Evergreen ILS Website

IRC log for #evergreen, 2016-03-02

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat

All times shown according to the server's local time.

Time Nick Message
00:07 sandbergja joined #evergreen
01:04 pinesol_green [evergreen|Dan Pearl] LP#1312699 - Add feature to allow user to edit their checkout history. - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=6987fc3>
01:04 pinesol_green [evergreen|Jake Litrell] LP#1312699: Whitespace cleanup. - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=712f588>
01:04 pinesol_green [evergreen|Jake Litrell] LP#1312699: Minor modifications to patron checkout history deletions - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=a1120ea>
01:16 pinesol_green [evergreen|Dan Scott] LP#1507845: Improve search engine optimization - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=7d92d37>
01:29 kmlussier joined #evergreen
01:29 mceraso joined #evergreen
01:29 bshum joined #evergreen
02:43 mnsriz joined #evergreen
02:46 book` joined #evergreen
06:55 csharp @quote add < jeff> reminds me of an old math joke. "what did the acorn say when he grew up?" ... "GEOMETRY!"
06:55 pinesol_green csharp: The operation succeeded.  Quote #140 added.
07:35 rjackson_isl joined #evergreen
07:50 ericar joined #evergreen
07:56 Dyrcona joined #evergreen
08:05 Mark__T joined #evergreen
08:16 _bott_ joined #evergreen
08:40 mmorgan joined #evergreen
08:44 rhamby joined #evergreen
08:45 mmorgan left #evergreen
08:51 jeff okay, of 83 unique barcodes having SIP2 failures with a 5s timeout, only 67 have errors with a 30s timeout.
08:55 jeff 22 patrons with hold_items / unavail_holds errors, 45 patrons with fine_items errors.
08:57 Dyrcona Sounds about right.
08:58 * Dyrcona figures we have more than that, but we have about half a million patrons, depending on how you count them.
08:58 * jeff nods
09:00 jeff i've found encoding/length errors that i can fix and re-test, and i've found a failure related to holds on titles with zero volumes, which i can also fix and re-test.
09:01 mmorgan joined #evergreen
09:05 Dyrcona jeff: You know you can set the encoding per SIP2 account, yeah?
09:05 Dyrcona We use ascii for a couple.
09:06 jeff yeah.
09:06 Dyrcona It started with accents in call numbers of all things.
09:07 Dyrcona Then in patron names.
09:16 jeff I don't remember if the encoding makes a difference with this particular issue, but I think I see the bug.
09:22 jeff 2016-03-02 07:02:50 hostname SIPServer.pm: OILS: Patron->unavail_holds()
09:22 jeff 2016-03-02 07:02:50 hostname opensrf_sip[31789]: Reaping child 8588
09:22 jeff (heh)
09:24 Dyrcona If you think UNIX is family-friendly, remember that all children must die. ;)
09:28 lualaba joined #evergreen
09:29 lualaba Hello. Barcode not visible in Opac. every thing is looks like. asset.calls,asset.copy in staff client is visible. only differense is asset.opac_visible_copies (not insterted in this table)
09:30 lualaba which process store into this table?
09:34 pastebot "lualaba" at 64.57.241.14 pasted "Barcode is not visible" (16 lines) at http://paste.evergreen-ils.org/35
09:35 phasefx lualaba: could try select asset.refresh_opac_visible_copies_mat_view();
09:36 lualaba No function matches the given name and argument types. You might need to add explicit type casts.
09:41 bwicksall joined #evergreen
09:44 lualaba evergreen=# select count(*) from asset.opac_visible_copies where copy_id is null;  count  -------      0 (1 row)
09:47 yboston joined #evergreen
09:47 lualaba please help to fix this problem. asset.opac.visible_copies not updated
09:49 csharp lualaba: can you pastebin the full output of what phasefx suggested?
09:50 pastebot "lualaba" at 64.57.241.14 pasted "refresh_copies" (6 lines) at http://paste.evergreen-ils.org/36
09:50 jeff lualaba: you have a typo -- try again, exactly as phasefx suggested
09:51 lualaba same
09:51 jeff lualaba: can you paste (here) the command/query that you are executing?
09:52 lualaba select asset.refresh_opac_visible_copies_mat_view();
09:53 jeff okay, it appears that your database is missing critical functions, or may be very old?
09:53 lualaba 2.8.1
09:53 jeff lualaba: what version of Evergreen is this?
09:53 jeff hrm.
09:55 mmorgan1 joined #evergreen
09:55 jeff yes, it appears that your database may be missing at least one function. that wouldn't *cause* this symptom, but it may be another symptom of a larger problem.
09:55 lualaba solution?
10:05 lualaba i rebulid the copy OPAC visibility cache.
10:12 berick kmlussier: looking at bug 1312699, i'm curious if you tested disabling circ history, where it deletes all history items, after applying the patch.
10:12 pinesol_green Launchpad bug 1312699 in Evergreen "Editable Checkout History" [Wishlist,Fix committed] https://launchpad.net/bugs/1312699
10:13 Lualaba_ joined #evergreen
10:13 kmlussier berick: I can't remember now. I know I tested far enough to get the warning for disabling. Are you having trouble with it?
10:14 berick kmlussier: i haven't tried, i just saw something in the code that caught my eye as a potential issue
10:14 * kmlussier can take another look
10:15 Lualaba_ Upgrade to 2.84 will.fix my problem?
10:16 Dyrcona Lualaba_: Not likely if you're already missing things.
10:17 Lualaba_ How can I fix?
10:17 Dyrcona Lualaba_: what's the output from: select count(*) from asset.opac_visible_copies where copy_id is not null;
10:18 Lualaba_ 0
10:18 Dyrcona Lualaba_: You'll very likely need to find out what is missing and recreate it in the database.
10:18 Lualaba_ Compare to data schema?
10:19 Dyrcona Lualaba_: Start with the asset.refresh_opac_visible_copies_mat_view() function.
10:19 Dyrcona Find the code in the Open-ILS/src/sql/Pg/ directory that creates that function and run it in the database.
10:20 Dyrcona I'd copy and paste just that create function call to start.
10:20 Dyrcona I forget what file that is in.
10:20 Lualaba_ Ok thank you I will check
10:21 Lualaba_ After rebuild functions what is next step to fix existing problem?
10:21 kmlussier berick: Figures. I think that's the one test that I did on your branch that I didn't fully try out on jlitrell's branch. You're right. That's not clearing out the circ history.
10:23 sandbergja joined #evergreen
10:23 berick kmlussier: ok, i think I know why.  I'm create a patch
10:23 kmlussier berick++
10:23 Dyrcona Lualaba_: Try running the function.
10:25 Dyrcona kmlussier & gmcharlt: I'm updating master on my dev system because we want to look at the checkbox to exclude electronic resources.
10:25 Dyrcona It shouldn't affect what you're doing, I hope.
10:25 gmcharlt joined #evergreen
10:25 Dyrcona hah.
10:26 Dyrcona I have not pulled in the last wip branch changes, but if you want me to, I can.
10:26 gmcharlt joined #evergreen
10:28 berick heh, just updated to master, now when I look at the catalog, I see some bare TT text just udner "Link 1" -> ctx.metalinks.push('');
10:30 bshum berick: Oops, maybe that's related to the patch from dbs that I pushed.  I think that had a metalink thingy in it.
10:30 mllewellyn joined #evergreen
10:33 jwoodard joined #evergreen
10:34 berick kmlussier: fix pushed to tip of working/user/berick/lp1507845-circ-id-query-repair
10:35 bshum berick: In /openils/var/templates/opac/home.tt2, if I move the ctx.metalinks above the ctx.page_title, it works better
10:35 bshum So I think it might be a missing semicolon or some other thingy
10:36 kmlussier berick: OK, loading it on a server now
10:36 bshum Oh no it's just the way the process works I guess
10:37 berick bshum: as it stands, ctx.metalinks.push is outside of the [% ... %] block, so TT thinks it's supposed to be bare text (html, etc.)
10:37 bshum berick: Ah that'll do it
10:38 berick so, yeah, moving it above page_title will work
10:38 bshum I see the %] now
10:38 bshum berick++
10:48 jeff that's interesting. every patron with failing fine_items SIP2 retrieval is an... old barcode schema.
10:49 Dyrcona kmlussier | gmcharlt: I had to kill a psql on my dev vm to reload the database.
10:49 jeff but i know that "old bibs" aren't the only thing that we've tripped over, but those newer bibs i know have caused issues with item info messages, not patron info with details.
10:49 * jeff mumbles to himself
10:49 gmcharlt Dyrcona: ok
10:49 jeff off to see if we're constructing the title/author via different means in those two messages
10:50 Dyrcona kmlussier asked me to update the activity metric branch and I'm confused by what I saw after a pull, so I figured the safe thing was nuke the db and start over.
10:51 * berick pushes a fix to the home template to working/user/berick/lp15078​45-seo-home-template-repair
10:53 * Dyrcona is starting to hate pg_restore.
10:53 kmlussier berick: I can test that one too
10:53 berick kmlussier++
10:55 Dyrcona That's nice. Pg says my superuser isn't a superuser.
10:56 mrpeters joined #evergreen
10:58 Dyrcona Now, pgadmin lets me into the postgres database with that user, but tells me the password is wrong for any other database on that server.
10:58 Dyrcona And my .pgpass looks good.
10:59 Dyrcona And, it worked yesterday.
10:59 * Dyrcona suspects ssl updates.
10:59 * Dyrcona installs updates and bounces postgresql.
11:00 Dyrcona And now, I cannot connect to the server at all.
11:00 lualaba_ joined #evergreen
11:04 dbwells gmcharlt: Are we feature frozen right now?  I thought we were, yet feature code appears to still be going in.  I am not looking to push anything specific, just seeking clarification.  Thanks!
11:06 gmcharlt dbwells: still slushy; I will be making a pass through pull requests today; any that are sufficiently non-disruptive may make it in, but it will definitely be frozen when I cut the beta this evening
11:06 dbwells gmcharlt: okay, thanks again
11:09 * dbwells had reviewed a branch and was holding off, but it got pushed in already anyway :)
11:17 jeff item info messages don't have author, and author is the most common place where we encounter accented characters...
11:17 Dyrcona So, for those not following along in #postgresql: my account's pg password expired at midnight last night.
11:18 jeff but even that doesn't seem to explain why 100% of my "failing fine_items detail SIP2 request" patrons are "older" patron barcode schema.
11:18 jeff Dyrcona: ah!
11:20 mdriscoll joined #evergreen
11:23 lualaba_ Hello i add two functions: asset.refresh_opac_visible_copies_mat_view(); and asset.cache_copy_visibility which solved my problem.
11:24 lualaba_ Do i need add or replase something else?
11:24 berick guessing we're not having today's regularly scheduled dev meeting, since we had one recently
11:24 jeff lualaba_: great! i worry about how they came to be missing in the first place.
11:25 lualaba_ i create DB with standart method not first time
11:26 Christineb joined #evergreen
11:27 kmlussier berick: Yeah, we talked about that yesterday, but then I guess we never followed through with canceling. gmcharlt was thinking it would be good to have one sometime between beta and the full release.
11:29 kmlussier berick++ #Fix to circ history looks good.
11:30 berick great, thanks for confirming
11:30 lualaba_ i have another problem may it also issue with functions? icos e-book, audio etc not vicible in opav but label ##URI## exist in DB also MArc is good
11:41 dbwells Hello all.  I have a question about using the translator vs the gateway and am looking for some insight.
11:42 dbwells I am trying to retrieve a pretty large group of something (3832 objects).
11:43 dbwells The query actually finishes though the gateway, but if I try the translator, it runs for a while (maybe 30 seconds?), then I get a 500 error for the server unexpectedly closing the connection.
11:43 dbwells I've tried bumping up various timeouts, but nothing helps yet.
11:44 berick dbwells: have you looked at the apache error log?  500 error is unusual and does not suggest a timeout
11:44 dbwells I can't find the error in any log, which is frustrating, so I am not even sure what layer is dying.
11:44 berick :)
11:44 dbwells I agree with the 500 error being odd.
11:45 dbwells I am using LWP to make the request.
11:47 berick hmm, then my next guess is the LWP client code is not calling the translator correctly, with all the required headers, etc.
11:47 dbwells It is a pcrud request.  If I do the same request in srfsh, it seg faults, but I am guessing that's unrelated (probably a size problem, I am guessing).
11:47 berick well, that could def. be related
11:47 dbwells okay
11:48 berick srfsh and translator use the same code underneath
11:48 dbwells All I have learned at this point is that the gateway is somehow more robust than srfsh and the translator.
11:48 dbwells :)
11:49 dbwells Is the gateway more distinct from those two?
11:49 Dyrcona dbwells: Did you try a streaming method? That should give you chunked responses, IIRC.
11:49 Dyrcona Oh, I should have read about the pcrud bit.
11:50 gmcharlt dbwells: I'm now testing bug 1548147 as a possible beta candidate
11:50 pinesol_green Launchpad bug 1548147 in Evergreen "Quick Export Feature for Vandelay Queues" [Wishlist,New] https://launchpad.net/bugs/1548147
11:51 dbwells gmcharlt: cool, thank you
11:51 berick dbwells: code underneath is the same for all 3.  big difference is gateway will collect all data from a streaming request, then return an array of the full data set, whereas the translator will relay each response of a streaming request to the caller.
11:52 berick and pcrud is streaming, unless you tack .atomic on the end
11:52 Dyrcona yes, i realized that after I said that about the streaming bit.
11:53 dbwells berick: right.  I was trying to use streaming/translator because I know it's a big request.  Sorta surprised the gateway finished at all for this.
11:53 * Dyrcona shuts up now, 'cause he's only ever logged in through the translator.
11:53 bmills joined #evergreen
11:54 pinesol_green [evergreen|Bill Erickson] LP#1312699 Circ history clear-all repair - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=4aa3564>
11:54 berick dbwells: I wonder what happens when you call the translator w/ .atomic
11:54 dbwells One more possible puzzle clue.  If I add 'atomic' in srfsh, it just hangs forever.  At that point, I think I need to restart the pcrud backend to get it working again.
11:54 dbwells :)
11:54 berick heh
11:55 * berick now wants to know how websockets handles the request
11:55 dbs lualaba_: by "MARC is good" do you mean you have the fixed fields set to designate the thing as an electronic book? The format icons don't care at all about 856 fields.
11:56 dbwells Usually I can find an error in one of the many logs, but keep coming up empty so far.
11:56 lualaba_ yes i am just checking and modifing LDRs. thank you
11:57 dbwells If I follow the log, I see a bunch of object requests / perm checks, then they just stop.  I should go back and check if they always stop at the same place.
11:59 JBoyer joined #evergreen
11:59 * csharp has never uttered the phrase "MARC is good"
11:59 Dyrcona MARC is MARC.
11:59 csharp trudat
11:59 * Dyrcona goes to get some lunch.
12:00 * Dyrcona also thinks he's finally sorted his restoring Evergreen to a "new" server worked out.
12:00 Dyrcona Err, something like that.
12:00 JBoyer MARC is the MARC-est standard still in use at libraries.
12:00 berick dbwells: dmesg show any segfaults?
12:03 dbwells berick: I do see the srfsh segfaults, but I don't really know what I am looking at.  I'll paste one somewhere.
12:03 dbwells most relevant info looks like "error 6 in libopensrf.so.2.1.1"
12:05 pastebot Someone at 64.57.241.14 pasted "srfsh segfault on large pcrud request" (1 line) at http://paste.evergreen-ils.org/37
12:06 berick dbwells: ah, well, you already said srfsh was segfaulting, so I'm not too concerned w/ that.  was mostly curious if the apache child proc running the translator was segfaulting
12:07 berick though, if the translator is segfaulting in the same place and it's not getting logged for some reason, then you could use srfsh to diagnose
12:07 berick since it's easier to attach gdb to srfsh than apache
12:08 berick (that is, if you want to get into all that...)
12:08 dbwells berick: nope, sorry, no other segfaults in dmesg
12:08 pinesol_green [evergreen|Bill Erickson] LP#1507845 SEO home.tt2 template repair - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=1fe4981>
12:10 dbwells berick: I've used gdb like one time since college days, but I'll do whatever I gotta do
12:10 dbwells If you think it will help.
12:11 berick dbwells: in one terminal, run srfsh.  in another, execute:  sudo gdb /openils/bin/srfsh <running-srfsh-pid>
12:11 berick after gdb loads, type 'continue'
12:11 berick then back in srfsh, execute your command
12:11 mmorgan joined #evergreen
12:11 berick then back in gdb, after it dies, type 'bt' (for backtrace)
12:11 berick after it shows the segfault
12:12 berick with any luck, it will give you a line number
12:18 dbwells berick: thanks!  The srfsh window actually spit out a new error when doing this, "Buffer reached MAX_SIZE of 10485760"
12:19 dbwells will paste the bt
12:19 berick oh, nice.  was not expecting that ;)
12:19 dbwells :)
12:20 pastebot "dbwells" at 64.57.241.14 pasted "srfsh segfault bt" (16 lines) at http://paste.evergreen-ils.org/38
12:22 kmlussier dbwells: We might have somebody willing to test bug 1422379 on Bug Squashing Day, but, based on your last comment on the bug, I wasn't sure if you're planning to do more work on it.
12:22 pinesol_green Launchpad bug 1422379 in Evergreen "Move money.billing timestamps back to moment of fine" [Medium,Triaged] https://launchpad.net/bugs/1422379
12:22 kmlussier Is that something that's still ready for testing?
12:22 berick dbwells: aha.. buffer_expand failed on BUFFER_MAX_SIZE, freeing the gb object, then buffer_free was called with a pointer to a fee'd object
12:23 dbwells kmlussier: It is.  I did mention a plan to make the upgrade script more conservative, but haven't found time yet.  It should work as-is.
12:23 kmlussier dbwells: OK, thanks!
12:25 dbs kmlussier++ berick++
12:26 dbs thanks for fixing what I assume was my bad dbs--
12:27 dbwells berick: I need to run for a bit.  Before I do, does this get us anywhere?
12:28 berick dbwells: this issue is likely unique to srfsh, becuase of how it collects resopnses into a growing_buffer before displaying to the user
12:28 berick (unless using raw_print)
12:28 berick so, probably not helping to diagnose the translator
12:28 dbwells okay, I was afraid of that.  Oh well, I guess the search will continue later.  berick++
12:29 berick dbwells: for when you get back to this, are you seeing responses from the translator up until the point that it fails?
12:35 dbwells berick: I may not be doing the LWP stuff exactly optimal.  When it works, I end up with 'content' as a single large JSON array, and then each array member has a payload->content.  When it fails, I just have the 500 error.
12:36 * dbwells steps out
12:37 berick dbwells: then that's the problem.  you should get a series of results, not a single JSON array.  likely the data is collecting in the apache server and filling up some buffer.
12:38 berick maybe an outbound network buffer
12:51 gmcharlt dbwells: are you particularly attached to XML as the hard-coded export format for non-imported records?
12:51 gmcharlt if not, I'd suggest MARC
13:00 Callender_ joined #evergreen
13:20 jlitrell joined #evergreen
13:25 dbwells gmcharlt: The choice of XML is in large part a carryover from the initial implementation (where we simply dumped out from the DB to a file).  It has been handy for doing quick edits and reimporting, but I can see how MARC might be better for general consumption.  So no, not particularly attached.
13:25 gmcharlt dbwells: OK, I'm then inclined to make that change, write release notes, and push
13:28 vlewis joined #evergreen
13:28 dbwells gmcharlt: sweet, thanks again, especially for offering to write release notes.  Hopefully I put something reasonably useful in the commit message to possibly help with that.
13:29 gmcharlt yeah, it's pretty straightforward
13:29 dbwells gmcharlt++
13:31 vlewis When I run \set ON_ERROR_STOP on at the top of my psql file I get the error : syntax error at or near "\" Any suggestions? Thanks
13:34 kmlussier tsbere: Looking at Dpearl's comment at https://bugs.launchpad.net/eve​rgreen/+bug/1501781/comments/8, if a new dependency is added, would the script on the VMs install that new dependency?
13:34 pinesol_green Launchpad bug 1501781 in Evergreen 2.9 "Patron name search should be diacritic-insensitive" [Medium,Triaged]
13:37 tsbere kmlussier: So long as it is added to the Makefile.install chunks
13:39 kmlussier tsbere: That's what I thought. Thanks!
13:41 jeffdavis vlewis: that seems odd. How are you running that script?
13:41 stompro__ joined #evergreen
13:42 vlewis jeffdavis: In a query window
13:42 jeffdavis vlewis: in pgAdmin?
13:43 vlewis jeffdavis: Yes
13:45 tsbere pgAdmin is not psql and does not support most (any?) psql specific commands
13:50 jeffdavis vlewis: yeah, "\set" is a command that's specific to the psql command-line client. I think the "ON_ERROR_STOP" option may be psql-specific too, not sure.
13:50 vlewis jeffdavis: I'm using PostgreSQL
13:52 jeffdavis So, PostgreSQL is the database software, pgAdmin is a separate web interface for interacting with it, and psql is a command-line interface for interacting with it.
13:52 jeffdavis The "\set" command is only supported by psql, it doesn't work in pgAdmin (or so it seems - I don't use pgAdmin myself).
13:57 tsbere jeffdavis / vlewis: I suspect the "Auto-Rollback" option under the "Query" menu is effectively what you would want in pgAdmin
13:58 tsbere (or, you know, switch to running psql by going to Plugins->PSQL Console)
14:00 pinesol_green [evergreen|Dan Wells] LP#1548147 Quick Export Feature for Vandelay Queues - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=212cf48>
14:00 pinesol_green [evergreen|Galen Charlton] LP#1548147: set export format to MARC - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=3460776>
14:00 pinesol_green [evergreen|Galen Charlton] LP#1548147: add release notes - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=8a6f4ba>
14:16 pinesol_green [evergreen|Josh Stompro] LP#1545226 - Fix MARC Batch Editor status screen. - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=54114dc>
14:17 kmlussier As of 2.8, I see we can now match authority records when uploading them in Vandelay. Are there other ways people are uploading authority records?
14:18 Dyrcona kmlussier: I'm using a script of my own devising to load authority records from Backstage.
14:18 kmlussier Dyrcona: That's in the MVLC git repo, right?
14:18 Dyrcona kmlussier: yes. http://git.mvlcstaff.org/?p=j​ason/backstage.git;a=summary
14:20 pinesol_green [evergreen|Ben Shum] LP#1544723: i18n - Add webstaff to update_pofiles - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=5f49532>
14:21 kmlussier And then when the authority record is updated, it automatically updates the bib record, right?
14:21 * kmlussier can test it, but thought somebody would know off the top of their head. :)
14:25 Dyrcona kmlussier: I believe the backstage program runs authority bib linker after.
14:27 krvmga joined #evergreen
14:28 Dyrcona I don't know about Vandelay, but there is a db trigger that will do it, IIRC. (I could be wrong.)
14:29 kmlussier Dyrcona: Thanks!
14:30 kmlussier Dyrcona: Do you run your authority to bib linker on a regular basis? To link the newly-entered bibs?
14:30 gmcharlt grabbing 0965
14:31 Dyrcona kmlussier: It runs as part of our Backstage imports, so no.
14:32 Dyrcona It pretty much only runs the bibs that were updated, and then with a days back option if authorities were updated.
14:38 pinesol_green [evergreen|Terran McCanna] LP#1519911: fix typos in some seed A/T tempaltes - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=160310e>
14:38 pinesol_green [evergreen|Galen Charlton] LP#1519911: upgrade script to fix self-checkout receipt A/T template - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=ba689b9>
14:39 jeff a typo in a commit message for a commit that fixes typos! i see what you did there!
14:41 Dyrcona heh.
14:42 jeff > You hit a guy... with glasses... that, heh, that's well played.
14:46 pinesol_green [evergreen|Thomas Berezansky] LP#1531976: fix bringing up Mesage Center and Triggered Events - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=315beb4>
14:59 kitteh__ joined #evergreen
15:04 kitteh___ joined #evergreen
15:04 mmorgan1 joined #evergreen
15:09 pinesol_green [evergreen|Galen Charlton] LP#1516022: new reporting source for hold/copy ratios - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=4c9f565>
15:09 pinesol_green [evergreen|Galen Charlton] LP#1516022: add release note - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=3ac253a>
15:09 pinesol_green [evergreen|Josh Stompro] LP#1516022: new reporting source for hold/copy ratios - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=f2ca2d7>
15:11 dbwells berick: I've been poking at my earlier problem a bit and am still not getting anywhere.  As far as I can tell, I am not getting any data back from the translator at all.  Could still be something I am doing wrong.  Are there any examples of Perl code making translator requests?
15:13 tsbere dbwells: I wrote this: http://git.evergreen-ils.org/?p=working/random.git​;a=shortlog;h=refs/heads/collab/tsbere/remoteauth
15:17 kitteh__ joined #evergreen
15:18 dbwells tsbere: Thanks.  Not seeing anything different, unfortunately.  I've also tried adding a content callback and a read size to the ua request, but neither helped (I see the content of shorter messages processed in hunks, but nothing from this one).
15:18 dbwells It just says "500 Server closed connection without sending any data back".
15:19 berick i'm not aware of any Perl code apart from tsbere's
15:19 * tsbere would have found his code easier to write if he had something else to base it on ;)
15:20 berick :)
15:21 Dyrcona I did a little thing in Java to login through the translator but that's about all I've done.
15:21 Dyrcona I've never shared it.
15:27 dbwells I am going to try to boil my script down to just a test, then hopefully find out if others have this problem.  Login and other typical size messages work fine.
15:28 berick dbwells: hm, are you setting the X-OpenSRF-multipart header to true?
15:28 berick if not, it will not stream back results.  it will collect them all on the server then send the final result.
15:29 dbwells berick: let me try that, I'd been meaning to and forgot.  I saw scary messages in the code that said multipart was deprectaed and going away or something to that effect, but must have misunderstood.  One moment.
15:30 berick it is deprecated and you may have to write some client side multipart parsing code to read it properly
15:30 berick without, though, you may as well use the gateway
15:31 dbwells ah
15:31 dbwells I see
15:31 berick multipart is the main advantage of the translator.  well, that and the ability to do transactional conversations
15:33 dbwells Why was it deprecated?
15:35 berick dbwells: multipart/x-mixed-replace was only ever supported by FF/Mozilla, and they eventually deprecated it as well
15:35 berick still works in our old XUL client, though ;)
15:36 dbwells ok, makes sense
15:37 dbs first the translator was making the gateway deprecated. Then the translator was deprecated. Welcome back gateway!
15:39 Bmagic Is there a dev meeting today?
15:43 pinesol_green [evergreen|Bill Erickson] LP#1516757: ensure that SIP2 returns DOB correctly - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=86c5da0>
15:43 pinesol_green [evergreen|Galen Charlton] LP#1516757: add Perl test case - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=524a3dc>
15:44 Dyrcona Bmagic: No there is not.
15:44 * dbs deletes the event from the calendar
15:44 gmcharlt jeffdavis: about?
15:45 jeffdavis gmcharlt: hi
15:45 gmcharlt jeffdavis: curious, have you been running your patch for bug 1499086 in production?
15:45 pinesol_green Launchpad bug 1499086 in Evergreen 2.9 "Slowness/timeout on loading bookbags in OPAC" [Medium,Triaged] https://launchpad.net/bugs/1499086
15:46 * berick keeps meaning to put together a perl/EG websockets client
15:47 berick tsbere will probably beat me to it
15:48 jeffdavis gmcharlt: yes. There are a couple of bugfixes in the working branch that would need to be merged with the commit I directly linked in the bug report.
15:49 gmcharlt jeffdavis: right, I see three commits that look relevant in the user/jeffdavis/lp1499086-bookbag-slowness brnach
15:49 tspindler joined #evergreen
15:52 jeffdavis yeah, top three commits in that branch
15:53 tsbere berick: At this point? Probably not until we don't need an alternate port for websockets and MVLC is running them. And I have a need for one. So you have a good chance of beating me still :P
15:54 kitteh_ joined #evergreen
15:55 berick tsbere: :)
15:56 tspindler berick: is your angular patron editor in for 2.10?
15:57 berick tspindler: yep, it's in there.  FYI, still some bug fixes in progress
15:58 tspindler berick: thanks
15:58 dbwells berick: okay, adding multipart seems to have allowed the data to come through.  Yay!  Thank you.  Now to figure out how to actually make use of it.
15:59 pinesol_green [evergreen|Jason Boyer] LP1492793: Support Application Name Postgres Option - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=f942193>
15:59 pinesol_green [evergreen|Galen Charlton] LP#1492793: teach Clark Kent to set the Pg app name - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=efb194d>
15:59 pinesol_green [evergreen|Galen Charlton] LP#1492793: add release notes - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=b8eb0b6>
15:59 berick dbwells: yay, indeed
16:00 jvwoolf1 joined #evergreen
16:01 dbwells berick: that said, any idea why letting the translator group the messages would die/fail, but the gateway succeeds?
16:08 berick dbwells: yeah.. looks like the gateway writes each response body to the apache output buffer as it arrives, so apache can deliver it, then forget about it, whereas the translator collects each response in an array, then writes the final array to the caller.
16:08 berick translator behaves very similar to srfsh in that regard
16:08 dbwells ah, interesting.  Guess that explains them failing in harmony.
16:08 berick yeah, could be the same issue
16:25 tspindler left #evergreen
16:27 terran joined #evergreen
16:35 * dbwells after writing the world's worst multipart parser, enjoys the fruits of his labours.
16:38 Bmagic So, I thought that checkout limit sets didn't care what circ mods were on the already-checked-out items. We have been telling our libraries that limit sets enforce simply on the quantity already checked out. The code suggests otherwise.
16:40 Bmagic scenario - patron has 10 books checked out. There is a checkout limit of 5 dvd's. The patron tries to checkout a dvd, it's blocked because he has more than 5 things checked out (even though those items are not dvds)
16:41 Dyrcona Bmagic: Then the limits were not set up correctly.
16:41 Bmagic I guess so
16:41 Bmagic This subject has always confused me - which is probably why they are not setup correctly
16:43 Dyrcona Off the top of my head, there are a couple of things that it could be.
16:44 Dyrcona 1) They are not properly linked to circ modifiers via config.circ_limit_set_circ_mod_map.
16:45 Dyrcona or 2) The one for dvds is actually linked to the wrong the wrong matchpoint.
16:45 Dyrcona Or maybe they're in the same group when they shouldn't be.
16:46 Bmagic hmm, I will have to do some experimenting
16:46 Dyrcona But, honestly, I usually have to consult with tsbere before I get it right half the time.
16:46 maryj joined #evergreen
16:46 Bmagic haha
16:47 Dyrcona I'd check first if the limit sets in question are in a group.
16:47 Dyrcona IIRC, that would treat them as the same thing.
16:47 Bmagic so, if the circ policy mentions a circ mod - then we should probably match that on the limit set?
16:48 Dyrcona Probably. I should review the code again before I give a definitive answer.
16:48 Dyrcona It's almost quitting time for me.
16:49 Bmagic thanks for your time!
16:49 jeff also keep in mind that some configurations "tag" circulations (in action.circulation_limit_group_map)
16:49 Bmagic Looking forward to some cut thoat Hanabi!
16:49 Dyrcona heh
16:50 Bmagic jeff: that table has 0 rows for us
16:50 jeff though maybe that's -- yeah, i was just noting that here.
16:50 jeff i don't know if that's because we're not "some configurations" or if it's because the table is deprecated.
16:51 Dyrcona We've got 1,780,248, and I don't think it is deprecated, is it?
16:52 Dyrcona You probably need to configure the tags and link them with circ_matrix_matchpoints.
16:53 Dyrcona Those are meant for future reference, if you decide you do want to put limits on after the fact.
16:56 Dyrcona Bmagic: Maybe this will help? https://jason.mvlcstaff.org/lo​oking-glass/notes.html#_limits
16:56 Dyrcona I can't find part 1 anywhere.
16:56 Dyrcona Meh, probably not that helpful.
16:57 Dyrcona Just presentation slides from 4 years ago.
16:57 mmorgan joined #evergreen
16:58 Dyrcona That doesn't explain how they work.....
16:58 Dyrcona Anyway, good night, #evergreen.
17:00 kmlussier StomproJ: Are you still looking at bug 1496522?
17:00 pinesol_green Launchpad bug 1496522 in Evergreen "Wishlist - Action Trigger for periodic billing statement" [Wishlist,Triaged] https://launchpad.net/bugs/1496522 - Assigned to Josh Stompro (u-launchpad-stompro-org)
17:02 Bmagic StomproJ: I was wondering the same thing
17:06 Bmagic So limit set's depth is confusing to me. If you set a depth, and the limit set is mapped to a circ policy. The circ policy also has a "Depth" inferred from the org_unit/patron home ou
17:08 gmcharlt claiming 0966
17:11 jvwoolf1 left #evergreen
17:13 pinesol_green [evergreen|Mike Rylander] LP#1470957: Normalize date1 (pubdate sorter) to null when not numeric - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=d9348b9>
17:13 pinesol_green [evergreen|Galen Charlton] LP#1470957: update upgrade script - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=90dc59a>
17:13 pinesol_green [evergreen|Galen Charlton] LP#1470957: provide pgTAP test for expected pubdate normalization - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=d9f6868>
17:13 pinesol_green [evergreen|Galen Charlton] LP#1470957: adjust a previous regression test - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=4e6dd04>
17:13 pinesol_green [evergreen|Galen Charlton] LP#1470957: stamp schema upgrade - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=8460d7a>
17:19 pinesol_green [evergreen|Terran McCanna] LP #1466173 Selfcheck Hide UI Elements Until Patron Auth - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=f49b010>
17:21 mmorgan left #evergreen
17:37 pinesol_green [evergreen|Mike Rylander] LP#1375043: Support for in-A/T telephony configuration - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=76ace6a>
17:37 pinesol_green [evergreen|Galen Charlton] LP#1375043: add release notes - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=fc9027a>
17:43 pinesol_green [evergreen|Josh Stompro] LP#1391282: Add Auto-Submitted: auto-generated header to included A/T email templates. - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=022f895>
17:50 berick http://thenewstack.io/google-p​reps-angular-2-final-release/ -- Google will support angular 1 "for at least another year"
17:53 csharp ugh
17:57 berick well, it's kind of nice to have clear motivation to keep our stuff up to date
17:57 maryj joined #evergreen
18:02 csharp yeah
18:07 pinesol_green [evergreen|Jason Boyer] LP 1371647: Additional Fixed Fields and Values - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=9a6cc7e>
18:07 gmcharlt grabbing 0967
18:07 pinesol_green [evergreen|Galen Charlton] LP#1371647: tweak pinned IDs - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=76c80e1>
18:09 pinesol_green [evergreen|Galen Charlton] LP#1371647: stamp DB upgrade - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=fbe13ad>
18:23 mrpeters left #evergreen
18:59 jihpringle joined #evergreen
19:07 dbs berick: csharp: wheeeeeee
19:44 mrpeters joined #evergreen
20:27 gmcharlt grabbing 0968
20:30 pinesol_green [evergreen|Josh Stompro] LP#1392396 - New patron welcome email action/trigger event def - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=bd88e5f>
20:30 pinesol_green [evergreen|Josh Stompro] LP#1392396 - Renamed hook from "create" to "au.created" to make it - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=d814675>
20:30 pinesol_green [evergreen|Josh Stompro] LP#1392396 - Changed the hook description to match the other - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=187713a>
20:30 pinesol_green [evergreen|Josh Stompro] LP#1392396 - Release notes for New Patron Notice A/T. - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=04f5e29>
20:30 pinesol_green [evergreen|Galen Charlton] LP#1392396: stamp DB update - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=b3ebb9c>
20:38 mrpeters left #evergreen
20:48 kmlussier gmcharlt++ #Working through dusty pullrequests that have been lingering in LP
20:48 * gmcharlt sneezes
20:50 bshum gmcharlt++
21:00 gmcharlt note to self: tests that depend on patron opt-in are more likely to pass if you actually turn on opt-in :)
21:02 kmlussier gmcharlt: I see you're following the kmlussier method of testing.
21:10 * gmcharlt claims 0969
21:14 pinesol_green [evergreen|Jeff Davis] LP#1533329: Selectively disallow opt-in based on patron's home library - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=eb7f942>
21:14 pinesol_green [evergreen|Jeff Davis] LP#1533329: live tests for opt-in feature - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=4b0e086>
21:14 pinesol_green [evergreen|Jeff Davis] LP#1533329: release notes for restricted opt-in - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=bc9e73f>
21:14 pinesol_green [evergreen|Galen Charlton] LP#1533329: bail out of live Perl test if opt-in not enabled - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=2180dac>
21:14 pinesol_green [evergreen|Galen Charlton] LP#1533329: stamp DB update - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=634d20e>
21:18 pastebot "gmcharlt" at 64.57.241.14 pasted "A more pressing election?" (5 lines) at http://paste.evergreen-ils.org/39
21:19 kmlussier heh
21:20 kmlussier Maybe we can change that at the same time we're changing all the instances of copies to items
21:20 gmcharlt we may need to hold ceremonial coinflips at the Evergreen conference
21:21 b_bonner_ joined #evergreen
21:24 jcamins_ joined #evergreen
21:25 mtj_ joined #evergreen
21:28 * kmlussier notes that we seem to be getting some failures on tests. http://testing.evergreen-ils.org/~live/test.html
21:30 pinesol_green [evergreen|Terran McCanna] LP #1422932 TPAC: Holds history pager typo - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=e46ee22>
21:30 kmlussier Looks like they started failing around 2-24
21:38 bshum Right
21:38 bshum There's a bug with the source and potential solution
21:39 bshum Broke because of https://bugs.launchpad.net/evergreen/+bug/1505286
21:39 pinesol_green Launchpad bug 1505286 in Evergreen "set limit on facets retrieved during search" [Wishlist,Fix committed]
21:39 bshum Which we have to resolve the question of fixing the SQL or making PG 9.2+ the required base version (and updating necessary docs)
21:41 * gmcharlt makes the decision
21:41 gmcharlt 9.1+ it will remain, although I think now is time to announce a deprecation to take effect with 2.11
21:42 bshum Makes good sense.
21:43 kmlussier Ah, ok. I saw talk of that, but didn't realize it was related.
21:43 * kmlussier misses the days when pinesol_green published qa reports in channel.
21:43 bshum kmlussier: It's supposed to, but I think it's buggy.
21:44 bshum Or perhaps the RSS for live tests doesn't work the way that some other RSS does
21:44 kmlussier Yeah, I've seen a couple of attempts to get it working again.
21:44 kmlussier Anyway, it's late. Have a nice night everyone!
21:48 bshum @rss add qatests http://testing.evergreen-ils.org/~live/test.html
21:48 pinesol_green bshum: The operation succeeded.
21:49 bshum @rss announce qatests
21:49 pinesol_green bshum: Error: 'qatests' is not a valid integer.
21:49 bshum @list rss announce
21:49 pinesol_green bshum: (list [--private] [--unloaded] [<plugin>]) -- Lists the commands available in the given plugin. If no plugin is given, lists the public plugins available. If --private is given, lists the private plugins. If --unloaded is given, it will list available plugins that are not loaded.
21:49 bshum @rss announce
21:49 pinesol_green bshum: Error: Couldn't get RSS feed.
21:49 bshum Hmm
21:49 bshum @rss announce add qatests
21:49 pinesol_green bshum: The operation succeeded.
21:51 pinesol_green News from qatests: Test Failure - http://testing.evergreen-ils.org/~live/test.html <http://testing.evergreen-ils.org/~live/test.html>
21:52 bshum :)
21:52 bshum Let's see how that goes for awhile.
21:58 pinesol_green [evergreen|Galen Charlton] clear slate of release notes for master - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=64b518b>
22:21 remingtron_ joined #evergreen
22:22 abowling1 joined #evergreen
22:25 jeff___ joined #evergreen
22:27 mtj_ joined #evergreen

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat