| Time |
Nick |
Message |
| 06:00 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 07:50 |
rjackson_isl_hom |
csharp nothing stuck out |
| 07:58 |
rjackson_isl_hom |
after our upgrade to 3.4.4 the actor.usr_delete function has become corrupted. There are two entries attempting to set guardian to null so function fails always. |
| 07:59 |
rjackson_isl_hom |
two questions for the community: where do I find the function definitions that should be in place |
| 08:00 |
rjackson_isl_hom |
and how to I run the create or replace function? Can I do that while system is online? (well I guess 3 questions...) |
| 08:08 |
|
Dyrcona joined #evergreen |
| 08:10 |
|
rfrasur joined #evergreen |
| 08:17 |
|
mantis2 joined #evergreen |
| 08:20 |
csharp |
rjackson_isl_hom: https://git.evergreen-ils.org/?p=Evergreen.git;a=blob;f=Open-ILS/src/sql/Pg/999.functions.global.sql;h=45321a0d2ba9522a2f07844c8f51818a8e116a30;hb=refs/heads/tags/rel_3_4_4#l747 |
| 08:21 |
csharp |
rjackson_isl_hom: you should be able to copy/paste that CREATE OR REPLACE FUNCTION section into a file and run it in psql |
| 08:22 |
rjackson_isl_hom |
csharp++ - I just now made it to that function and was hoping we had the latest thanks - think this is one that is really clobbering our system when the 7:30 AM cron kicks off to clean up old accounts |
| 08:31 |
|
mmorgan joined #evergreen |
| 08:44 |
Dyrcona |
rjackson_isl_hom: A job to delete users can run for a long time, and if run at an inopportune moment it can interfere with circulation depending on who is being deleted and how much "junk" they have in the database. |
| 08:44 |
Dyrcona |
If you use the standard srfsh version of the script, I think it can also time out and not finish. |
| 08:45 |
|
alynn26 joined #evergreen |
| 08:45 |
Dyrcona |
We have a SQL version that we run once per week at 2:00 AM. |
| 08:46 |
Dyrcona |
Looks like every Saturday morning. |
| 08:46 |
* Dyrcona |
now decides to check the logs to see what he is replying to. :) |
| 08:47 |
rjackson_isl_hom |
Dyrcona ours runs daily at 7:30 AM - just catching up enough to figure out this issue - look slike it could have been causing loads of other issues since it spun for a long time daily with errors |
| 08:47 |
rjackson_isl_hom |
updated the function def ( csharp++) and now seems to be processing as intended |
| 08:48 |
Dyrcona |
That's good. Did you have a custom version of the function previously? |
| 08:49 |
Dyrcona |
We did, and I had to keep it up to date across upgrades. We now use the standard function, since 3.2, at least. |
| 08:50 |
rjackson_isl_hom |
I didn't look thru line per line but hopefully it was standard :( |
| 08:50 |
rjackson_isl_hom |
looked close on the mstching I did do |
| 08:50 |
rjackson_isl_hom |
JBoyer might be able to chime in on that one ;) |
| 08:51 |
Dyrcona |
OK. Seems odd that it would end up with with two lines setting the guardian. I could see that happening from a bad conflict resolution, though. |
| 08:53 |
JBoyer |
Yeah, I don't recall ever customizing that, unless it might have been to apply some fix that was in a later version than was being used. |
| 08:53 |
JBoyer |
(which shouldn't be an issue now anyway) |
| 08:54 |
Dyrcona |
Yeah, we've done that among other things. It could have been a bad resolution in an upgrade script, too, so maybe someone else's fault. :) |
| 08:57 |
rjackson_isl_hom |
JBoyer++ - a great resource ;) |
| 09:33 |
|
dbwells joined #evergreen |
| 09:42 |
|
jvwoolf joined #evergreen |
| 09:52 |
|
jvwoolf1 joined #evergreen |
| 09:58 |
|
Christineb joined #evergreen |
| 10:32 |
Dyrcona |
The shuffle on Rhythmbox has not let me down this morning. |
| 10:45 |
Dyrcona |
cache-- |
| 10:45 |
Dyrcona |
cache++ |
| 10:45 |
Dyrcona |
:) |
| 10:55 |
csharp |
troubleshooting something we just found on 3.6.1 when doing the open-ils.circ.holds.test_and_create.batch call |
| 10:55 |
csharp |
CALL: open-ils.cstore open-ils.cstore.json_query.atomic {"from":["actor.org_unit_ancestor_setting","circ.hold_boundary.soft","\"171\""]} |
| 10:56 |
csharp |
since those quotes are escaped, the function is failing because it's text, not an integer |
| 10:56 |
csharp |
and I can't seem to see where the quotes are escaped (or why they would be) |
| 10:59 |
berick |
csharp: i'd start with the activity log entry for the open-ils.circ.holds.test_and_create.batch call, see if it's coming from the client |
| 11:00 |
berick |
though, additionally, that api should probably coerce it to a number regardless since that's what it needs |
| 11:00 |
csharp |
yeah, it's like that in the activity log too |
| 11:01 |
mmorgan |
Could the ou setting be stored oddly in the database? Like with curly quotes? |
| 11:01 |
berick |
k so the client is sending a number as a string, which is less than ideal, but not strictly wrong |
| 11:01 |
csharp |
open-ils.circ open-ils.circ.holds.test_and_create.batch "<nunyabusiness>", {"patronid":22067,"pickup_lib":"\"171\"","hold_type":"T","email_notify":true,"phone_notify":null,"thaw_date":null,"frozen":false,"sms_notify":null,"sms_carrier":null,"holdable_formats_map":null}, [4998761] |
| 11:01 |
berick |
oh wait |
| 11:02 |
berick |
no, that's not ok |
| 11:03 |
berick |
mmorgan: yeah, i wonder if a setting somewhere is not stored correctly, like a default pickup lib user setting |
| 11:03 |
berick |
"\"171\"" is still valid JSON, so the DB would accept it |
| 11:03 |
Dyrcona |
Yeah, I was about to suggest it might be a user setting. |
| 11:04 |
csharp |
mmorgan: that was my first thought, but I stopped when it wasn't the OU setting - I'll dig in that direction |
| 11:05 |
csharp |
well, I should have mentioned that we're testing the fix to bug 1889128 on *top* of 3.6.1 |
| 11:05 |
pinesol |
Launchpad bug 1889128 in Evergreen "Angular Staff Catalog: Place Another Hold & Multi-Holds" [High,Confirmed] https://launchpad.net/bugs/1889128 |
| 11:07 |
Dyrcona |
csharp: You removed the sticker, opened the seal, and voided the warranty. :) |
| 11:07 |
csharp |
Dyrcona++ |
| 11:08 |
* csharp |
glares at Open-ILS/src/eg2/src/app/staff/catalog/hold/hold.component.ts |
| 11:08 |
berick |
csharp: you have that code in its current form? |
| 11:09 |
csharp |
sec... |
| 11:10 |
csharp |
berick: https://git.evergreen-ils.org/?p=evergreen/pines.git;a=blob;f=Open-ILS/src/eg2/src/app/staff/catalog/hold/hold.component.ts;h=3ac30414eb7e01718fa32b722c5f41cccd6ab7ca;hb=refs/heads/rel_3_6_1 |
| 11:11 |
berick |
csharp: yeah, you're missing some needed patches |
| 11:11 |
csharp |
oh hmm |
| 11:12 |
berick |
csharp: for starters, https://bugs.launchpad.net/evergreen/+bug/1887429 |
| 11:12 |
pinesol |
Launchpad bug 1889128 in Evergreen "duplicate for #1887429 Angular Staff Catalog: Place Another Hold & Multi-Holds" [High,Confirmed] |
| 11:13 |
berick |
which was linked as a dupe for the multi-holds patch |
| 11:13 |
mmorgan |
berick: Did my signoff branch get them all? https://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/mmorgan/lp1889128_signoff |
| 11:13 |
mmorgan |
Thought I did. |
| 11:13 |
berick |
mmorgan: was just checking... |
| 11:14 |
berick |
well my branch may have missed it too |
| 11:14 |
berick |
mmorgan: hm, no, the patch is there in your s/o branch |
| 11:15 |
mmorgan |
I did my testing on terranm's server, and it worked well there. |
| 11:15 |
berick |
but it has a different LP number so csharp may have missed it |
| 11:16 |
csharp |
oh - yeah - that's it |
| 11:16 |
csharp |
I didn't realize they were all of a piece because of the different LP numbers |
| 11:16 |
csharp |
berick++ mmorgan++ |
| 11:17 |
mmorgan |
csharp++ terranm++ berick++ |
| 11:28 |
|
rfrasur joined #evergreen |
| 11:43 |
|
Cocopuff2018 joined #evergreen |
| 11:47 |
|
mantis2 left #evergreen |
| 12:01 |
|
jihpringle joined #evergreen |
| 12:08 |
|
Jp joined #evergreen |
| 12:28 |
Jp |
. |
| 13:15 |
|
sandbergja joined #evergreen |
| 14:21 |
csharp |
we're seeing the same kind of actor drone exhaustion that Dyrcona and jeffdavis have previously reported |
| 14:21 |
csharp |
I'll see if I can test berick's fix for bug 1896285 |
| 14:21 |
pinesol |
Launchpad bug 1896285 in Evergreen "Use batch methods for multi-row grid actions" [Medium,Confirmed] https://launchpad.net/bugs/1896285 |
| 14:23 |
jeffdavis |
^ should mention I re-tested those fixes, one seemed to work the second time around but I still saw drone exhaustion on the other test case. I need to revisit and properly report results. |
| 14:44 |
pinesol |
[evergreen|Chris Sharp] Forward-port 3.4.4-3.4.5 version upgrade script - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=a76b279> |
| 14:46 |
pinesol |
[evergreen|Chris Sharp] Forward-port 3.5.1-3.5.2 version upgrade script - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=f7fe764> |
| 14:52 |
pinesol |
[evergreen|Jason Boyer] Forward-port 3.6.1 upgrade script - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=0297d91> |
| 15:02 |
csharp |
JBoyer++ |
| 15:03 |
JBoyer |
csharp++ |
| 15:59 |
|
sandbergja joined #evergreen |
| 16:12 |
Dyrcona |
JBoyer++ csharp++ |
| 16:24 |
jeff |
anyone here love/hate a particular hold locker / package locker make/model? recommendations (and warnings) appreciated, here or via msg. :-) |
| 17:02 |
|
mmorgan left #evergreen |
| 17:30 |
|
jeffdavis joined #evergreen |
| 17:30 |
|
Cocopuff2018 joined #evergreen |
| 18:00 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 18:39 |
|
Cocopuff2018 joined #evergreen |
