| Time |
Nick |
Message |
| 00:15 |
|
remingtron__ joined #evergreen |
| 02:26 |
|
eby_ joined #evergreen |
| 02:27 |
|
sard_ joined #evergreen |
| 02:30 |
|
Bmagic_ joined #evergreen |
| 05:12 |
|
beanjammin joined #evergreen |
| 07:15 |
|
rjackson_isl joined #evergreen |
| 07:50 |
|
rlefaive joined #evergreen |
| 07:53 |
|
jvwoolf joined #evergreen |
| 08:43 |
|
mmorgan joined #evergreen |
| 09:15 |
|
dbwells joined #evergreen |
| 09:24 |
|
terran joined #evergreen |
| 09:32 |
|
yboston joined #evergreen |
| 09:36 |
|
kmlussier joined #evergreen |
| 09:47 |
kmlussier |
Good morning #evergreen! |
| 09:47 |
dbs |
morning! |
| 09:47 |
* kmlussier |
thinks we need additional caffeine in here to help us recover from the time change. |
| 09:47 |
kmlussier |
@coffee [someone] |
| 09:47 |
* pinesol_green |
brews and pours a cup of Ethiopia Yirgacheffe, and sends it sliding down the bar to eby |
| 09:47 |
kmlussier |
@tea [someone] |
| 09:47 |
* pinesol_green |
brews and pours a pot of San Mateo Air Dried Loose Yerba Mate, and sends it sliding down the bar to jeff (http://ratetea.com/tea/guayaki/san-mateo-air-dried-loose-yerba-mate/4327/) |
| 09:52 |
Bmagic_ |
I'm getting a "returning NULL" error when opening a report template folder with over 1000 templates in it. Should that be a problem for Evergreen? |
| 09:52 |
Bmagic_ |
Returning NULL from app_request_recv after timeout: open-ils.reporter.folder_data.retrieve |
| 09:54 |
|
jvwoolf joined #evergreen |
| 09:54 |
Bmagic |
It will open the template folder no problem with any of the limiters (10,25,50) but not "all" |
| 10:01 |
berick |
Bmagic: maybe dig a little deeper for the source error. could be timeout, jabber message too big, etc. |
| 10:01 |
berick |
there's a turtle at the bottom of the stack |
| 10:02 |
csharp |
Bmagic: we have trouble with "all" too |
| 10:03 |
Bmagic |
berick: which log then? I thought I had them all tailed |
| 10:04 |
Bmagic |
request en-us open-ils.reporter-store.direct.reporter.template.search.atomic [{"folder":1070},{"flesh":1,"order_by":{"rt":"create_time DESC"},"flesh_fields":{"rt":["owner"]}}] |
| 10:04 |
Bmagic |
followed by about 100 lines of |
| 10:04 |
Bmagic |
[open-ils.reporter-store] sent 189 bytes of data to opensrf private.localhost/open-ils.reporter_drone_at_localhost_130291 |
| 10:05 |
Bmagic |
Message processing duration 0.175308 |
| 10:05 |
Bmagic |
editor[0|1] open-ils.reporter-store.direct.reporter.template.search.atomic: returned 250 result(s) |
| 10:05 |
|
mmorgan1 joined #evergreen |
| 10:05 |
Bmagic |
and then the gateway.log spits out the aforementioned error |
| 10:06 |
berick |
Bmagic: total time from start to end of Perl API call? |
| 10:07 |
berick |
which I'm guessing is open-ils.reporter.folder_data.retrieve |
| 10:08 |
Bmagic |
wait a minute.... it looks like one of the services is using a different time |
| 10:08 |
Bmagic |
[2018-03-12 14:07:24] open-ils.reporter vs. [2018-03-12 10:07:25] open-ils.reporter within a split second |
| 10:11 |
berick |
ok so not a timeout. xmpp stanza too big? |
| 10:11 |
berick |
i know that's not supposed to happen anymore, but i seem to recall some edge cases |
| 10:12 |
pastebot |
"Bmagic" at 64.57.241.14 pasted "open-ils.reporter-store 2018-0" (32 lines) at http://paste.evergreen-ils.org/597 |
| 10:15 |
Bmagic |
paste.evergreen-ils.org/597 |
| 10:15 |
Bmagic |
Sorry that's not clickable http://paste.evergreen-ils.org/597 |
| 10:15 |
Bmagic |
CStoreEditor.pm is reporting times in the log with UTC (maybe) and Transport.pm is reporting times in EST. Could that have an impact here? |
| 10:15 |
|
Bmagic joined #evergreen |
| 10:16 |
pinesol_green |
[evergreen|Remington Steed] Docs: Remove docs not applicable in web client - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=3173dcf> |
| 10:16 |
Bmagic |
Did I miss something? I DCed I guess |
| 10:18 |
Bmagic |
http://paste.evergreen-ils.org/597 |
| 10:18 |
Bmagic |
CStoreEditor.pm is reporting times in the log with UTC (maybe) and Transport.pm is reporting times in EST. Could that have an impact here? |
| 10:18 |
berick |
all we can see is the Perl API took 5.936 seconds total. well within the default 60 second timeout. and then the gateway reported a timeout. which still has the appearance like a xmpp message size error. |
| 10:19 |
berick |
the clock differences should not matter, since the services don't compare times. |
| 10:19 |
Bmagic |
ok good to know |
| 10:19 |
Bmagic |
There is a config with ejabberd that needs to change to accomodate 1000 templates in a single folder? |
| 10:20 |
Bmagic |
accomodate / accommodate |
| 10:20 |
berick |
Bmagic: don't know for sure that's the problem. still just a theory |
| 10:20 |
berick |
but maybe |
| 10:21 |
berick |
doesn't jabber log an error when a message is too big? i would expect some noise somewhere |
| 10:21 |
Bmagic |
hmm, I didnt look |
| 10:21 |
|
kmlussier joined #evergreen |
| 10:22 |
Bmagic |
2018-03-12 10:07:45.537 [info] <0.1017.3>@ejabberd_socket:send:181 Timeout on gen_tcp:send |
| 10:22 |
berick |
huh, not familiar with that error |
| 10:23 |
Bmagic |
it's followed up with this |
| 10:23 |
Bmagic |
2018-03-12 10:07:45.537 [info] <0.1017.3>@ejabberd_c2s:terminate:1788 ({socket_state,gen_tcp,#Port<0.10692>,<0.1016.3>}) Close session for opensrfpublic.localhost/translator_nccard-upgrade.nccardinal.org_1520850307.050388_88848 |
| 10:23 |
Bmagic |
if that helps |
| 10:24 |
berick |
well, we've angered ejabberd, that much is clear |
| 10:25 |
berick |
you could always try raising the max stanza size and seeing what happens. simple to test. |
| 10:25 |
Bmagic |
sure |
| 10:27 |
Bmagic |
upped from 2000000 to 5000000 - same error |
| 10:28 |
|
Christineb joined #evergreen |
| 10:28 |
Bmagic |
I guess add a zero? |
| 10:30 |
bshum |
If it's already that high, seems like that's probably not the thing |
| 10:30 |
bshum |
Since historically, that's already a high number |
| 10:31 |
Bmagic |
it doesn't seem to matter |
| 10:31 |
Bmagic |
same error |
| 10:31 |
berick |
ok, yeah, that's already pretty well padded |
| 10:32 |
Bmagic |
I would like the log to report the right time. where is the config for that?it's weird that a single perl module would report times in the logs 4 hours off while the rest of the modules are "correct" |
| 10:32 |
pinesol_green |
[evergreen|Remington Steed] Docs: Fix broken include in root.adoc - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=506d1c2> |
| 10:32 |
berick |
Bmagic: all i've ever had to do is modify the system time zone |
| 10:33 |
Bmagic |
yeah, I have both /etc/timezone and /etc/localtime setup correctly |
| 10:33 |
Bmagic |
the bash date command reports the right thing |
| 10:35 |
Bmagic |
csharp: you cannot load template folders with more than x number of templates? This might help find such folder: select folder,count(*) from reporter.template group by 1 order by count(*) desc; |
| 10:36 |
dbwells |
Bmagic: we've noticed the differing log timestamps since upgrading to 3.0 in December. Maybe it happened before that, but that's the first time we've noticed. |
| 10:36 |
Bmagic |
dbwells: interesting |
| 10:36 |
dbwells |
It's been annoying, but we haven't looked into at all yet. |
| 10:37 |
dbwells |
I assumed it was somehow related to the timezone-related feature changes, but not sure how. |
| 11:02 |
jeff |
are you both using syslog or file-based logging for the opensrf log messages in question? |
| 11:07 |
|
kmlussier joined #evergreen |
| 11:19 |
|
kmlussier joined #evergreen |
| 11:27 |
|
kmlussier joined #evergreen |
| 11:27 |
|
mmorgan joined #evergreen |
| 11:35 |
Bmagic |
jeff: file-based |
| 11:43 |
kmlussier |
Bmagic / dbwells: We noticed the difference log timestamps back in 2.12. I remember seeing them when we were trying to troubleshoot the search stalling issues. |
| 11:44 |
berick |
kmlussier: are you using file-based logging as well? |
| 11:44 |
berick |
i.e. not syslog |
| 11:48 |
|
yboston joined #evergreen |
| 11:48 |
|
Jaswinder joined #evergreen |
| 11:49 |
Jaswinder |
hey Guys, just an update with my setup with Evergreen, I did the setup using ubuntu 16.04 32 bit version. My next challenge is to find the existing APIs and learn how I can call external API for search |
| 11:50 |
Jaswinder |
Anyone know, where I should look at in the code for API? |
| 12:02 |
|
ngf42 joined #evergreen |
| 12:07 |
|
beanjammin joined #evergreen |
| 12:09 |
Bmagic |
Jaswinder: did you see this https://wiki.evergreen-ils.org/doku.php?id=osrf-devel:user_s_guide |
| 12:15 |
Jaswinder |
Bmagic - Thanks for sharing the link. I am looking at this one now. The documentation is limited. I do see snippet of code for API on the server. It will be helpful to have complete example that shows all areas needed for API development. |
| 12:16 |
Bmagic |
dbwells and remingtron are working on some awesome documentation project that might help |
| 12:16 |
Jaswinder |
That will be great! When should I expect it to be out? |
| 12:18 |
Bmagic |
I believe pieces of it are out there. I am waiting for them to respond. They did a presentation on it last year and I would have expected to see a link here https://evergreen-ils.org/conference/2017-evergreen-international-conference/2017-conference-programs-and-schedule/ |
| 12:19 |
dbwells |
Jaswinder: Well, it's an iterative process. Here in the current public, temporary, rough form: http://library.calvin.edu/devdocs_project/doku.php?id=start |
| 12:20 |
dbwells |
It mostly just draws a bunch of existing content together at the moment. |
| 12:21 |
Jaswinder |
Alright! I will go over these document and then bug you guys if I have further questions. |
| 12:22 |
Bmagic |
Jaswinder++ # Go forth! |
| 12:22 |
Bmagic |
dbwells++ |
| 12:22 |
Bmagic |
remingtron++ |
| 12:23 |
dbwells |
Jaswinder: There isn't really a comprehensize API doc of any kind. There is some generated documenatation for seach here, but it is really just a frame in many cases: https://webby.evergreencatalog.com/opac/extras/docgen.xsl?service=open-ils.search&all=on&offset=0&limit=25 |
| 12:24 |
dbwells |
In your shoes, I would consider just cranking the logging up and poking at the system. Many complete API calls will be revealed that way, though in some cases they are obscured. |
| 12:42 |
terran |
dbwells++ remingtron++ |
| 13:10 |
|
kmlussier joined #evergreen |
| 13:11 |
|
jihpringle joined #evergreen |
| 13:21 |
Jaswinder |
I agree dbwells, I am digging up the code as we speak and certainly anything that sheds a light on APIs will help |
| 13:30 |
|
kmlussier joined #evergreen |
| 13:40 |
pinesol_green |
[evergreen|Jane Sandberg] Docs: adding some funding orgs to release notes acknowledgements - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=b44eb58> |
| 13:40 |
pinesol_green |
[evergreen|Jane Sandberg] Docs: adding code contributors to 3.1 release notes acknowledgements - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=86ab5a9> |
| 13:40 |
pinesol_green |
[evergreen|Jane Sandberg] Docs: adding translators to 3.1 release notes acknowledgements - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=daaa5de> |
| 13:52 |
|
kmlussier joined #evergreen |
| 13:54 |
|
rsoulliere joined #evergreen |
| 13:59 |
rsoulliere |
Hi all, should the webclient support LDAP auth out of the box? If old staff client can connect via LDAP shouldn't the webclient? |
| 14:16 |
dbwells |
rsoulliere: That seems like a reasonable assertion. We use LDAP for patron accounts only, so we haven't missed it, and the same might be true for others. So, probably just an oversight. I would open a bug about it if one doesn't exist, or add a "me too" if it does exist. |
| 14:20 |
rsoulliere |
thanks dbwells. I will do that. |
| 14:37 |
pinesol_green |
[evergreen|Jane Sandberg] Docs: adding more funders to 3.1 release notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=ceca743> |
| 14:37 |
Bmagic |
Anyone been able to create a report that shows average checkout length for a given circulation dataset? |
| 14:40 |
dbs |
rsoulliere++ |
| 14:41 |
dbwells |
Bmagic: an actual report, or just a query? We have a few saved queries for things like that. |
| 14:41 |
Bmagic |
An actual report :) |
| 14:42 |
dbwells |
Sorry, can't help you there :) |
| 14:42 |
* dbs |
thought adding a circ rule defined by copy location would be as simple as adding a new entry with a non-NULL copy_location, copying the same grp/org_unit/circ_lib as before, but apparently not :/ |
| 14:42 |
Bmagic |
yeah, I am thinking that the formula avg(checkin_date - checkout_date) is not possible with reporter |
| 14:44 |
dbs |
Ah, config.weight_assoc tells me that I am wrong |
| 14:44 |
bshum |
Weights |
| 14:44 |
bshum |
Heh, you got it |
| 14:44 |
Bmagic |
weights are heavy |
| 14:45 |
dbs |
I guess http://docs.evergreen-ils.org/reorg/3.0/staff_client_admin/_circulation_policies.html could use some more touch-ups |
| 14:45 |
Bmagic |
</ marty_mcfly> |
| 14:45 |
bshum |
"There's that word again: "heavy." Why are things so heavy in the future? Is there a problem with the Earth's gravitational pull?" |
| 14:45 |
Bmagic |
bshum++ |
| 14:45 |
bshum |
Bmagic++ |
| 14:45 |
Bmagic |
lol |
| 14:46 |
dbwells |
Bmagic: FWIW, we usually do "coalesce(checkin_time, stop_fines_time) - xact_start" to avoid lost circs of infinite length. It's like the roach hotel in reverse, "They check out, but they don't check in!" |
| 14:47 |
Bmagic |
dbwells: I need it to work in the reporter |
| 14:47 |
dbwells |
I know, just a hint for the future. |
| 14:48 |
Bmagic |
roger that! One of the critera was xact_finish is not null |
| 14:48 |
Bmagic |
or perhaps checkin_scan_time is not null |
| 14:49 |
|
StomproJosh joined #evergreen |
| 14:50 |
dbwells |
Yeah, if you limit to items that actually came back, it is easier. We just had cases where we inadvertantly inflated the numbers, cause once they are averaged out, it isn't obvious that there was some 3000 day checkout for a lost item from 10 years ago thrown into the mix. |
| 14:52 |
Bmagic |
I can see that being problematic |
| 14:56 |
mmorgan |
In the xul receipt template editor, is there any way to strip out spaces and punctuation in a string returned by a macro? |
| 14:57 |
mmorgan |
For example, for a hold slip, we would like to turn a patron last name of O'Connor into OCO |
| 14:58 |
Bmagic |
I want to know the answer too! |
| 14:58 |
mmorgan |
:) |
| 14:59 |
Bmagic |
Does Evergreen Authority have a mechanizm for translating search for authors who used* to be called X but are being searched with Y? Example Jennifer Frantz is now Jennifer Fox and Authority control is fixing it. Searches for Frantz returns 0 |
| 15:06 |
mmorgan |
%SUBSTR(0,3)%%PATRON_LASTNAME%%SUBSTR_END% gives us the first three letters of the last name and we're using text-transform: uppercase; |
| 15:26 |
|
kmlussier joined #evergreen |
| 15:28 |
kmlussier |
Bmagic: Re your authority questions. Do you mean something like this? https://gapines.org/eg/opac/browse?blimit=20&qtype=author&bterm=archer%2C+ur&locg=1 |
| 15:28 |
Bmagic |
I suppose, but I was hoping it would work in a basic search |
| 15:30 |
kmlussier |
Bmagic: Nope, not in keyword search. But your timing is good. We've been chatting at MassLNC about "Did You Mean?" searching and having some suggestions generated from authority data in that way. |
| 15:30 |
Bmagic |
or even an author advanced search? |
| 15:31 |
kmlussier |
Bmagic: Yeah, by keyword search, I meant any kind of keyword searching (author, title, general keyword). |
| 15:31 |
|
mmorgan1 joined #evergreen |
| 15:32 |
kmlussier |
Bmagic: If you want those cross references to show up in a browse search, you need to run the authority_authority linker. |
| 15:33 |
Bmagic |
kmlussier++ # thanks |
| 15:37 |
Bmagic |
I am getting search timeouts often. What can I do about it? I am looking at two different examples in osrfsys.log and it just seems like the issue is that it's taking to long. The query doesn't see particularly broad, paste inc. |
| 15:37 |
pastebot |
"Bmagic" at 64.57.241.14 pasted "Example 1" (59 lines) at http://paste.evergreen-ils.org/598 |
| 15:38 |
kmlussier |
Bmagic: What release are you on? |
| 15:38 |
Bmagic |
2.12.7 |
| 15:39 |
pastebot |
"Bmagic" at 64.57.241.14 pasted "Example 2" (59 lines) at http://paste.evergreen-ils.org/599 |
| 15:39 |
kmlussier |
Bmagic: bug 1704396 |
| 15:39 |
pinesol_green |
Launchpad bug 1704396 in Evergreen 2.12 "Slowness for metarecord and one-hit searches in 2.12" [High,Fix released] https://launchpad.net/bugs/1704396 |
| 15:39 |
Bmagic |
oh! it's that? |
| 15:39 |
kmlussier |
Bmagic: We had major issues with that. |
| 15:40 |
Bmagic |
I remember that of course, I just didn't connect the two. |
| 15:40 |
|
yboston joined #evergreen |
| 15:42 |
Bmagic |
Are there any other cherry-picks that I need to include on this vanilla install? |
| 15:42 |
kmlussier |
Bmagic: The title is misleading. I think it was easier to reproduce for metarecord and one-hit searches, but we saw the problem with a lot of our searches. |
| 15:42 |
Bmagic |
The paste reports that there were "49 rows"... |
| 15:43 |
Bmagic |
and the first example says "80 rows" |
| 15:49 |
pinesol_green |
[evergreen|Chris Sharp] LP#1750912 - Add add_translations.py convenience script to i18n tools. - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=f09ae17> |
| 16:22 |
|
mmorgan joined #evergreen |
| 16:30 |
Bmagic |
kmlussier: thanks! I think that is the issue on those slow searches for that installation |
| 16:31 |
kmlussier |
Bmagic: Awesome! There were cheers across all the Evergreen libraries in Massachusetts when that patch got loaded. |
| 16:31 |
Bmagic |
kmlussier: Along these same lines, have you seen an Evergreen installation where attempting to open a reporting template folder with over 1000 templates in it fails? |
| 16:31 |
|
yboston joined #evergreen |
| 16:32 |
Bmagic |
berick and csharp and I were talking about it this morning. I have the logs http://paste.evergreen-ils.org/597 |
| 16:33 |
kmlussier |
Bmagic: I'm not aware of that problem, no. |
| 16:33 |
Bmagic |
having over 1000 templates in a folder seems like a bad idea but it happens I guess |
| 16:38 |
kmlussier |
Bmagic: Yes, I was going to say we probably don't have that many templates in a folder, but I could be wrong. I would think it would be difficult to find the template you need with that many in one folder. |
| 16:38 |
* kmlussier |
heads out for the day. |
| 16:43 |
pinesol_green |
[evergreen|Jane Sandberg] Docs: correcting funder names for 3.1 release notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=c9faf11> |
| 16:43 |
pinesol_green |
[evergreen|Jane Sandberg] Merge branch 'master' of git.evergreen-ils.org:Evergreen - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=4900b49> |
| 17:05 |
|
mmorgan left #evergreen |
| 17:15 |
pinesol_green |
[evergreen|Remington Steed] Docs: Update "Circulation - Patron Record" for web client - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=6790dd6> |
| 17:28 |
|
jvwoolf left #evergreen |
| 17:49 |
|
gmcharlt joined #evergreen |
| 17:50 |
|
gmcharlt joined #evergreen |
| 18:19 |
|
abowling joined #evergreen |
| 18:48 |
|
Christineb joined #evergreen |
| 18:52 |
|
abowling1 joined #evergreen |
| 19:43 |
|
jonadab joined #evergreen |
| 21:50 |
|
jonadab joined #evergreen |
