Evergreen ILS Website

IRC log for #evergreen, 2018-09-13

| 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
07:06 rjackson_isl joined #evergreen
07:31 bdljohn joined #evergreen
07:53 jvwoolf joined #evergreen
08:00 Dyrcona joined #evergreen
08:04 jvwoolf1 joined #evergreen
08:29 aabbee joined #evergreen
08:29 * csharp wonders if we can disable the "Answers" feature of launchpad and add a redirect to our site's FAQ
08:34 rlefaive joined #evergreen
08:40 JBoyer I think that's come up before and I thought the answer was "can't do that" :/
08:40 csharp looks like that is possible for someone with higher privileges that I have: https://help.launchpad.net/Answers
08:40 mmorgan joined #evergreen
09:28 gmcharlt csharp: those instructions seem slightly out of date
09:29 gmcharlt but...
09:30 gmcharlt bshum: berick: could you add cesardv and csharp to the "Evergreen Release Team" team on Launchpad
09:30 gmcharlt and bump up me (and anybody else on it whose interested to admin privs on that team?)
09:31 yboston joined #evergreen
09:32 gmcharlt also, I don't see an option to turn off Answers
09:32 gmcharlt although there's a way to specify that it is external
09:32 gmcharlt but then, LP displays something along the lines of "we don't know how this project receives support requests"
09:32 gmcharlt i.e., not seeing a way to put in a URL or text saying where to go to get help :(
09:40 jvwoolf joined #evergreen
09:42 csharp yeah :-/
09:43 csharp gmcharlt: thanks for looking and for trying to up my privs :-)
09:45 bshum gmcharlt: csharp and cesardv added to that team in LP
09:48 bshum And I don't think we can disable answers
09:48 aabbee noting in case anyone else sees similar error: after installing 3.2: i got 500 errors in the browser and /var/log/syslog said 'egweb: Context Loader error: Can't locate object method "retrieve_config_global_flag" via package "OpenILS::Utils::CStoreEditor" at /usr/local/share/perl/5.22.1​/OpenILS/WWW/EGCatLoader.pm line 284.\n'
09:48 aabbee resolved by restarting apache2. found these logs, which helped: http://irc.evergreen-ils.org/​evergreen/2016-09-02#i_265134
09:48 aabbee tsbere++
09:48 aabbee Dyrcona++
09:56 * gmcharlt claims 1130 in the name of folks who don't want to be squished everywhere!
10:03 terran joined #evergreen
10:04 bshum "I shall call him Squishy and he shall be mine and he shall be my Squishy."
10:05 * gmcharlt also claims 1131 in the name of enforcing consistency!
10:06 pinesol [evergreen|Jason Stephenson] LP 1786534: Don't merge a user with itself. - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=ecae2cc>
10:06 pinesol [evergreen|Galen Charlton] LP#1786534: make update script reflect other recent changes in actor.usr_merge - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=7c58067>
10:06 pinesol [evergreen|Galen Charlton] LP#1786534: stamp schema update - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=6129d52>
10:07 * csharp is looking at bug 1773434
10:07 pinesol Launchpad bug 1773434 in Evergreen "Item Status - missing option to "Show in Catalogue"" [Undecided,Confirmed] https://launchpad.net/bugs/1773434
10:08 jwoodard joined #evergreen
10:08 csharp looks like we have working code there with a semi-signoff from kmlussier, but it ends with a question about menus
10:09 csharp wondering if we shouldn't accept the fix as-is and let kmlussier open a new bug if she thinks it makes sense to change it - thoughts?
10:09 * csharp was about to test and sign off
10:10 pinesol [evergreen|Galen Charlton] LP#1786534: ensure that changes don't regress - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=4894212>
10:11 gmcharlt csharp: yeah, I think it makes sense to either leave that for a follow-up bug or just do a quick follow-up patch
10:12 gmcharlt either way, so long as it doesn't block /something/ getting in for 3.2
10:15 csharp ok - that's sane - moving forward with that something :-)
10:30 pinesol [evergreen|a. bellenir] LP#1773434 missing option to "Show in Catalog" - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=b19a728>
10:30 pinesol [evergreen|a. bellenir] LP#1773434: show in catalog - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=fab9098>
10:31 Dyrcona Kernel update, going to reboot. BBIAB.
10:35 Dyrcona joined #evergreen
10:36 pinesol [evergreen|Kathy Lussier] LP#1738688: Add cancel time to Most Recent Transits - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=8a1a43b>
10:44 pinesol [evergreen|Remington Steed] LP#1774724: Fix copy-pasto's in Library Settings - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=564ba9f>
10:45 berick @band add Copy Pasta
10:45 pinesol berick: Band 'Copy Pasta' added to list
10:48 JBoyer With openers The Electric Italian Heinz Crimes
10:48 rebecca_ joined #evergreen
10:49 khuckins joined #evergreen
10:53 yboston Hello everyone, I am planning relying on batch adding standing penalties to a bunch of users to support a new initiative here
10:54 yboston I have figured out how to batch add with SQL the standing penalty that I want to the users I want to target. My question for you folks, is how do I go about using SQL batch deleting standing penalties if the need arises.
10:54 yboston It might be as simple as deleting specific rows in actor.usr_standing_penalty, but I wanted to make sure.
10:54 yboston I looked over the source code and got as far as seeing entried like this $e->delete_actor_user_standing_penalty($penalty) but I have not been able to find the next logical steps in the code to see how standing penalty deletions are done.
10:54 yboston TIA
10:55 Dyrcona yboston: That $e->delete_actor_usr_standing_penalty bit just does a SQL delete under the covers.
10:55 Dyrcona Any of the CStoreEditor functions end up doing SQL, evenually.
10:56 berick and the $e-> methods are auto-generated, so grep won't find them.
10:56 yboston Dyrcona: thanks, that is what I suspected. I just never made the leap to understand how the code interacts wtht he CStoreEditor
10:56 Dyrcona Right. They're autogenerated from the IDL.
10:56 yboston Dyrcona: that is what I thought
10:57 Dyrcona So, you can just do a SQL delete if that's easier than figuring out the Perl, which it usually is.
10:57 Dyrcona berick++ :)
10:58 yboston Dyrcona: for this case I will use SQL, but this is very helpful, it confirmed what I thought was happenng.
10:58 yboston Dyrcona: can you tell me a little more of what I should look for in the IDL to at least go a couple more steps deeper in the code logic
10:59 yboston Dyrcona: I want to learn a bit more of the CStoreEditor logic today from you (or someone else) if you have a second
11:00 Dyrcona Well, the IDL is mainly for Fieldmapper to turn database data into Perl objects, but pcrud, cstorage, and storage use it generate methods to create, retrieve, update, and delete objects.
11:01 mmorgan left #evergreen
11:02 Dyrcona The CStoreEditor generates search_ delete_ create_ and update_ functions for the objects from the IDL. It uses the oils_obj:fieldmapper attribute with the :: converted to an underscore.
11:03 Dyrcona It also requires a transaction for delete, create, and update.
11:03 Dyrcona If you have any specific questions, please ask.
11:05 csharp jeffdavis: FYI, PINES staff are about to test bug 1715767 - hopefully we'll see a signoff so it can make it to 3.2
11:05 pinesol Launchpad bug 1715767 in Evergreen "Allow others to use my account (privacy waiver)" [Wishlist,New] https://launchpad.net/bugs/1715767
11:06 csharp (unless it's too late for getting that done)
11:07 berick csharp: too late on that one for 3.2, but a sign-off now so we can merge as soon as rel_3_2 is branched would be great.
11:07 mmorgan joined #evergreen
11:07 csharp berick: understood
11:21 yboston Dyrcona: thanks a lot for the recap. I knew the basics of what the IDL and fieldmapper did, i.e. creating objects verisosn of DB tables/fields and assist in creating correspnding methods to act upon them, from reading some dev docs. I just had not followed the process before of the other end of callign comands on the CStoreEditor object. ¡¡¡Thanks again!!!
11:23 Christineb joined #evergreen
11:29 gmcharlt csharp: for that last thing you pushed, the schema update needs to be stamped
11:29 gmcharlt do you know how to do that?
11:30 csharp oh oops
11:31 csharp I've not done it before
11:31 gmcharlt ok - so for the room, the process is to
11:33 gmcharlt 1. look at 002.schema.config.sql for the line INSERT INTO config.upgrade_log
11:34 gmcharlt 2. claim the next number in sequence in this channel
11:34 gmcharlt 3. edit 002.schema.config.sql; note that the comment on that line by conventino should be changed to the IRC nicks of the people folks involved in the patch (author, tester, committer)
11:35 gmcharlt 4. under upgrade/, do a git mv XXXX./whatever/ to /new_stamp/./whatever/
11:36 gmcharlt 5. edit /new_stamp/./whatever and ensure that the SELECT evergreen.upgrade_deps_block_check line is uncommented and has the correct new stamp
11:36 gmcharlt 6. commit and push the result
11:37 berick gmcharlt++
11:37 csharp 7. profit
11:37 csharp thanks I'll do that
11:38 gmcharlt NOTE
11:38 gmcharlt when backporting, while you can cherry-pick the update
11:38 gmcharlt note that you can often expect a merge conflict in 002.schema.config.sql
11:38 gmcharlt but of much more import: the schema update itself needs to be looked at for correctness
11:38 gmcharlt particularly for changes to functions
11:38 gmcharlt since the master version of function foo() may be different from the version in previous rel_* branches
11:40 * Dyrcona thinks it is a bad idea to backport db changes, but I do it.... ;)
11:40 gmcharlt yeah, it can very much introduce risks
11:41 gmcharlt csharp: and that's why commits 4894212ed19a4ab6d90b9afd8505d71043fc6e93, 7c580671be47cb2ea800b3e005f6a281acee4829, and aac44b2b02f791351d7505221d8929a6ee2362eb are what they are
11:41 pinesol gmcharlt: [evergreen|Galen Charlton] LP#1786534: ensure that changes don't regress - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=4894212>
11:41 pinesol gmcharlt: [evergreen|Galen Charlton] LP#1786534: make update script reflect other recent changes in actor.usr_merge - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=7c58067>
11:41 pinesol gmcharlt: [evergreen|Galen Charlton] LP#1786534: sync schema update with rel_3_x actor.usr_merge - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=aac44b2>
11:41 gmcharlt as part of my pushing the fix for bug 1786534
11:41 pinesol Launchpad bug 1786534 in Evergreen 3.1 "Possible to merge a user with itself" [Medium,Fix committed] https://launchpad.net/bugs/1786534
11:42 Dyrcona Yeah, I should have rebased that one after something else went in.
11:42 csharp gmcharlt++ # thanks for the thorough explanation!
11:42 gmcharlt cesardv: also, please take note of ^^^
11:44 csharp ok, then I'm calling 1132 :-)
11:48 csharp oops - didn't sign off on it, but I guess that's probably a minor mistake
11:50 beanjammin joined #evergreen
11:51 pinesol [evergreen|Chris Sharp] LP#1774724 - stamp upgrade script - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=f594e28>
11:52 gmcharlt csharp++
12:15 jvwoolf joined #evergreen
12:25 plux joined #evergreen
12:38 plux left #evergreen
12:38 plux joined #evergreen
12:39 JBoyer git --amend --notary
12:44 plux working on an upgrade from 3.0.3 to 3.2 beta on a test box - hit issues with the reingest at 1103.data.virtual_index_defs.sql - tried running with just one valid biblio.record_entry id and it just hangs - ditto for other records
12:45 plux anyone else seeing issues with that?
12:46 beanjammin joined #evergreen
12:47 JBoyer plux, is postgresql completely crashing on you? depending on your database you might be running into bug 1764542
12:47 pinesol Launchpad bug 1764542 in Evergreen "Incorrect format on config.metabib_field insert results in segmentation fault" [High,Confirmed] https://launchpad.net/bugs/1764542
12:54 plux that’s the behaviour ….so I set config.metabib_field format = to mods33 for everything that had an xpath with mods but let 1103 reset the xpath value for id 16 (geographicCode)
12:55 plux to all the xpath in config.metabib_field shows mods 33 and all the format columns for those are mods 33 with the exception of that one xpath field set by 1103
12:55 plux but it does indeed crash the db to run the 1103 reingest even with just one record
13:00 plux reverted that one xpath and it appears to have run OK on that one record so I’ll try to run it again on all
13:01 plux not sure if I can accomplish that with pingest.pl anymore or if it has to run sequentially
13:03 Dyrcona plux: Browse ingest is the only that has to run serially. Everything else can run in parallel.
13:03 jvwoolf1 joined #evergreen
13:03 plux @drycona Thanks!
13:03 pinesol plux: You probably want hard-boiled eggs.
13:04 jihpringle joined #evergreen
13:04 Dyrcona pinesol: I do! Bring me some!
13:04 pinesol Dyrcona: Fire BAD! Reading GOOD!
13:04 JBoyer Sorry to be distracted, what you can do before applying that patch is to just change the default value for the format field to mods33, you don't need to change any other existing values. Then you can apply that upgrade script and reingest without issue.
13:04 * JBoyer notices that his prod db STILL doesn't have that set...
13:15 csharp hmm - that's interesting :-/ :https://pastebin.com/nrrRPgrX
13:16 yboston joined #evergreen
13:16 berick would have been more interesting had it returned a result ;)
13:32 csharp trying to ferret out an issue where I register a workstation, log in with it, then am brought back to workstation (though it sees the one I registered) and when I click "Use Now" it goes back to the login page with "egAuth found no valid authtoken" in an endless loop
13:33 csharp this is on FF and Chrome and cache/cookie clears have all happened
13:34 csharp I've applied several patches at the behest of my colleagues for testing bugs, so I don't know what broke it :-/
13:36 Dyrcona charp: Have you tried doing the npm, make, and make install steps again?
13:36 Dyrcona csharp, that is.
13:36 csharp no, I haven't
13:36 csharp we installed with a deb created from the 3.2-beta1 tarball
13:36 csharp I'm going to revert to a known-working state
13:38 Dyrcona I find reinstalling sometimes helps weird issues when mixing branches.
13:39 csharp yeah - definitely gonna be more careful installing a bunch of branches at once :-/
13:39 JBoyer csharp, re those logs: I regularly see things like that with metarecord url params in it. It grabs the MR id, but keeps going (a ';' vs '&' things somewhere?) I haven't been able to track it down.
13:39 csharp I've seen things like that before in our prod logs too
13:56 rlefaive joined #evergreen
14:03 miker csharp: just a thought: https://bugs.launchpad.net/evergreen/+bug/1786987
14:03 pinesol Launchpad bug 1786987 in Evergreen "Autogen does not refresh unlocalized org tree cache" [Undecided,New]
14:04 miker but ;-separator is probably a better thread to pull first
14:05 csharp miker: it will probably recur and I'll bookmark that bug just in case - thanks!
14:05 * miker looks around for tuits to address that...
14:06 * csharp reverted to "pure" PINES-ified 3.2-beta1
14:06 miker oh, I already did..
14:06 csharp heh
14:10 gmcharlt berick and anybody else who reviewed the recent hold sorting patch - you might be interested in looking at the pull request for bug 1792429
14:10 pinesol Launchpad bug 1792429 in Evergreen "relative queue positions reversed" [High,New] https://launchpad.net/bugs/1792429
14:41 berick gmcharlt: thanks for the heads up
14:53 khuckins joined #evergreen
15:06 mmorgan1 joined #evergreen
15:49 jvwoolf1 left #evergreen
16:05 rlefaive joined #evergreen
17:01 mmorgan joined #evergreen
17:01 bdljohn joined #evergreen
17:09 mmorgan left #evergreen
17:16 jvwoolf joined #evergreen
17:54 pinesol [evergreen|Galen Charlton] LP#1792429: calculate queue position in correct order - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=65f5afc>
18:26 khuckins_ joined #evergreen
21:02 akilsdonk joined #evergreen
21:02 dkyle1 joined #evergreen
21:02 phasefx_ joined #evergreen
21:05 dbs joined #evergreen
21:08 jeff between hurricanes and gas explosions, I hope all of you on the east coast are staying safe.
23:02 csharp_ joined #evergreen
23:02 miker joined #evergreen
23:03 gmcharlt joined #evergreen
23:03 cesardv joined #evergreen
23:05 pinesol joined #evergreen
23:16 agoben joined #evergreen

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