Perl 6 - the future is here, just unevenly distributed

IRC log for #opentreeoflife, 2014-06-11

| Channels | #opentreeoflife index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:02 towodo org.opentree.taxonomy.plugins.TNRS.autocompleteBoxQuery - no class def found for org/opentree/exceptions/MultipleHitsException
00:04 towodo so taxomachine on ot10 is not happy.
00:14 mtholder It is not well documented, but https://github.com/OpenTreeOfLife/peyotl/tree/master/examples/ot-api has some scripts for testing that the api described on the wiki for the hackathon is working
00:15 mtholder I was going to turn the service-status-check.py in that dir into a cron-deployed job that tested that the services on api.opentreeoflife.org are working. But I got bogged down with other things...
00:16 mtholder that class is in ot-base, so it looks like another dependencies+maven thing
00:16 mtholder towodo ^
00:16 towodo yes, i just figured that out
00:16 josephwb joined #opentreeoflife
00:17 towodo going to blow away ot-base from .m2
00:17 towodo i wonder what best practice is - delete .m2 on every build? ...
00:17 towodo or use make?  I'm getting very close to plunging in to creating a makefile for this system.
00:21 mtholder make + ant seems like a good combo IMHO. but I don't know how complex it is to deal with neo4j
00:22 mtholder (by which I mean, it may be easier to stick with maven if we are just tweaking neo4j maven examples)
00:22 mtholder -U flag to maven might help
00:22 towodo i was thinking make + maven
00:23 mtholder (but I think -U is for checking online repos for jars)
00:23 towodo ok, flushed ot-base from .m2, and deleted taxo jar, and it's still losing…
00:24 towodo will check on ot-base version
00:25 towodo ot-base: branch master, up to date, rebuilt, installed into .m2
00:27 towodo class org.opentree.exceptions.MultipleHitsException is in there
00:28 mtholder is it in the jar that gets put in the plugin dir?
00:29 towodo yes, according to jar tf … | grep ...
00:30 mtholder is the "neo4j restart" getting called? I think that was 1 problem I had today
00:30 towodo yes
00:30 towodo looking at taxomachine sources again now, i was consulting wrong version
00:31 towodo confusing since there's a 2nd MultipleHitsException class (ironically)
00:31 mtholder Ha!
00:32 towodo I need to get home...
00:32 towodo I will put my deployed-systems changes in a branch and push, but not commit
00:33 mtholder ok
00:33 josephwb joined #opentreeoflife
00:40 towodo bye
01:03 josephwb joined #opentreeoflife
01:10 mtholder joined #opentreeoflife
01:53 towodo joined #opentreeoflife
01:59 mtholder joined #opentreeoflife
03:17 josephwb joined #opentreeoflife
03:17 travis-ci joined #opentreeoflife
03:17 travis-ci [travis-ci] OpenTreeOfLife/phylesystem-api#556 (master - 7db88c5 : Mark T. Holder): The build passed.
03:17 travis-ci [travis-ci] Change view : https://github.com/OpenTreeOfLife/phylesystem-api/compare/2facf54d1700...7db88c5d9426
03:17 travis-ci [travis-ci] Build details : http://travis-ci.org/OpenTreeOfLife/phylesystem-api/builds/27276594
03:17 travis-ci left #opentreeoflife
03:20 josephwb joined #opentreeoflife
04:48 mtholder joined #opentreeoflife
04:51 mtholder well. I've sync'ed phylesystem-1 with phylografter and then added the specifiedRoot and the unrootedTree info from gcmdr
04:52 mtholder so, in theory, we could deploy phylesystem-1 to ot10 and have the curator keep things for posterity.
09:59 kcranstn joined #opentreeoflife
10:40 towodo joined #opentreeoflife
10:51 josephwb joined #opentreeoflife
12:22 josephwb joined #opentreeoflife
12:27 towodo joined #opentreeoflife
12:47 towodo joined #opentreeoflife
12:52 kcranstn joined #opentreeoflife
12:56 mtholder joined #opentreeoflife
13:04 blackrim joined #opentreeoflife
13:12 kcranstn @mtholder: are we pointing people to https://github.com/OpenTreeOfLife/phylesystem who just want to clone the repo?
13:13 mtholder That should work. They have to then run the pull-studies.bash in that repo
13:13 kcranstn is phylesystem-1 better?
13:13 kcranstn actually, the docs are good
13:14 kcranstn for the unversioned one
13:16 mtholder I updated phylesystem-1 last night with info from phylografter and with the blessing of candidateTreesForSynthesis from gcmdr.
13:16 mtholder so we could deploy it if we want to.
13:17 kcranstn yay!
13:20 blackrim mtholder: what do you need from gcmdr
13:22 mtholder i think that I got what I needed. Basically phylografter does not know which trees are preferred for synth.
13:22 blackrim gotcha
13:22 mtholder i grabbed that from the list in gcmdr
13:23 mtholder and the trees in that list are now labeled in the nexsons as being rooted in a non-arbitrary way.
13:23 blackrim fantastic
13:23 kcranstn so now we need to point curator at phylesystem-1 and we are good to start pointing people there?
13:23 mtholder do we have a broader list of studies for which the rooting is correct?
13:24 mtholder if we want to start writing to the permanent repo. Yes.
13:24 kcranstn this is exciting
13:25 mtholder towodo would it cause problems for you if I pushed a different phylesysytem repo to ot10?
13:25 mtholder i know that you were in the midst of a deployment there...
13:26 blackrim mtholder: we kind of do have a broader list, but these have not been confirmed by anyone else and in looking more closely at these they often are wrong, so i will say no
13:36 towodo FYI I'm working on ot10 taxomachine
13:36 towodo no problem pushing any phylesystem to ot10, except that I will have to do a config file merge. no big deal.
13:36 towodo be sure to commit deployed-systems when done
13:37 mtholder will do. jimallman: are you OK with me swapping out phylesystem-1 for phylesystem-0 on ot10?
13:39 towodo still getting the no class defin found error.  flushed .m2, flushed the .jar files for taxomachine and ot-base.  what did i forget?  I guess I can blow away the taxomachine and ot-base repos, but they're pretty fresh
13:40 towodo but I am desperate. i think that's what I'll do
13:40 mtholder you removed the jars from the target and the plugin dirs
13:40 mtholder ?
13:40 towodo rm repo/ot-base/target/ot-base-0.0.1-SNAPSHOT.jar
13:41 towodo rm repo/taxomachine/target/taxomachine-neo4j-plugins-0.0.1-SNAPSHOT.jar
13:41 towodo mv .m2 .m2-20140610
13:41 towodo that's what I did just before './push.sh … taxomachine'
13:42 mtholder hmmm. did you try cussing? I did that alot yesterday. seemed to help...
13:42 towodo maybe the ot-base repo needs a refresh… no, 'git fetch' does nothing. branch = master.
13:43 towodo same in taxomachine repo
13:45 towodo taxomachine imports that class from ot-base… the compile-time dependency is satisfied otherwise we'd get a compiler error...
13:46 towodo but it's not in the taxomachine .jar file.
13:46 blackrim towodo: let me see if cody is around
13:46 towodo there is a class with the same name but in a different package.  that should be irrelevant
13:46 towodo looking at pom.xml
13:46 mtholder does moving the ot-base jar into the neo4j plugins dir for taxomachine help?
13:47 towodo will try.
13:47 towodo pom.xml looks fine to me
13:47 mtholder I guess there is also a neo4j.../lib dir that you could try
13:47 towodo pom.xml is the wrong file, it's pom.serverplugins.xml
13:48 towodo yes, there is a lib/ , will try in a sec
13:48 towodo pom.serverplugins.xml  looks fine
13:49 towodo i gotta go feed my parking meter.  back in ten minutes, then will try putting ot-base[etc].jar in neo4j/lib
13:49 josephwb joined #opentreeoflife
13:49 mtholder or the good ole CLASSPATH env var hack when launching neo4j might work.
13:54 blackrim towodo: ok, cody is in, any one have the kiwiirc link handy to get him on here
13:54 codiferous joined #opentreeoflife
13:55 blackrim codiferous: you made it to irc!
13:55 codiferous i did!
13:55 mtholder but we lost scrollback. I miss that bot already
13:55 scrollback joined #opentreeoflife
13:56 mtholder woohoo it's back
13:56 blackrim phew
13:56 kcranstn it’s a proper opentree party now
13:56 mtholder is there an anti-Turing test where I try to convince ilbot3 that I'm one of them?
13:58 jimallman mtholder: phylesystem-1 on ot10 is fine by me
13:58 mtholder ok. I'll try it.
14:01 jimallman kcranstn: i’ve added the server-details link to the footer in both web-apps. It’s mostly a dump of the private/config file, but the curator app also lists all shards (repo names) used by its phylesystem-api instance. (this information was surprisingly hard to retrieve!)
14:01 jimallman working now on devtree
14:02 jimallman aaaand she’s gone :(
14:02 mtholder did the phylesystem_config method not work? jimallman
14:02 towodo back
14:02 mtholder i'm pushing the api to ot10 right now towodo
14:02 mtholder if it works, I'll push the config.
14:02 jimallman mtholder: i believe that’s what i ended up using, yes. fetched via ajax, parsed for shard list. i kept expecting to find it closer to home…
14:03 mtholder hmmm. I can move it.
14:03 towodo ok. i'm looking at taxomachine, should be independent.
14:05 towodo I don't get it. ot-base in in the serverplugins jar file, but the needed class isn't being copied into the taxomachine plugins jar file. no wonder it's reported missing.
14:05 towodo jar tf neo4j-taxomachine/plugins/taxomachine-neo4j-plugins-0.0.1-SNAPSHOT.jar | grep MultipleHitsException
14:05 towodo ==>  org/opentree/tnrs/MultipleHitsException.class   which is the wrong class (different package)
14:07 towodo move the ot-base jar file into the neo4j lib directory, and now it works.
14:08 towodo codiferous, any thoughts? not even sure how to report this as an issue.
14:08 markholder joined #opentreeoflife
14:09 blackrim mtholder: by move it do you mean login again
14:09 towodo the weird thing is that taxomachine works on ot12, which is a practically identical setup.
14:10 mtholder joined #opentreeoflife
14:11 mtholder sorry, blackrim. I meant to say to jimallman: i can rearrange the info in the phylesystem_config response.
14:11 towodo I think I will blow away (or put aside) the ot-base and taxomachine repos, and .m2, and try again. there must be some bit of state that's confusing maven. we need for this to be repeatable, not enough that it works this once
14:13 jimallman mtholder: no, it’s all good. i just found it surprising that the curation app is oblivious to the source repos it’s using. i’ve had a notion that we might link from the History tab entries to the actual commits on GitHub, but it would be a minor project to get the URLs required.
14:15 towodo oops… mtholder I need your ot10.config, don't want to break anything you're doing
14:16 mtholder joined #opentreeoflife
14:18 mtholder towodo. I pushed.
14:18 towodo tnx
14:19 mtholder sorry I was having internet connectivity issues and was booted off twice...
14:21 towodo changing ot9 to ot10, and dev to devtree … shouldn't matter for present purposes
14:21 towodo ok… rebuilding ot-base and taxomachine on ot10
14:23 codiferous sorry for the delay. i have never seen that. You said the other MultipleHitsException.class file is not the right one? Where is that one coming from?
14:24 towodo the server plugins .jar file needs to have two classes with that name, different packages.  it has one (the one defined in taxomachine), but not the other (the one defined in ot-base)
14:25 codiferous i wonder if that is the problem.
14:25 towodo I don't see why it would be… different packages means completely different classes, java is designed for that.
14:25 codiferous where is the error occurring?
14:25 codiferous but it could be confusing to maven
14:25 towodo curl -X POST http://devapi.opentreeoflife.org/taxomachine/v1/autocompleteBoxQuery -H "content-type:application/json" -d '{"queryString":"Endoxyla","contextName":"All life"}'
14:26 towodo but don't try that now, it's rebuilding
14:26 towodo "cause" : {
14:26 towodo "message" : "org.opentree.exceptions.MultipleHitsException",
14:26 towodo "exception" : "ClassNotFoundException",
14:26 codiferous stack trace? what line of code is triggering the exception?
14:27 towodo it's in the class loader.
14:27 towodo "Stacktrace" : [ "java.net.URLClassLoader$1.run(URLClassLoader.java:366)", "java.net.URLClassLoader$1.run(URLClassLoader.java:355)", "java.security.AccessController.doPrivileged(Native Method)", "java.net.URLClassLoader.findClass(URLClassLoader.java:354)", "java.lang.ClassLoader.loadClass(ClassLoader.java:425)", "sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:308)", "java.lang.ClassLoader.loadClass(ClassLoader.java:358
14:27 towodo "org.opentree.taxonomy.plugins.TNRS.autocompleteBoxQuery(TNRS.java:128)", "java.lang.reflect.Method.invoke(Method.java:606)", "org.neo4j.server.plugins.PluginMethod.invoke(PluginMethod.java:57)", "org.neo4j.server.plugins.PluginManager.invoke(PluginManager.java:168)", "org.neo4j.server.rest.web.ExtensionService.invokeGraphDatabaseExtension(ExtensionService.java:300)",
14:27 towodo "org.neo4j.server.rest.web.ExtensionService.invokeGraphDatabaseExtension(ExtensionService.java:122)", "java.lang.reflect.Method.invoke(Method.java:606)", "org.neo4j.server.rest.security.SecurityFilter.doFilter(SecurityFilter.java:112)" ],
14:28 towodo z
14:28 towodo ok, it works now, but i had to blow away the taxomachine and ot-base repos to get there.
14:28 codiferous interesting
14:29 towodo I don't want to have to do that on every push.  The way things are now problems like this could easily happen again
14:30 kcranstn joined #opentreeoflife
14:30 towodo I can create a makefile to track dependencies better, but I'm not confident that would catch this problem
14:30 mtholder ok. jimallman et al: phylesystem-1 is now deployed on the api. I was able to save a silly tweak to a study, so I think that we are in business
14:30 towodo I *thought* I was refreshing the repos, deleting the derived files (the .jar files), and rebuilding… that ought to be the same as a clean build
14:30 mtholder one sec...
14:31 * mtholder is redeploying to test a tweak to the deploy script...
14:32 towodo I still have the broken copies of the repos, I could do a recursive diff, I guess
14:33 towodo a big problem is that I have no model of how maven works.  the documentation has a "trust me, just do this" flavor, which is useless for troubleshooting
14:34 codiferous i'm removing the duplicate multiplehitsexception classes, the only one should be the one in ot-base
14:34 codiferous not that that is much consolation regarding maven arcana
14:35 josephwb joined #opentreeoflife
14:41 mtholder OK. I pushed my tweak to the deploy (one line pull request to towodo - no rush).
14:43 towodo I'm not confident that removing the 2nd class will fix the problem, but it's worth a try.  let me know when pushed to master. tnx
14:48 mtholder So, now that I have installed a new version of the data repo. I need to clear out oti and reindex.
14:48 mtholder is that part of the deployment system?
14:48 mtholder towodo^
14:48 towodo right. the answer is complicated. the short answer is reindexing is a deployment command, but clearing out is not.
14:49 towodo there is a piece i have yet to program and i've been doing it manually
14:49 mtholder ok
14:49 towodo i can tell you how though, it's not hard
14:49 mtholder are there docs?
14:49 towodo um. hmm. there may be email
14:49 mtholder is it just the unindexNexsons call?
14:50 towodo the new thing is the ability for oti to make use of taxonomy.
14:50 towodo I don't know anything about that call.
14:50 towodo here's what I would do if you really want a fresh start:
14:50 towodo 1. blow away or set aside the current oti graph.db directory
14:51 towodo 2. install the taxomachine db as the oti db (hang on, i'll get you that command)
14:51 towodo 3. ./push.sh -c … index
14:51 towodo that's it.  step 2 could take 10 minutes, step 3 could take up to an hour
14:52 towodo 2. would be:  setup/install-db.sh downloads/taxomachine.db.tgz oti mtholder    (run on ot10)
14:52 towodo this needs to be either documented, automated, or both.  haven't got to it
14:53 mtholder OK. thanks. here goes...
14:53 towodo need to pay some attention to memory use.  we were fine with this on ot12, IIRC, and ot10 is bigger.
14:53 towodo I'm sort of distracted now, making progress on rehabilitating the RAID array.
15:03 jimallman towodo: fyi - i was wrapping up some other fixes last night, working on ssl config now…
15:03 towodo ok
15:05 mtholder note: the curator app on devtree will not be available until I finish the reindexing steps that towodo mentioned. So if you are seeing it hang, that is why.
15:06 mtholder If you know the study ID, you can still edit it (it is just the study list from oti that is not ready). the urls for editing are like: http://devtree.opentreeoflife.org/curator/study/view/pg_1019
15:13 jimallman mtholder: got it, thanks
15:25 mtholder whoa. step 2 took awhile.... ~30 minutes. on to step 3...
15:28 mtholder step 3 is launched. I'll be off line for a bit.
15:32 josephwb when will the new taxomachine code be up and running? nothing urgent, just testing curl calls
15:36 towodo RAID is back up!  can get all OTT versions now (I hope!).
15:37 towodo I think taxomachine is up and running.  the last test I did passed.
15:37 towodo that is, the curl command in the API docs.
15:37 towodo would be nice to have more testing, but I guess that can be done through the UI.
15:38 josephwb i am testing the getSyntheticTree call
15:38 towodo I would need to check to see whether ot11 is accessing ot10.  may need to update ot11
15:38 josephwb the treeID argument is now optional (as of yesterday)
15:38 towodo that's treemachine
15:38 josephwb doh, yes
15:39 josephwb that code should be updated
15:41 towodo I can do that if it's urgent
15:41 josephwb not urgent
15:41 towodo remind me later if not done...
16:00 towodo ok, going home, where parking and lunch are both cheaper. will work from there. back on line soon.
16:10 mtholder joined #opentreeoflife
16:17 mtholder looks like the indexing (oti indexing phylesystem-1 on ot10) has completed.
16:17 mtholder all systems go. I guess.
16:17 kcranstn is this temporary? http://files.opentreeoflife.org/ott/
16:18 kcranstn (instead of the other page with far more info)
16:41 * jimallman just pushed a tweak to devtree (ot11), to show shard list from the main study-list page of curation app
16:51 jimallman i need to test my SSL (HTTPS) configuration, which may cause some short-term breakage on devtree… I should be able to clean up pretty quickly, or revert if it all goes pear-shaped.
17:13 towodo joined #opentreeoflife
17:14 jimallman towodo: i think something’s blocking port 443 on ot1. what’s the active firewall? (if no answer handy, i’m googling now for typical debian setup)
17:15 towodo you mean ot10
17:15 towodo you mean ot11
17:15 towodo let me check security settings at AWS
17:15 towodo I thought I set them up to allow 443
17:17 towodo AWS says port 443, protocol TCP, any source is allowed.
17:17 towodo did you restart apache?
17:18 towodo config for HTTPS is identical to that for HTTP.
17:18 jimallman towodo: sorry, that’s ot11 (web-app server)
17:18 towodo right, that's what I was talking about (I corrected myself)
17:18 towodo ot10 has no firewall at all
17:18 jimallman trying again with *:443 instead of _default_:443 (in VirtualHost)…
17:19 towodo yes, that would make a difference
17:19 towodo the server doesn't know its own name, after all
17:19 towodo you could also say devtree.opentreeoflife.org:443, but then we'd have to vary that between development and production
17:21 mtholder joined #opentreeoflife
17:22 jimallman no joy… i’ll keep sniffing here. there’s no sign of the expected /var/log/apache2/ssl_error.log, so I’m guessing the HTTPS request isn’t making it to apache for some reason.
17:30 towodo telnet ot11.opentreeoflife.org 443
17:30 towodo Trying 54.189.63.165...
17:30 towodo telnet: connect to address 54.189.63.165: Connection refused
17:30 towodo telnet: Unable to connect to remote host
17:30 towodo bash-3.2$
17:31 towodo ^^^ that means there is no firewall problem.  there is just no process listening on that port.
17:31 jimallman hrm. maybe mod_ssl is not installed..? i’ll check
17:31 towodo (you can see what a firewall block looks like by trying to telnet to 442)
17:31 jimallman thanks
17:31 towodo mods_available I think
17:31 towodo and mods_enabled or something like that
17:34 jimallman yes, also    $ sudo apache2ctl -M
17:34 towodo ?? never seen that before
17:35 jimallman lists all ‘static’ and ‘shared’ modules. nice! ssl doesn’t seem to be listed here, working on it now
17:35 towodo there's a command to install an apache module, don't remember what it's called
17:36 jimallman looks like a2enmod is the recommended tool
17:36 towodo yes. that's it.  we can add that to the deploy system
17:36 towodo assuming it's idempotent...
17:39 jimallman ‘sudo a2enmod ssl’ did the trick… running it again just generates a sensible courtesy message: “Module ssl is already enabled.”
17:39 jimallman this reveals goofy errors in my apache config, chasing these now…
17:39 towodo maybe it's built in to the base apache.
17:46 jimallman progress!
17:46 jimallman https://devtree.opentreeoflife.org/curator
17:47 towodo excellent.
17:47 kcranstn loading… loading…
17:47 towodo not getting the study list, but I assume that's a separate problem
17:47 jimallman odd, it loads quickly for me
17:47 towodo interesting, the browser eats CPU while waiting
17:50 jimallman just out of curiosity: Try the “Show/hide server details” link in the footer. Does this give you a shart list at the top?
17:50 jimallman (that should hit the phylesystem-api and show ‘phylesystem-1’)
17:50 jimallman s/shart/shard
17:50 kcranstn footer doesn’t show up while the “loading” message is there
17:50 jimallman :-/
17:50 kcranstn still loading, BTW
17:51 jimallman odd. maybe clear browser cache and retry?
17:51 jimallman could be a JS snafu from the footer stuff, this was just fixed awhile ago.
17:51 towodo still just busy-hangs for me. can't see footer.
17:52 towodo I tried https://devtree.opentreeoflife.org/ and it says 'whoops'
17:52 jimallman check for JS errors? (in most modern browsers: right-click on the page, choose Inspect Element, switch to Console tab)
17:52 kcranstn re-loaded and works for me - loads quickly on chrome and firefox
17:53 towodo fixing this taxomachine problem, which turns out to be the same problem that vexed me a few weeks ago, is going to require a chunk of hacking.
17:53 towodo need to introduce a makefile that runs on the server.
17:53 kcranstn can I delete that ‘please delete’ study?
17:53 jimallman please do
17:54 jimallman (it’s literally asking for it)
17:54 mtholder what could be clearer?
17:54 mtholder ;-)
17:54 kcranstn delete and confirmed but still shows up in list
17:54 kcranstn but get error when I try and load
17:54 jimallman oti must be lagging..?
17:54 kcranstn (still in oti, right?)
17:54 mtholder yup
17:55 kcranstn want me to create an issue for that?
17:55 mtholder fwiw curator loads for me in http by not https
17:55 towodo hmm. maybe firefox caching lossage.  i can see the page just fine from chrome
17:55 jimallman we can try this one, also tagged “delete me”: https://devtree.opentreeoflife.org/curator/study/view/2584
17:56 jimallman for sure these pages are a mix of HTTP and HTTPS, so maybe the browsers are balking…
17:56 josephwb can't get into http://dev.opentreeoflife.org
17:56 josephwb oops, nevermind.
17:56 josephwb it is back
17:56 kcranstn from that https link, I can’t get back to the study list (i.e. loading message...)
17:57 jimallman did you get a generic Apache page? plain HTML, “this is the default page”? cuz i’m seeing that once in awhile.
17:57 jimallman josephwb: ^
17:57 josephwb should i not upload a study at this point?
17:57 kcranstn nope, just hanging while loading thes study list
17:58 jimallman josephwb: i think if you stick to http, you should be OK. but i might be messing around here shortly, and i’d hate to block your changes..
17:58 jimallman kcranstn: do you have a JS console handy?
17:58 josephwb i can wait
17:59 jimallman thanks josephwb , i’ll holler when i’ve got this cleared up.
17:59 josephwb ok
17:59 towodo do we need another server for testing?… ot7 is available
18:00 towodo mtholder, I figured out the taxomachine problem, and it was another D'oh! moment.
18:00 jimallman hm, i suppose we could point that at services on ot10 and ot12 (match config for ot11)
18:00 kcranstn “Blocked loading mixed active content "http://ot10.opentreeoflife.org/oti/ext/QueryServices/graphdb/findAllStudies""
18:00 mtholder towodo: what was it?
18:01 jimallman kcranstn: thanks, that definitely helps. “mixed content” meaning a mix of HTTP and HTTPS. this will take a little time to sort out… just curious, what browser is blocking this?
18:01 kcranstn learn more link points me here: https://developer.mozilla.org/en-US/docs/Security/MixedContent
18:01 kcranstn firefox
18:01 jimallman got it, thanks
18:02 towodo I think it has to do with taxomachine getting rebuilt only once instead of twice.  I did a diff of the taxomachine checkout losing vs. working and the only differences were the .jar files
18:03 towodo remember there are two taxomachine .jar files, made by separate builds. only one build was happening. not sure i understand it completely.  but the fix I think is to use make.
18:03 mtholder make certainly sounds good to me.
18:08 towodo kcranst, yes the index page on files.opentreeoflife.org/ott/ is temporary.  i have the right version and will put it in place
18:12 kcranstn thanks, @towodo
18:16 mtholder kcranstn: we have an issue about oti getting out of sync: https://github.com/OpenTreeOfLife/phylesystem-api/issues/89
18:16 mtholder (though it probably needs a more general name)
18:17 mtholder to my knowledge this morning was the first time that we've cleared away the oti db and started fresh. previously it was a bit unclear if we just had cruft from previous testing filling oti.
18:23 jimallman i’ll check the webhook that nudges oti. it’s likely that this needs to mirror some of our recent configuration changes.
18:23 mtholder quite likely. I can check later if you are in the midst of https hell
18:25 jimallman no prob, i’m already there. it should be nudging ot10: http://ot10.opentreeoflife.org/api/search/nudgeIndexOnUpdates
18:27 jimallman i’m not seeing the study deletion. nudging oti now (from GitHub settings page)…
18:29 towodo I never set up the webhook for ot10. my bad.
18:29 jimallman no, it’s already there. i think the problem is that changes aren’t being pushed to GitHub:
18:29 jimallman https://github.com/OpenTreeOfLife/phylesystem-1/tree/master/study/pg_19
18:30 jimallman note that pg_1019 (“delete me”) is still here
18:31 mtholder doh. let me check on the server...
18:31 towodo kcranstn, index page for files.../ott/ is recovered.  working on recovering a few more of the files that were there
18:32 towodo the right thing is to put it on s3.  you can now host sites from s3
18:33 towodo jimallman, i just restarted apache on ot10 (without clobbering apache-admin first, i.e. not through deployment system)
18:33 towodo sorry, it was in a script, forgot it was going to do that
18:33 jimallman ok, no prob
18:33 mtholder looks like a bug in phylesystem-api...
18:34 mtholder the WIP is not merging with master...
19:10 mtholder fix implemented, need to test...
19:37 travis-ci joined #opentreeoflife
19:37 travis-ci [travis-ci] OpenTreeOfLife/phylesystem-api#557 (master - a5247be : Mark T. Holder): The build passed.
19:37 travis-ci [travis-ci] Change view : https://github.com/OpenTreeOfLife/phylesystem-api/compare/7db88c5d9426...a5247beed518
19:37 travis-ci [travis-ci] Build details : http://travis-ci.org/OpenTreeOfLife/phylesystem-api/builds/27341326
19:37 travis-ci left #opentreeoflife
19:42 mtholder OK, so the DELETE verb is now working correctly (the WIP with the study deleted is merged back to master), but the deleted study is still showing up in the study_list. I'll check the logic on the nudging of oti.
19:55 towodo ok. spent too much time on files.opentreeoflife.org.  it's back now.  RAID works too, so multiple copies of all files.
20:00 mtholder great. thanks
20:01 towodo everything that we expect to be working, seems to be working.  I'm not going to launch into this makefile project right now.  anything else I need to do right now?  (jimallman / mtholder)
20:01 towodo ot3 doesn't point to ot10 any more… good...
20:01 mtholder I think we're good (wrt deployment).
20:02 mtholder we have an issue with ot:studyId fields, which may be causing our weirdness with oti not detecting file deletion...
20:03 mtholder I implemented aliasing of study # to pg_# in the phylesystem API. I think that I did this for the sake of backward compatibility with the phylografter study IDs
20:03 mtholder (and the fact that phylografter does not use a pg_ namespace
20:03 mtholder ).
20:04 mtholder but now it is confusing because the ot:studyId field in the NexSONs is sometimes pg_# and sometimes just a #
20:04 towodo at this point I'd say program defensively and update opportunistically, rather than grovel over the entire phylesystem repo to fix field values… collect together a bunch of normalizations and do them all at the same time
20:04 towodo but ymmv
20:06 towodo ot10 still says 'opentree_branch opentree production'
20:06 towodo that's not right.
20:07 towodo will fix in a branch and submit to jimallman for review
20:07 mtholder I think that the only reason we deploy opentree to that machine is for the wonky routes.py side effect on web2py
20:08 mtholder but yes. it should go to master
20:09 towodo removing opentree from list of components… and removing the opentree_branch command
20:12 towodo assigned to mtholder
20:16 mtholder looks good
20:19 codiferous joined #opentreeoflife
20:22 towodo I'll send some email reviewing state of hosts
20:34 mtholder joined #opentreeoflife
20:43 towodo oh yes… routes.py …
20:44 towodo damn, I didn't take notes.  but I think I kludged it, i see this in push_api in push.sh :
20:44 towodo rsync -p -e "${SSH}" ../SITE.routes.py "$OT_USER@$OPENTREE_HOST":web2py/routes.py
20:45 towodo with an apologetic comment.
21:05 1JTAAEY9O joined #opentreeoflife
21:18 josephwb joined #opentreeoflife
21:29 mtholder the studyId thing does appear to be the source of the oti sync issues. I'll see if I can fix it tonight....
21:29 jimallman good to know, thanks
21:32 mtholder left #opentreeoflife
22:01 josephwb joined #opentreeoflife
22:04 josephwb joined #opentreeoflife
22:06 josephwb joined #opentreeoflife
22:21 josephwb joined #opentreeoflife
22:51 josephwb joined #opentreeoflife
23:07 mtholder joined #opentreeoflife
23:19 josephwb joined #opentreeoflife
23:27 kcranstn joined #opentreeoflife
23:59 mtholder joined #opentreeoflife

| Channels | #opentreeoflife index | Today | | Search | Google Search | Plain-Text | summary