| Time |
Nick |
Message |
| 00:03 |
|
sandbergja joined #evergreen |
| 07:00 |
|
agoben joined #evergreen |
| 07:08 |
|
rjackson_isl joined #evergreen |
| 07:09 |
|
agoben_ joined #evergreen |
| 08:04 |
|
collum joined #evergreen |
| 08:48 |
|
mmorgan joined #evergreen |
| 09:07 |
|
Dyrcona joined #evergreen |
| 09:20 |
|
yboston joined #evergreen |
| 09:23 |
|
dbwells joined #evergreen |
| 09:30 |
|
jvwoolf joined #evergreen |
| 09:46 |
|
sandbergja joined #evergreen |
| 09:48 |
|
JBoyer joined #evergreen |
| 09:58 |
|
mmorgan1 joined #evergreen |
| 10:24 |
|
yboston joined #evergreen |
| 10:32 |
gsams |
JBoyer: I ran into LP1786100 recently and just added the log info you requested. |
| 10:48 |
Dyrcona |
Lp 1786100 |
| 10:48 |
pinesol |
Launchpad bug 1786100 in Evergreen "Removed copy tags are not removed" [Undecided,Confirmed] https://launchpad.net/bugs/1786100 |
| 10:50 |
JBoyer |
gsams, yeah, that's what I thought it was. The function is being called on DELETE but still trying to look at NEW, which doesn't exist. It may have been possible that the functions are created with 2 different names depending on when your database was first setup vs what upgrades have been run on it. (or PG versions, etc.) |
| 10:53 |
JBoyer |
do you have upgrade script 1081 on that database? |
| 10:53 |
JBoyer |
1063 is where they're initially generated, with names chosen programmatically, which is why my fix in 1081 may have missed them. |
| 10:53 |
|
sandbergja joined #evergreen |
| 10:57 |
gsams |
JBoyer: 1081 appears to be applied. |
| 10:59 |
JBoyer |
in psql, does \d+ asset.copy_tag_copy_map give you 1 trigger or two? |
| 10:59 |
JBoyer |
(Or, does it list DELETE in the AFTER list) |
| 11:01 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 11:07 |
|
rfrasur joined #evergreen |
| 11:08 |
gsams |
JBoyer: Two triggers, second trigger has DELETE in the AFTER list where the first does not. The first trigger is inherit_asset_copy_tag_copy_map_copy_fkey and the second is inherit_copy_tag_copy_map_copy_fkey |
| 11:37 |
|
khuckins joined #evergreen |
| 11:43 |
|
sandbergja joined #evergreen |
| 11:46 |
|
jihpringle joined #evergreen |
| 11:46 |
|
yboston joined #evergreen |
| 12:27 |
|
sandbergja_ joined #evergreen |
| 12:30 |
|
mmorgan joined #evergreen |
| 12:31 |
* JBoyer |
just gets back from lunch. |
| 12:34 |
JBoyer |
gsams, ok, I think I can figure out what happened. The SELECT in 1063 uses the constraint name to determine the trigger name, so depending on whether PG put the schema name in the constraint name determines if this problem persists after 1081 is applied. |
| 12:34 |
JBoyer |
In my case 'asset; was in the name, so thats |
| 12:35 |
JBoyer |
's all I changed. If you drop that trigger (and any other duplicates from 1063) you'll be back in business. If this happened to all of the triggers I don't think you can remove copy notes or delete copies from buckets either. |
| 12:42 |
JBoyer |
I suppose one more upgrade script that just drops all of the no-asset triggers should finally put this thing to rest. |
| 12:42 |
JBoyer |
(I was interested in the logs because I didn't know what the other name for the triggers could be) |
| 12:43 |
gsams |
Well I'm glad I had a chance to look into it at least |
| 12:43 |
JBoyer |
gsams++ |
| 12:43 |
JBoyer |
If you want to go ahead and drop those triggers for your users they'll be able to do their thing without any problems and it won't affect the eventual upgrade script. |
| 12:44 |
|
collum joined #evergreen |
| 12:44 |
gsams |
JBoyer++ #I will definitely be doing that soon. |
| 13:01 |
|
khuckins joined #evergreen |
| 14:26 |
pinesol |
[evergreen|Mike Risher] lp1739292 merge UI displays record summary - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=e689c8c> |
| 14:34 |
|
khuckins joined #evergreen |
| 15:07 |
|
yboston joined #evergreen |
| 15:12 |
|
mmorgan1 joined #evergreen |
| 16:58 |
|
mmorgan1 left #evergreen |
| 17:54 |
|
khuckins joined #evergreen |
| 23:01 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
