Evergreen ILS Website

IRC log for #evergreen, 2015-09-09

| 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
00:17 jwoodard_tablet joined #evergreen
01:29 Mark__T joined #evergreen
02:43 jwoodard_tablet joined #evergreen
03:03 pgardella joined #evergreen
03:30 jwoodard_tablet joined #evergreen
04:21 jwoodard_tablet joined #evergreen
04:40 jwoodard_tablet joined #evergreen
05:12 jwoodard_tablet joined #evergreen
07:35 jwoodard_tablet joined #evergreen
07:45 rjackson_isl joined #evergreen
07:55 jboyer-isl joined #evergreen
08:02 artunit_away joined #evergreen
08:03 mrpeters joined #evergreen
08:04 ericar joined #evergreen
08:22 _bott_ csharp: for the record, we've been running Pg 9.4.1, with EG 2.8, under Precise for a few months now.
08:29 Shae joined #evergreen
08:33 Stompro csharp: the Debian Jessie EG prereq install does install PG 9.4.4, that is the version debian packages for Jessie.
08:33 mrpeters joined #evergreen
08:39 _bott_ joined #evergreen
08:41 akilsdonk joined #evergreen
08:47 mmorgan joined #evergreen
08:52 csharp _bott_: awesome - I've been looking forward to upgrading for a while
08:52 csharp Stompro: thanks - I figured that out almost immediately :-)
08:53 csharp _bott_: oh yeah, I remember now that dkyle mentioned using 9.4 at GRPL without issues at the EG conf
08:53 * csharp knew someone mentioned it but wasn't remembering who
08:54 _bott_ We poked it with a stick for awhile before upgrading and didn't have any issues, so we ran with it.
08:54 ericar_ joined #evergreen
08:54 _bott_ Using streaming replication and PGPool along with
08:57 Callender joined #evergreen
09:09 maryj joined #evergreen
09:18 graced joined #evergreen
09:26 collum joined #evergreen
09:31 abowling1 joined #evergreen
09:32 sarabee joined #evergreen
09:33 bshum That's better.
09:44 kmlussier phasefx / gmcharlt / eeevil: I noticed that the ESI demo server at http://evergreen-ils.org/dokuwik​i/doku.php?id=community_servers is running 2.6
09:44 kmlussier Do you have any plans to update it or should I take down the link?
09:46 gmcharlt kmlussier: we'll update it
09:46 gmcharlt thanks for mentioning it
09:48 kmlussier gmcharlt: Do you think there is any value to the MassLNC and ESI demo servers running different releases?
09:49 kmlussier Of course, not 2.6, but maybe a 2.8 and 2.9.
09:49 gmcharlt kmlussier: yeah, an N and N-1 convention seems like a good idea
09:49 abowling joined #evergreen
09:49 kmlussier gmcharlt: I can take 2.9 if you want 2.8
09:50 gmcharlt kmlussier: beats trying to figure out how to play rock paper scissors over IRC ;)
09:50 kmlussier Well, there's always this:
09:50 kmlussier @coin
09:50 pinesol_green kmlussier: heads
09:50 gmcharlt kmlussier++ :)
09:54 artunit joined #evergreen
10:02 rlefaive joined #evergreen
10:18 * bshum idly wonders if 2.next will be 2.10
10:24 bshum Hmm
10:24 bshum csharp: I haven't tested the working code bits yet, but I think we might want to try some different ways of explaining the options in the README.
10:25 bshum Looking at https://bugs.launchpad.net/evergreen/+bug/1493824
10:25 pinesol_green Launchpad bug 1493824 in Evergreen "Evergreen/PostgreSQL 9.4 support" (affected: 1, heat: 6) [Undecided,New]
10:25 bshum If we do adopt the postgresql apt repo by default, I wonder too whether it makes more sense to just make PG 9.4 the standard target, and avoid dancing about between multiple make targets for Trusty
10:27 * bshum also wonders when we should probably yank out all the Squeeze targets and references.
10:27 bshum And 16.04 is next year already?  Come on....
10:27 * bshum grumbles to himself
10:34 rlefaive joined #evergreen
10:40 gmcharlt +1 for using the Pg project's apt repos
10:42 gmcharlt though I'm not entirely comfortable with /silently/ adding a repo *and* apt key
10:44 gmcharlt bshum: re Squeeze -- I'm not in favor of yanking it just for the sake of yanking it until such time as Squeeze LTS goes out of support
10:45 bshum gmcharlt: I'm fine with that, better decision to be made by actual Debian users imo :)
10:45 * bshum goes back to his corner
10:47 gmcharlt of course, there may come a time when we want deps that aren't conveniently available in Squeeze LTS, in which case we can resconsider
10:57 Dyrcona joined #evergreen
11:09 jboyer-isl Has anyone seen this: I’ve got a bunch of BRE’s with holdings tags in the marc field, this disappears when the record is edited.
11:09 jboyer-isl 1. Is this likely something like this: cataloger exports records, accidentally clicks “include holdings” box, edits in MarcEdit or whatever, re-imports.
11:10 bshum jboyer-isl: With Evergreen, I would normally expect the MARC record in BRE to be stored without any holding tags in them.
11:10 jboyer-isl 2. is this any kind of potential issue? Aside from being out of date the holdings data doesn’t appear anywhere in the client.
11:10 bshum And there is some sort of code I thought that would remove any holding tag data out of the bib
11:10 bshum Depending on how it was formatted of course
11:11 gmcharlt demo.evergreencatalog.com is now running 2.8.3
11:11 bshum Like our old Horizon holding tag was like 949 and thus custom, not 852
11:11 bshum So many of our bibs still retain a "holding tag" for the items that does not truly reflect reality.
11:11 bshum You know what, actually strike everything I said
11:11 bshum I'm thinking of marc_export
11:12 bshum Which does strip holdings tags out of the data during export to attach new holdings tags based on actual item data.
11:12 jboyer-isl bshum: I don’t mean an 852 or 949, I’m talking about “<holdings xmlns="http://open-ils.org/spec/holdings/v1”><counts><count type=“public”…
11:12 jeff jboyer-isl: under what circumstances are you seeing tags removed, and what were the tags before removal?
11:12 jeff oh, you have non-marcxml xml in biblio.record_entry.marc fields?
11:12 jeff neat!
11:12 gmcharlt jboyer-isl: indeed - MARC::File::XML will (in effect) remove anything that's not in the marc21slim namespace when it parses records
11:13 bshum @quote get 46
11:13 pinesol_green bshum: Quote #46: "<_bott_> I am not a cataloger, but I speak enough MARC to be fun at parties" (added by gmcharlt at 11:43 AM, March 15, 2013)
11:14 jboyer-isl gmcharlt: that’s good, so I assume it doesn’t cause problems, though now that I have a list I’m still tempted to hand it to someone to cleanup.
11:15 jeff jboyer-isl: do you have any idea how it managed to get in there? manual importing/migration?
11:16 jeff fwiw, no sign of any such records in our db.
11:16 jboyer-isl My only thought is what I mentioned above, export with holdings, do whatever, re-import+overlay. I guess that workflow should also strip out the holdings tag, so if that’s the case I’m at a loss.
11:16 gmcharlt one of SuperCat's export modes includes embedding such holdings elements
11:17 jboyer-isl This is something I can test, I guess I should get to it.
11:21 jboyer-isl Sigh. Good job RAID battery, why don’t you condition yourself in the middle of the day, why would this not be the perfect time.
11:23 Christineb joined #evergreen
11:23 gmcharlt jboyer-isl: http://demo.evergreencatalog.com/opac/extras/u​napi?id=tag:demo.evergreencatalog.com,2015:bib​lio-record_entry/3122&amp;format=marcxml-full
11:27 jeff jboyer-isl: writeback cache disabled during battery conditioning?
11:28 jboyer-isl jeff: yes. on the host with the logging collector. Also, the check_raid script we’re using doesn’t understand that this is a warning, not critical. (I can use negate to change that I think, it just doesn’t happen often enough to annoy me into doing it)
11:28 jeff ah. if it was just the nagios check i was going to counter with "would you rather get the alert at 3 AM?" :-)
11:29 jeff but then suspected it wasn't just a spurious alert that was causing you annoyance.
11:29 jboyer-isl jeff, heh, well… my iPhone goes into timed DND between 9:00 and 7:15, so 3AM would actually be perfect. :D
11:30 jboyer-isl Not ideal, IRC logs, I know.
11:36 jboyer-isl gmcharlt: I tried importing xml and binary records with holdings through vandelay on our migration server (2.7.2) and I can’t seem to cause it. Perhaps this is an old problem, since nothing shows up for the records I’m looking at in the audit tables.
11:36 gmcharlt a direct insert into bre could have done it
11:36 bmills joined #evergreen
11:37 jboyer-isl (audit tables which we clear after some number of months, so that’s not definitive)
11:38 jboyer-isl I wonder if there was a project done where something was done via perl and we started with the wrong export options.
11:40 jwoodard joined #evergreen
11:41 * jwoodard joined the wrong channel and just noticed
11:45 kmlussier gmcharlt++ # Updating the ESI demo server
11:46 bshum jwoodard: Which channel did you end up in?  :D
11:46 bshum Maybe it's a better place.
11:46 gmcharlt a bunch of very confused automobile jockeys now wondering how to power their cars using MARC records? ;)
12:23 jwoodard bshum: #evergreen# lol
12:23 jwoodard I have it set to auto login in so I do no understand how I ended up there.
12:23 jwoodard All I know is that it is a cold empty place.
12:27 berick but so much room for activities
12:30 jeff berick: looks like the fix for that ssl issue is going to make it into Chrome 46
12:30 jeff (if you hadn't seen that already)
12:30 berick jeff: oh, great, thanks for the heads up
12:53 bmills joined #evergreen
12:54 csharp gmcharlt: re: silently adding the PG repo & key - what would you rather see?  a warning to the user?  a prompt that needs to be accepted?  or should that be an optional step in the install instructions?
12:54 * csharp is totally open to suggestions ;-)
12:54 gmcharlt csharp: optional step in the install instructions
12:55 gmcharlt adding a prompt would also work for me
13:05 ericar_ joined #evergreen
13:11 * Dyrcona prefers an optional step.
13:12 Dyrcona While Pg 9.4 is great, there's nothing particularly wrong with Pg 9.3.
13:12 * Dyrcona is late to the party.
13:21 jihpringle joined #evergreen
13:22 * csharp starts to question if this is still worth contributing as a feature...
13:23 csharp I mean, we're interested in running 9.4 on Ubuntu trusty, but if no one else cares, I can let it go
13:25 csharp are there other advantages to using postgres's apt repos other than "my packaged version isn't the newest one?"
13:26 bshum That's generally a good advantage in my opinion.
13:27 csharp well, I guess the question is, do we want that as a documented and/or "sanctioned" option for Evergreen in general, or would we just leave that up to individual admins to implement? (i.e., the status quo)
13:27 csharp if the latter, then I'll stop working on it as an Evergreen feature and just move one ;-)
13:27 csharp s/one/on/
13:29 * csharp moves on for now
13:49 mixo joined #evergreen
13:51 mixo hello. does circulator need "VIEW_BILLING_ADDRESS" permission?
13:58 mixo When I try login with staff client I get error "Permition Denied: VIEW_BILLING_TYPE
13:59 mixo another staff member with the above permission may authorize this specific action
14:00 tsbere mixo: Well, view billing type would be "permission to see the list of billing types", I believe, if only to cache them for later use.
14:00 tsbere mixo: The billing address one, however, I am less sure about if you are getting that at login
14:03 mixo yes I am getting error at login. I think, I did  some mistake during update. osrfsys
14:04 mixo osrfsys.log does not says anything about error
14:05 Dyrcona mixo: If you're getting messages that the account needs those permissions, then it needs those permissions.
14:05 mixo When I login with cataloger user all is ok
14:06 Dyrcona mixo: Cataloger is a different profile from circulator with different permissions.
14:07 Dyrcona Cataloger also needs different permissions.
14:07 dMiller_ joined #evergreen
14:08 Dyrcona mixo: It could also be that the profile/group or user has the permissions, but they need to be at a different depth.
14:10 vlewis joined #evergreen
14:10 vlewis_ joined #evergreen
14:15 mixo I gave this permission to circulators group but get same error
14:15 mixo I've run autogen.sh after that
14:17 Dyrcona mixo: The depth of the permission probably needs to be 0.
14:18 Dyrcona You shouldn't have to do anything special after granting a permission.
14:18 mixo ok. I will try
14:20 mixo depth 0 is CONS or SYS1?
14:22 Dyrcona depth 0 is CONS.
14:22 Dyrcona It grants the permission everywhere basically.
14:27 Bmagic Has anyone developed a script to delete unreferenced report files?
14:28 bshum Bmagic: I think we end up just piping results from the orphaned reports script to rm -r
14:28 * bshum looks
14:30 jwoodard_tablet joined #evergreen
14:30 Bmagic oh, that's simple enough
14:31 bshum Bmagic: http://wiki.evergreen-ils.org/doku.php​?id=scratchpad:random_magic_spells#reg​ularly_scheduled_report_output_purging
14:31 bshum That's the approach I occasionally use
14:32 Bmagic bshum++
14:32 Bmagic thanks
14:32 Bmagic I was working on developing something then I realized that this has probably been done already
14:33 kmlussier @love the random magic spells page
14:33 pinesol_green kmlussier: The operation succeeded.  kmlussier loves the random magic spells page.
14:40 gsams @love the random magic spells page
14:40 pinesol_green gsams: The operation succeeded.  gsams loves the random magic spells page.
14:40 gsams I firmly believe it is the best page
14:42 Dyrcona gsams: But not "The Best Page in the Universe" :)
14:43 Stompro Interestings, I just tried to do a pg_restore on a 9.1 system from a 9.4 backup with the custom directory format, and putty locks up going to 100% cpu usage?  I wonder why that happens?  It looks like it isn't possible to do downgrade dumps to older versions with the custom formats of pg_dump.
14:44 miker Stompro: no, it's not
14:44 miker Stompro: you could try taking the pg_dump with the 9.1 binary against the 9.4 server, but that might fail as well
14:46 Stompro Shouldn't a text/sql only dump work?
15:01 miker Stompro: not necessarily. feaures change, etc
15:02 jboyer-isl Stompro: There’s always the “insert” style dump; It’s practically guaranteed to work but is extremely slow.
15:10 vlewis__ joined #evergreen
15:10 vlewis___ joined #evergreen
15:16 rlefaive joined #evergreen
15:30 jlitrell joined #evergreen
15:52 ericar_ joined #evergreen
16:05 Bmagic is it possible that the column default_copy_count could impact which provider is assigned to invoices created via EDI ?
16:10 kmlussier Bmagic: Which vendor are you working with?
16:12 Bmagic BT
16:14 kmlussier Bmagic: My short answer to that question is I hope not.
16:15 kmlussier Bmagic: But I'm sitting across an acq person right now who tells me that the only time she's had trouble with the wrong provider on an invoice is when the proider is BT and there isn't a vendor assigned code in the EDI account settings for that provider.
16:15 kmlussier proider? ugh
16:19 Bmagic hmmm
16:21 Bmagic All of our have a vendcode assigned (in the EDI account)
16:27 kmlussier Bmagic: Are the lineitem ids being truncated in the EDI message?
16:40 ericar joined #evergreen
16:40 mrpeters left #evergreen
16:47 kmlussier I'm updating http://docs.evergreen-ils.org​/2.8/_settings_overview.html to add negative balance settings. I see Long-Overdue Max Return Interval is identified as a True/False setting. Should that be duration?
16:52 Bmagic kmlussier: I had to run out, let me check on your questeion
16:52 vlewis joined #evergreen
16:53 jlitrell joined #evergreen
16:55 Bmagic kmlussier: no, it doesn't appear that the edi messages are getting truncated
16:56 Dyrcona kmlussier: Yes, it should be an interval according to the upgrade script.
16:56 kmlussier Dyrcona: Thanks! I'll update the docs accordingly
16:56 kmlussier Bmagic: OK, in our EDI messages from BT, the NAD+BY part has the library SAN followed by the vendcode. Do you see the same thing in your messages?
16:57 kmlussier Bmagic: I have to get started on my commute home, but I'm out of ideas anyway if that isn't the problem.
16:57 Bmagic There is a column in acq.edi_message called "account" which is selected upon processing the FTP file. One would think that the account would be the same for all of the related EDI messages to a purchase order
16:57 Bmagic kmlussier: yes, we get the SAN but no vendcode (which we addressed with BT with your help
16:59 Bmagic kmlussier: drive safe! Have a good evening
17:01 kmlussier Bmagic: Wait, I'm confused. You said you're getting no vendcode in the message? Then I think that would be the problem.
17:01 * kmlussier leaves for real now.
17:02 dMiller_ joined #evergreen
17:04 Bmagic kmlussier: yeah, we thought that was the problem but it's hard to say that it's the issue, because it's only happening for 3 of the 9 vendcodes related to BT
17:04 Bmagic and it just so happens that those 3, have a different value in default_copy_count than the rest
17:07 Bmagic kmlussier: Other EDI invoices do not contain the vendcode and are assigned the correct provider (which I would image can be inferred by referencing the PO and therefore the vendcode is not required)
17:12 mmorgan left #evergreen
17:25 vlewis_ joined #evergreen
17:29 bmills joined #evergreen
17:31 dMiller_ joined #evergreen
17:49 jwoodard @librarian
17:49 pinesol_green jwoodard: Management:13, Cataloging:14, Acquisitions:12, Reference:10, Circulation:16, Systems:13, Research:12, Custodial:13
17:53 gmcharlt @developer
17:53 pinesol_green gmcharlt: Communication:9, BigPicture:10, DetailOriented:16, KungFu:15, GetsStuffDone:8, FlakeFactor:9, JavaAvoidance:14
17:53 gmcharlt @librarian
17:53 pinesol_green gmcharlt: Management:13, Cataloging:15, Acquisitions:13, Reference:12, Circulation:11, Systems:15, Research:14, Custodial:15
17:56 gsams @librarian
17:56 pinesol_green gsams: Management:6, Cataloging:15, Acquisitions:10, Reference:15, Circulation:16, Systems:13, Research:13, Custodial:15
17:57 gsams ouch, looks like my staff are on their own tonight.
17:57 gsams I have a hard time deciding how a 6 in management would affect us.  Would it mean that I'm getting in the way or simply not there?
17:58 gsams I suppose a score that low would be both simultaneously.
18:03 James joined #evergreen
18:03 James Hi, does anyone out there distribute an ISO or other image of a sample Evergreen server?
18:03 James With fake data, of course.
18:06 phasefx Guest91276: James: I don't think there is anything out there with modern EG, but I could be wrong.  There are quick ways to build a demo system though, if you have, say, a pristine image of Debian Wheezy
18:07 sbrylander joined #evergreen
18:37 dMiller_ joined #evergreen
19:08 jwoodard_tablet joined #evergreen
19:11 dMiller_ joined #evergreen
19:37 jwoodard_tablet joined #evergreen
20:02 kmlussier Bmagic: Well, the only feedback I can give on the invoice problem is that our acq expert who has been doing eg acq for more than 3 years  only sees that particular problem on B&T invoices where the vendcode is missing. When the vendcode is added, she has no trouble.
20:04 kmlussier @developer
20:04 pinesol_green kmlussier: Communication:16, BigPicture:15, DetailOriented:15, KungFu:8, GetsStuffDone:15, FlakeFactor:12, JavaAvoidance:10
20:44 gsams @dnd
20:44 pinesol_green gsams: strength:11 dexterity:15 constitution:8 intelligence:13 wisdom:11 charisma:12
20:52 jwoodard_tablet @dnd
20:52 pinesol_green jwoodard_tablet: strength:7 dexterity:15 constitution:13 intelligence:18 wisdom:15 charisma:15
20:53 jwoodard_tablet Its almost like real life. lol
20:54 jwoodard_tablet Wait...cut about 8 points from charisma.
23:15 dMiller_ joined #evergreen
23:33 jwoodard_tablet joined #evergreen

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