Evergreen ILS Website

IRC log for #evergreen, 2018-09-11

| 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
01:20 remingtron joined #evergreen
02:13 gsams__ joined #evergreen
07:52 rlefaive joined #evergreen
07:55 collum joined #evergreen
07:58 agoben joined #evergreen
08:01 rlefaive_ joined #evergreen
08:16 sandbergja joined #evergreen
08:17 _bott_ joined #evergreen
08:38 bos20k joined #evergreen
08:41 mmorgan joined #evergreen
09:01 jvwoolf joined #evergreen
09:06 jvwoolf1 joined #evergreen
09:07 agoben joined #evergreen
09:23 rlefaive joined #evergreen
09:26 yboston joined #evergreen
09:42 stephengwills note for self - after setting up circulation matchpoint weights and hold matchpoint weights, weights one can assign them to org units via the server->admin->weight association selection.
09:50 bdljohn joined #evergreen
10:05 Dyrcona joined #evergreen
10:23 rlefaive joined #evergreen
10:38 terran joined #evergreen
10:46 dteston joined #evergreen
10:47 csharp ok, seeing an issue moving between angJS and ang6 interfaces - on Chrome on Windows, we log into the /eg/staff UI as normal, but when we click on Acquisitions Administration, it prompts us to register a new workstation
10:47 csharp (3.2)
10:48 Dyrcona csharp: 1) Have you cleared authtokens/cookies beforehand? 2) Does it do that if you login with the eg2/staff first?
10:49 csharp it's not happening to me on Linux (Chrome or FF)
10:49 csharp yeah, we cleared cookies, then tried again
10:49 stephengwills left #evergreen
10:49 Dyrcona OK. I asked not because I have answers, but I didn't see that behavior when I tested it recently. I also used FF and Chromium on Linux.
10:50 Dyrcona I did log in with eg2/staff, first.
10:50 jeff csharp: What version of Chrome on each? By any chance, has one recently upgraded to Chrome 69 and the other is still on an earlier version?
10:50 csharp https://pastebin.com/xaHugXr2 - dev tools console output
10:51 csharp Chrome 68 on Windwos
10:52 berick win 10?
10:52 csharp just updated as I checked the version - let me try again
10:52 csharp yes, Windows 10
10:52 * berick fires up a win10 vm
10:52 csharp trying again
10:54 csharp ok - that worked
10:54 csharp we upgraded Chrome and cleared cookies
10:56 khuckins joined #evergreen
11:01 berick csharp: fwiw, unable to reproduce on win 10 chrome 68
11:02 * berick notices chrome is updating itself now
11:02 berick and ditto chrome 69
11:04 berick i didn't have any cookies to clear, though
11:05 csharp ok - thanks - probably a cookies issue - things look ok now - thanks for checking after us!
11:14 Christineb joined #evergreen
11:41 yboston joined #evergreen
11:59 yboston joined #evergreen
12:04 collum_ joined #evergreen
12:06 jihpringle joined #evergreen
12:15 bdljohn1 joined #evergreen
12:19 pinesol [evergreen|Jane Sandberg] Docs: Adding info about supercat filters - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=f58e975>
12:25 sandbergja joined #evergreen
12:37 rlefaive_ joined #evergreen
12:54 dbwells phasefx: Maybe just me, but I think this is /pretty/ clear evidence you are going to master time travel and have returned under an assumed identity: https://en.wikipedia.org/wiki/Henry_Spencer
12:54 dbwells ;)
12:55 phasefx spooky
12:59 jwoodard joined #evergreen
13:15 pinesol [evergreen|Dan Scott] LP#1774886 Distinguish Phys Char Wizard with an edit icon - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=1d38620>
13:21 beanjammin joined #evergreen
13:32 Dyrcona JBoyer: Would you mind sharing the SQL you used to cancel all of your duplicate transits? I am inclined to go with your solution to that problem.
13:33 JBoyer I'll see if I still have it around. It took a few rounds to make sure I wasn't getting rid of things we need. (and I can't exactly test a recreation since it's stopped...)
13:34 Dyrcona I can probably figure something out, but thought that if you did the work already, I could save myself some time. :)
13:41 berick in case it hasn't been said, +1 to the unique constraint being part of the fix for that bug
13:46 terran joined #evergreen
13:48 JBoyer I figured that'd be a definite, I wasn't sure if the index I put in place was loose enough. For instance, I don't look at prev_hop or anything like that, only canceled and recv times. If all "misdirected" transits (that's how they get a prev_hop, right?) are "received" or canceled when next scanned I suppose it's good as-is
13:51 berick JBoyer: prev_hop gets not love yet...  bug 1789300
13:51 pinesol Launchpad bug 1789300 in Evergreen "Wishlist: Stamp copy transits with a prev_hop value when routed to wrong location" [Wishlist,New] https://launchpad.net/bugs/1789300
13:51 berick but yeah, shouldn't matter
13:52 JBoyer ++
13:52 JBoyer (I vote the original transit gets canceled when that field finally sees use)
13:54 Dyrcona JBoyer: Anything that you can share on bug 1787274 would be greatly appreciated.
13:54 pinesol Launchpad bug 1787274 in Evergreen "Web Client: Transits Don't Always Clear" [Critical,Confirmed] https://launchpad.net/bugs/1787274
13:55 berick JBoyer: as in, create a new transit using the intermediate branch as the source?  if we do that, we wouldn't need the prev hop field.
13:55 berick i could see the argument for that, though
13:56 berick just having a new transit
13:56 JBoyer Well, part of the cleanup I did was to cancel anything where the transit was "open" but the item status was available. I didn't think too much about it except that it seemed obviously wrong.
13:56 JBoyer berick, I guess I thought that was the intent. Cancel the misdirected transit, start a new one "from" the current position and then prev_hop holds the real origin.
13:57 JBoyer (or most recent, if your transit system is bonkers)
13:57 berick oh, no, prev_hop would hold the intermediate
13:57 berick well, most recent intermediate
13:57 JBoyer Ah, I see. From src to dest via prev_hop.
13:57 berick yeah
13:57 JBoyer That makes sense.
13:58 Dyrcona Right, JBoyer. We've had some discussion here about canceling all transits where the copy is in one of a list of statuses that shouldn't have an open transit associated with it.
13:58 JBoyer (I may have a bad habit of making assumptions about incomplete features based on how I might do things at first glance)
14:01 JBoyer Dyrcona, I found the SQL I used (yay 5K+ line .psql_history files with ignore dups) cleaning it up for a paste + look over.
14:02 Dyrcona JBoyer++
14:02 Dyrcona psql_history++
14:05 sfchilds joined #evergreen
14:05 csharp we're seeing an issue in 3.2 where the dojo interfaces fail to load data with an error indicating that the org is undefined
14:06 bshum csharp: What's an example interface you were opening?
14:07 bshum Because mine seem to be opening fine
14:07 berick opening fine here too - back in linux
14:09 pastebot "Jboyer" at 64.57.241.14 pasted "action.transit_copy cleanup" (10 lines) at http://paste.evergreen-ils.org/14090
14:10 khuckins joined #evergreen
14:11 csharp this is the copy locations editor and funds in acq
14:11 yboston joined #evergreen
14:11 csharp maybe something local
14:13 Dyrcona JBoyer: That looks a lot like where my thoughts were heading. You should get all of the duplicates that way.
14:13 bshum Well the copy locations editor worked for me when I opened it on my test VM
14:14 bshum I don't remember how to navigate to funds :)
14:14 pinesol [evergreen|Bill Erickson] LP#1724083 Webstaff approve pending patron address - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=e763bd6>
14:14 bshum The only way I found forward was through the eg2 version of it
14:14 bshum But let's say that I'm rusty in my Evergreen navigation
14:14 bshum The copy editor is old html isn't it?  It doesn't look like dojo to me
14:15 bshum Did anyone muck around with the org units?  Perhaps you've got a stale org unit tree or something wonky?
14:15 JBoyer Dyrcona, if you do that enough times, anyway, heh.
14:15 JBoyer (I had 2 invalid indexes before I finally tracked down all of them)
14:16 Dyrcona Well, it should catch them all if they're from the same day.
14:16 Dyrcona Unless a later one got "sent" and the earlier one didn't.
14:17 berick csharp: bshum: yes, copy locations is pre-Dojo.  Acq funds are Dojo, but accessed via the eg2 splash page.
14:17 berick both loading OK for me
14:18 csharp yeah something must be broken locally - I'll track it
14:19 berick autogen could be it, like bshum said
14:21 berick also curious what the actual JS error is
14:22 pinesol [evergreen|Jane Sandberg] LP1010635: Make bre dates populate in Item Status grid - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=8310bfc>
14:57 Dyrcona Nice. I have 2 open transits from different dates and sources for the same copy. I should have expected that.
15:00 miker terran: we're looking into a suspected bug, but I'm wondering if there was a reason for a particular change... you may not recall since the patch is 4 years old ;) but in 2f252638d4a you added hold_pickup_lib=0 (rather than =1) to the KPACs hold pickup lib org selector.  do you recall if there was a reason to use 0 rather than 1? (it stops an org setting, one that can disable a lib for pickup selection, from working as it does in the main OPAC)
15:00 pinesol miker: [evergreen|Terran McCanna] LP#1282783: Use patron hold notification defaults for KPAC - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=2f25263>
15:03 terran miker: I'm sorry, I don't recall - my guess would be that I was trying to set it to the first item in an array if there wasn't a default setting
15:03 miker ah, ok. it's a bool flag... we'll LP it, thanks
15:05 khuckins_ joined #evergreen
15:09 terran Thanks Mike
15:21 mmorgan joined #evergreen
15:49 bdljohn joined #evergreen
15:56 csharp ugh - autogen.sh was the culprit of my dojo UIs not loading issue
15:57 phasefx anyone interested in setting up dedicated build slaves for the live tester? :)
15:57 csharp at least all of my problems so far are PEBKAC
15:57 phasefx csharp: jeff: ^  I have multi-host support worked out
15:58 csharp phasefx: there should be some VMs for the purpose (probably long out of date) on mundungus
15:59 csharp I created Ubuntu, Debian, and Fedora hosts IIRC, but I didn't have credentials to set up buildslaves
15:59 phasefx csharp: main things are 1) vm's need to be refreshed from snapshot periodically and 2) they need to accept an outside ssh connection (key-based)
15:59 bdljohn1 joined #evergreen
16:00 phasefx (optionally from a known IP if you want to tighten it further)
16:00 bdljohn joined #evergreen
16:02 bdljohn1 joined #evergreen
16:05 phasefx branched from collab/phasefx/wheezy_installer, we have http://git.evergreen-ils.org/?p=working/random.git​;a=tree;h=refs/heads/collab/phasefx/eg_live_tests   There's a public key in there for live@testing.evergreen-ils.org, and a config file in qa/test_runner.xml for managing remote hosts
16:06 bdljohn joined #evergreen
16:10 csharp which OSes need to be testing?  I have Ubuntu 14.04 & 16.04 and Debian jessie hosts running
16:10 csharp I can add others
16:10 bdljohn1 joined #evergreen
16:11 phasefx csharp: I think the more the merrier, but I'll likely be getting a Debian Jessie vm to replace the wheezy one we've been using
16:14 phasefx the script will fork a process for each server being tested
16:14 Dyrcona phasefx: I'd go for stretch so that you get a longer support lifetime.
16:15 phasefx Dyrcona: sounds sane to me
16:16 * phasefx needs to put together a multi-host dashboard/summary page
16:20 phasefx we could also test branches other than master if desired
16:32 Dyrcona JBoyer: I think we need a more thorough approach to remove "duplicate" transits. I added a query and I was able to create the index, but I still have 1,712 copies with duplicate transits since we upgraded to 3.0 using my first query from https://pastebin.com/AqZeeeh2
16:38 mmorgan1 joined #evergreen
16:42 Dyrcona But, I guess all of those transits can't be open....I think the open ones where another has been sent, should probably be canceled.
16:45 mmorgan1 Those duplicate transits sound scary :-(
17:07 mmorgan left #evergreen
17:34 Dyrcona You'd think that a song entitled "Last Leaf" would be the last song on the album, but it isn't.
17:47 csharp 3.0 is done except for security releases, right?
17:47 * csharp notices 3.0.12 target in launchpad
17:48 pinesol [evergreen|Bill Erickson] LP#1785887 edi_order_pusher.pl sets message account - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=6fb3aab>
18:08 sandbergja joined #evergreen
18:11 Christineb joined #evergreen
20:06 jvwoolf1 left #evergreen
22:22 finnx joined #evergreen
22:36 Guest32965 left #evergreen

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