
* ruebot still has to start the islandora proposal
dwilcox: excellent! i spent a bit more time on the upgration page for york yesterday. SUPER VERBOSE.12:00
<dwilcox>ruebot: Yeah, I saw that - great work! We can use this as an expanded template for other pilot projects; I think it captures most of the info we would want
* acoburn joins12:08
<ruebot>dwilcox: yay!12:12
<awoods>nice picture, mikeAtUVa.12:34
* dwilcox leaves12:57
* dwilcox joins12:58
<awoods>osmandin: coming?13:29
osmandin: https://wiki.duraspace.org/display/FF/2015-01-16+Sprint2015-1+Planning+Meeting
<ruebot>dwilcox, awoods: islandora foundation & the islandora/fedora4 project has the money to send me to code4lib. danny will be there too for dgi. do you think it is worth it for me to be there as well to meet with y'all?13:36
<dwilcox>ruebot: In a meeting right now but Andrew will respond when we're done. Answer is likely yes.13:37
<ruebot>cool. i'm around all afternoon.13:38
* osmandin leaves14:22
* mikeAtUVa leaves14:29
* github-ff joins14:40
[fcrepo-camel] acoburn opened pull request #40: added endpoint option for https protocol (master...ssl-support) http://git.io/Yag4pQ
* github-ff leaves
<awoods>ruebot: definitely come. We can organize an F3/4 migration meetup.15:30
<ruebot>awoods: yessir!15:40
<awoods>ruebot: that is good news.
<ruebot>awoods: it is excellent news!15:41
* github-ff joins
[fcrepo-camel] awoods pushed 2 new commits to master: http://git.io/Byjh-g
fcrepo-camel/master 77b51d4 Aaron Coburn: added endpoint option for https protocol
fcrepo-camel/master 7727e66 Andrew Woods: Merge pull request #40 from acoburn/ssl-support...
* github-ff leaves
<ruebot>awoods: is there a pre-conference thing i need to be at?15:43
* ruebot looking at the proposals now
not seeing anything15:44
<awoods>ruebot: not really. Right now, the only F4-specific sessions have been started by Hydra on Thursday.
<awoods>ruebot: but that session is cross-project, although Hydra folks initiated it.
<ruebot>awoods: i don't mind hanging out with hydranauts if i'm allowed to15:45
<awoods>ruebot: I will send a note out to see if folks are interested in another F3/4 side-event.
* travis-ci joins15:49
fcrepo4/fcrepo-camel#95 (master - 7727e66 : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4/fcrepo-camel/compare/675a9ae5ae8f...7727e66bb82c
Build details : http://travis-ci.org/fcrepo4/fcrepo-camel/builds/47293828
* travis-ci leaves
* dwilcox leaves15:54
* jgpawletko leaves16:30
* acoburn leaves17:00
* ksclarke leaves18:11
* ksclarke joins18:29
* awead leaves21:29

Generated by Sualtam