Perl 6 - the future is here, just unevenly distributed

IRC log for #opentreeoflife, 2015-08-03

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

All times shown according to UTC.

Time Nick Message
12:44 jar286 joined #opentreeoflife
15:45 jar286 joined #opentreeoflife
16:38 josephwb joined #opentreeoflife
17:49 kcranstn joined #opentreeoflife
18:11 kcranstn joined #opentreeoflife
18:13 kcranstn joined #opentreeoflife
18:58 jimallman joined #opentreeoflife
19:59 josephwb we should nail down the meanings of the various OTT flags
20:00 josephwb i made some edits to index.html at http://files.opentreeoflife.org/ott/ott2.9/. just typos (e.g. missing underscores, etc.)
20:05 josephwb jar286: is "incertae_sedis_direct" deprecated (like extinct_direct and major_rank_conflict_direct)? seems like it should be, and doesn't seem to be used in 2.9draft8
20:07 josephwb maybe you never used that (i.e. just in treemachine code)?
20:08 jar286 it should be documented as not-used-any-more.  the naming was inconsistent (as was the semantics) so I got rid of all the “_direct” variants
20:08 jar286 but I omitted it from https://github.com/OpenTreeOfLife/reference-taxonomy/blob/2.9draft8-finishing-up/doc/ott2.9.md
20:09 jar286 will fix
20:09 josephwb ok, thanks. just not specifically mentioned in the changes since 2.9
20:09 josephwb i mean 2.8
20:09 jar286 right, will fix
20:11 jar286 I am really stuck on this neo4j business.
20:11 jar286 it times out while starting up.
20:11 josephwb i am out of the loop. what is the business? oti?
20:12 jar286 all the neo4j apps are losing
20:12 jar286 it used to be you could start up treemachine with no db.  now it times out while starting.
20:13 josephwb without a db? never tried that.
20:13 jar286 will try it with a db, but that feels like voodoo (also it takes hours to transfer the db file, so I won’t know for a while)
20:13 josephwb actually, i probably have
20:13 jar286 used to work - it’s part of the normal setup sequence for the deployment system
20:14 josephwb right. yeah, should work. i don't know what could have changed
20:14 josephwb i will try it on my machine
20:15 josephwb are you sure it got completely compiled?
20:15 jar286 how would I know?
20:16 josephwb i will try it later; running synth at the moment, and don't want to mess things up
20:17 josephwb it should print the compilation process, and say something like "FAIL" if it didn't work.
20:17 jar286 it says ‘build success’ three times (jade, ot-base, treemachine)
20:17 josephwb hmm
20:19 josephwb is this on ot10?
20:22 jar286 no, ot24
20:23 jar286 newly provisioned instance (I want a backup api server in case we get deluged)
20:24 josephwb things seem to be running on ot10
20:25 jar286 if the question is what has changed, the answer is that ot24 is running debian jessie ‘natively’ i.e. it’s not an ugrade from debian squeeze
20:26 jar286 yes, ot10 and ot20 are doing just fine.  it’s ot24 that’s failing.  ot24 is a fresh install
20:26 josephwb oh. well, i have no idea then. its got java 8?
20:26 jar286 yes
20:27 jar286 it looks like it’s doing something with lucene when it times out
20:27 jar286 IndexProviderStore
20:28 jar286 does treemachine use lucene?  doesn’t make sense
20:28 josephwb yes, it does
20:28 josephwb use it, i mean
20:34 josephwb i only see it here: https://github.com/OpenTreeOfLife/treemachine/blob/2d0195f8d92bbb4e2c6a8f944b96ed2c45d2e050/src/main/java/opentree/synthesis/RankedSynthesisSubproblemExpander.java
20:34 josephwb but neo4j uses it heavily itself
20:34 josephwb brb
22:09 jimallman jar286: here are a couple of threads regarding neo4j timeouts and similar symptoms. TL;DR suggestions: tweak the ‘neo4j.properties’ file to allow store upgrade; give lots of RAM to neo4j (at least 300 MB); watch for badly mapped hostnames; allow for at least 40,000 open files.
22:10 jimallman http://stackoverflow.com/questions/17446653/how-to-start-neo4j-service-from-debian-package
22:10 jimallman https://groups.google.com/forum/#!topic/neo4j/BNOdOChrEKI
22:10 jimallman https://github.com/neo4j/neo4j/issues/3772
22:11 jimallman don’t know if any of that is helpful. it looks like neo4j error messages can’t be taken at face value.
22:23 jar286 should be helpful, thanks
22:43 jar286 I read the last of these already, a few days ago
22:44 jar286 My failure is the same as Iain’s, and there was no resolution for his woes.

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