| Time |
Nick |
Message |
| 01:44 |
|
khuckins joined #evergreen |
| 03:04 |
|
khuckins joined #evergreen |
| 04:48 |
|
khuckins joined #evergreen |
| 05:27 |
|
khuckins joined #evergreen |
| 06:01 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 07:28 |
|
rjackson_isl_hom joined #evergreen |
| 07:39 |
|
agoben joined #evergreen |
| 07:51 |
|
rfrasur joined #evergreen |
| 08:02 |
|
mantis1 joined #evergreen |
| 08:14 |
|
Dyrcona joined #evergreen |
| 08:45 |
|
mmorgan joined #evergreen |
| 09:08 |
Dyrcona |
Has anyone had success updating npm when it suggests to do so, with Evergreen? The couple of times that I have tried it, installing Evergreen was broken after. |
| 09:29 |
|
dbwells joined #evergreen |
| 09:31 |
|
sandbergja joined #evergreen |
| 10:24 |
gmcharlt |
calling a freeze on pushes to rel_3_4 and rel_3_5 starting at 11 a.m. ET / 8 a.m. PT for release-building |
| 10:25 |
gmcharlt |
(and if you're planning to merging anything in the next half hour, please say so) |
| 10:25 |
|
jvwoolf joined #evergreen |
| 10:25 |
csharp |
Dyrcona: I haven't done much with that, but I want to be able to update npm when prompted |
| 10:26 |
csharp |
seems like a Bad Idea™ to ignore deprecation, etc. warnings |
| 10:47 |
|
sandbergja joined #evergreen |
| 10:56 |
|
sandbergja joined #evergreen |
| 10:57 |
|
sandbergja_ joined #evergreen |
| 10:57 |
|
collum joined #evergreen |
| 10:59 |
|
nfBurton joined #evergreen |
| 11:01 |
nfBurton |
Are action-trigger errors stashed in a log somewhere? |
| 11:01 |
Dyrcona |
nfBurton: It depends. What errors are you asking about? |
| 11:02 |
nfBurton |
I'm getting an error state on some of the holds_emails and syslog isn't showing much |
| 11:03 |
Dyrcona |
Have you checked the action_trigger.event_ouput table? |
| 11:04 |
nfBurton |
probably something to do with the template if I had to guess |
| 11:04 |
Dyrcona |
Hang on a second. My Pg 12 database is not working. I |
| 11:04 |
Dyrcona |
have to tunnel to produciton. |
| 11:04 |
Dyrcona |
bleh. fingers-- |
| 11:04 |
nfBurton |
Oh no! |
| 11:05 |
nfBurton |
I have high hopes gor PG12 |
| 11:05 |
mmorgan |
nfBurton: I do see errors in the syslog, but often times they're not that useful. |
| 11:05 |
nfBurton |
yeah. In this case not so much |
| 11:05 |
Dyrcona |
nfBurton: For the error event does the error_output have anything in it? |
| 11:06 |
Dyrcona |
If so, it would be the id of a row in action_trigger.event_output. |
| 11:06 |
nfBurton |
I assume it's a templating issue but I mostly just copied the one I had for each branch and they're erroring now. I took out the variable I added, just waiting for more to trigger now but there IS outpus |
| 11:06 |
nfBurton |
output |
| 11:07 |
mmorgan |
nfBurton: Could it be an environment issue? I recently cloned some templates and for some reason the environment didn't come over. |
| 11:07 |
nfBurton |
Actually I'm missing some of the outputs too :S |
| 11:07 |
nfBurton |
Maybe I need to go down to just the one again |
| 11:08 |
nfBurton |
Wonder why that happens |
| 11:08 |
Dyrcona |
mmorgan nfBurton: I have a function for cloning event definitions. |
| 11:08 |
nfBurton |
No issues until I tried to split them |
| 11:08 |
nfBurton |
! |
| 11:08 |
nfBurton |
I assume its more than copying the row then -__- lol |
| 11:09 |
Dyrcona |
I'm not sure that I understand. What do you mean by "split them?" |
| 11:09 |
csharp |
nfBurton: check Open-ILS/src/perlmods/lib/OpenILS/Application/Trigger/Event.pm for "error" to see when it sets that state |
| 11:09 |
nfBurton |
I had an overreaching action trigger for holds that covered all branches and I tried to give one per branch |
| 11:10 |
Dyrcona |
Oh! Perfect candidate for my function! |
| 11:10 |
* mmorgan |
has done a lot of that lately. |
| 11:10 |
Dyrcona |
nfBurton: https://pastebin.com/2hcK8Xjj |
| 11:11 |
csharp |
at that point I conclude we need to add functionality to the A/T infrastructure to handle multiple branches |
| 11:11 |
Dyrcona |
Nope. It's can handle different events for different orgs already. |
| 11:11 |
csharp |
I haven't looked at it in a while |
| 11:11 |
nfBurton |
Dyrcona++ |
| 11:12 |
Dyrcona |
You will have to disable the one for all branches, though, or patrons may get two notices. |
| 11:12 |
nfBurton |
oh lol okay. I thought it would act as a fallback |
| 11:13 |
Dyrcona |
csharp: On a 9.6 database pg_restore'd to Pg 12: https://pastebin.com/s2K2KFXX |
| 11:13 |
mmorgan |
We had a single event def for years, but did disable that one and have been cloning it for each org unit when it's ready to start sending notifications. |
| 11:13 |
Dyrcona |
Looking at the code, it will run all events that match. |
| 11:13 |
|
khuckins joined #evergreen |
| 11:17 |
Dyrcona |
Re my message to csharp with a Pg 12 error: It is because I'm using the Pg 9.6 client. If I use the Pg 12 client, everything is fine. Also: https://stackoverflow.com/questions/58461178/how-to-fix-error-column-c-relhasoids-does-not-exist-in-postgres |
| 11:22 |
csharp |
ah |
| 11:22 |
Dyrcona |
nfBurton: Keep in mind that my function returns the id of the new event definition in case you want to change the template or granularity or whatever. I've used it a lot since writing it, because of COVID. |
| 11:26 |
Dyrcona |
I upgraded the VM to postgresql-client-12 and everything is good. Also means that I get the new --csv option that I have gotten used to using on my laptop. :) |
| 11:28 |
Dyrcona |
So far, Pg 12 has been nice. I've noticed that inserts seem to be a lot faster than on Pg 9.6. |
| 11:31 |
Dyrcona |
That said, I haven't tested anything really complicated, yet. |
| 11:31 |
csharp |
I'll start looking at it soon |
| 12:18 |
|
nfBurton joined #evergreen |
| 12:22 |
|
khuckins joined #evergreen |
| 12:22 |
|
Christineb joined #evergreen |
| 12:23 |
nfBurton |
Dyrcona++ |
| 12:39 |
|
sandbergja joined #evergreen |
| 12:44 |
|
jihpringle joined #evergreen |
| 13:02 |
|
mrisher joined #evergreen |
| 13:57 |
|
nfBurton joined #evergreen |
| 14:31 |
pinesol |
[evergreen|Remington Steed] Docs: Release notes for 3.4.3 - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=5652314> |
| 14:31 |
pinesol |
[evergreen|Jane Sandberg] Docs: 3.4.4 release notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=3491d38> |
| 14:51 |
|
jlundgren1 joined #evergreen |
| 14:54 |
|
sandbergja joined #evergreen |
| 15:31 |
Bmagic |
Dyrcona: That PG function for cloning seems like a good candidate for inclusion into stock. Cloning AT environment doesn't work most of the time anyway. What if EG didn't give you a choice and just cloned the AT with the environment, end of story? |
| 15:34 |
Dyrcona |
Bmagic: I don't use the admin interface, so I don't know how well or how poorly this works in the client. There are other ways besides a database function to make it copy the environment and parameters when cloning. |
| 15:35 |
Bmagic |
agreed, just seems that this solution has already been coded and "tested" in production. It seems straight forward as well. Just sayin' |
| 15:35 |
rfrasur |
I could have dreamed this. Has the library settings editor been angularized in an upcoming version? I'm not seeing it...but am hopeful that I'm just overlooking. |
| 15:36 |
Bmagic |
rfrasur: I read that somewhere too.... |
| 15:37 |
rfrasur |
Was it in a fantasy novel? Or something on the wiki? |
| 15:37 |
Dyrcona |
Bmagic: I wrote this function to fulfill a need that we had at CWMARS. It seems others have the same need. If you think add this function to stock and modifying the A/T admin interface to use it on the backend is a good idea, I'm OK with that. |
| 15:38 |
Bmagic |
Dyrcona: yeah, that's what I was thinking. But first, there needs to be an LP bug. There might be one already.... |
| 15:38 |
rfrasur |
Oh yes. I see it. |
| 15:38 |
Bmagic |
If there's not, there should be :) |
| 15:38 |
rfrasur |
(It's 3.6) |
| 15:39 |
Bmagic |
rfrasur++ |
| 15:39 |
Dyrcona |
rfrasur: Yes, I was thinking that it was future, not present, and I was about to check the code to verify. |
| 15:39 |
Bmagic |
mmmmm, code</homer> |
| 15:40 |
rfrasur |
Eh. I might still be wrong. If local admin circulation policy UI =/= library settings editor. |
| 15:40 |
Dyrcona |
Bmagic: If you want to open/find a Lp bug, that's OK. |
| 15:41 |
Bmagic |
groovy |
| 15:41 |
Bmagic |
Dyrcona: I like this plan! I'm excited to be a part of it! |
| 15:41 |
* rfrasur |
thinks they're not the same thing. |
| 15:41 |
Dyrcona |
rfrasur: Very likely works in progress. Everything is supposed to get angularized eventually. |
| 15:42 |
rfrasur |
++ |
| 15:43 |
rfrasur |
https://bugs.launchpad.net/evergreen/+bug/1839341 |
| 15:43 |
pinesol |
Launchpad bug 1839341 in Evergreen "Port Local Admin Org Unit Settings UI to Angular" [Wishlist,Confirmed] |
| 15:43 |
Dyrcona |
rfrasur++ |
| 15:46 |
|
sandbergja joined #evergreen |
| 15:48 |
|
sandbergja joined #evergreen |
| 16:06 |
Bmagic |
Dyrcona: bug 1891232 |
| 16:06 |
pinesol |
Launchpad bug 1891232 in Evergreen "Cloning Action Triggers fails to clone environment and parameters" [Undecided,New] https://launchpad.net/bugs/1891232 |
| 16:08 |
Bmagic |
csharp: any insight on the "EDI fetcher continuously downloading and processing the same file" issue ? |
| 16:09 |
Dyrcona |
Bmagic++ |
| 16:09 |
Bmagic |
My theory is that anything that staff do to a copy, like receive it (marking it received) is overwritten by the EDI fetcher when it processes the same EDI message again (marking it backordered) |
| 16:13 |
pinesol |
[evergreen|Jane Sandberg] Docs: tweaks to 3.4.4 release notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=0a157cd> |
| 16:13 |
pinesol |
[evergreen|Jane Sandberg] Docs: 3.5.1 release notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=c6748a1> |
| 16:18 |
Dyrcona |
sandbergja++ |
| 16:20 |
Bmagic |
berick might have a clue too? Reading the code in EDI.pm ~ line 129 suggest that it should* be ignoring files with the same name |
| 16:20 |
Bmagic |
Anyone care to run this on their instance: select remote_file,count(*) from acq.edi_message where create_time > now() - '100 days'::interval group by 1 having count(*) > 1 order by 2 |
| 16:23 |
|
mantis1 left #evergreen |
| 16:25 |
berick |
Bmagic: i have 265 instances of count(*) > 1 files. what's the question? |
| 16:25 |
Bmagic |
I have an example file that was processed 17,328 times in the last 100 days.... Seems like that's not normal. It runs every hour |
| 16:25 |
Bmagic |
berick: oh, so it is normal? |
| 16:26 |
Bmagic |
I am chasing down an issue where staff are claming to have recieved an item that was on backorder... then it returns to "back order" again - my theory is the EDI fetcher is changing it back |
| 16:27 |
berick |
the EDI should not be re-processing the same remote file |
| 16:28 |
berick |
my cron mail is full of stuff like: |
| 16:28 |
berick |
EDI: out/0201B5223127.inv already retrieved. Skipping at /usr/local/share/perl/5.22.1/OpenILS/Application/Acq/EDI.pm line 158. |
| 16:31 |
Dyrcona |
I filter that out with | grep -v 'already retrieved' on the crontab entry. :) |
| 16:31 |
Dyrcona |
We've not had reports of this issue with the EDI fetcher. |
| 16:35 |
berick |
Bmagic: having multiple edi_message's with the same file can also happen when a single EDI file contains multiple message (e.g multiple invoices) |
| 16:35 |
berick |
so yeah that part is normal |
| 16:40 |
berick |
well, 17,328 is, uh, atypical |
| 16:45 |
Bmagic |
thanks berick |
| 16:46 |
Bmagic |
pretty sure my logs show that it's NOT skipping it though. Double checking |
| 17:18 |
|
mmorgan left #evergreen |
| 17:41 |
Bmagic |
berick: would I expect these rows to have a status of anything other than "processed" ? |
| 17:44 |
berick |
Bmagic: that's what I'm seeing |
| 17:44 |
berick |
guess order resposnes never make it to 'complete' |
| 17:44 |
Bmagic |
Therefore, I can conclude that the only way I would know if it really digested it or not is the logs? |
| 17:45 |
berick |
Bmagic: do you have 17k entries in acq.edi_message for that problematic file name? |
| 17:45 |
Bmagic |
the logs have rolled off by now for most of those, but yeah, probably (if we kept em) |
| 17:46 |
Bmagic |
I'm zgrepping all the osrfsys.log.* for a different file (one that shows 974 times, related to this ticket) |
| 17:46 |
Bmagic |
I am seeing a "./edi_fetcher.pl [INFO:29071:CStoreEditor.pm:139:] editor[0|0] request en-US open-ils.cstore.direct.acq.edi_message.id_list.atomic" |
| 17:48 |
Bmagic |
call - which is where the software should be looking for a duplicate row |
| 17:48 |
Bmagic |
and, right now I am seeing a return of "1 result(s)" - which means (I think) it found a duplicate, and therefore, should skip that file |
| 17:51 |
berick |
Bmagic: you delete entries from acq.edi_message ? |
| 17:51 |
Bmagic |
no, should we? |
| 17:51 |
berick |
no i was confused by an earlier response .."if we kept them" -- must have been about logs |
| 17:52 |
Bmagic |
yeah, logs |
| 17:52 |
berick |
Bmagic: the one you said happaned 17k times, are there 17k entries for that file in acq.edi_message? |
| 17:52 |
Bmagic |
yes, 17k in edi_message |
| 17:53 |
berick |
Bmagic: how do the create dates compare across the set of 17k? |
| 17:54 |
berick |
do they span days or just happen all at once more or less? |
| 17:54 |
Bmagic |
every hour on the nose |
| 17:54 |
berick |
one each hour? |
| 17:54 |
Bmagic |
yep |
| 17:54 |
berick |
and that's when you run teh fetcher i assume |
| 17:54 |
Bmagic |
yep |
| 17:55 |
berick |
Bmagic: do they all show the same account value? |
| 17:55 |
berick |
and purchase_order value |
| 17:55 |
berick |
and are they all in 'processed' status? |
| 17:55 |
berick |
or complete |
| 17:56 |
Bmagic |
all identical "proccessed" - same PO, same File name, everything. create_time is different, id is different |
| 18:01 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 18:01 |
berick |
Bmagic: i can't explain that. what EG version? |
| 18:02 |
Bmagic |
3.3.4 |
| 18:02 |
Bmagic |
so you are saying that you don't have those same rows repeated? (I though you did) |
| 18:03 |
Bmagic |
I think if finally stopped because the vendor deleted the file from their ftp server |
| 18:03 |
Bmagic |
if/it |
| 18:03 |
|
jvwoolf1 joined #evergreen |
| 18:04 |
berick |
Bmagic: you should only have multiple rows in acq.edi_message for the same file name / account info if the message contains multiple EDI entries, e.g. multiple invoices in a single file |
| 18:04 |
berick |
in those cases, the messages would all be created at the same time |
| 18:04 |
berick |
17k duplicates, once per hour, should absolutely not be happening |
| 18:04 |
Bmagic |
I see |
| 18:04 |
Bmagic |
well, that's the case here |
| 18:05 |
berick |
no diffs for EDI.pm or edi_fetcher.pl from rel_3_3 to master |
| 18:05 |
Bmagic |
gotta go - I'll pick it up tomorrow. Thanks for your help! |
| 18:05 |
Bmagic |
I checked that too! |
| 18:06 |
Bmagic |
EDI.pm is the same. Didn't check edi_fetcher though... |
| 18:06 |
Bmagic |
nope - stock |
| 18:07 |
Bmagic |
checking against rel_3_3 |
| 18:08 |
Bmagic |
maybe my filename is tricking the search code? |
| 18:09 |
Bmagic |
it has a bunch of underscores in it |
| 18:09 |
berick |
odd thing is you said the query was finding a match... |
| 18:09 |
Bmagic |
those files are identical to stock 3_3 |
| 18:10 |
Bmagic |
yeah, it is, if I am reading it right |
| 18:10 |
Bmagic |
though.... I've not had a new duplicate row since 8/3 |
| 18:10 |
berick |
Bmagic: are you capturing cron emails? it will report in the email result which files it skipped because of dupes |
| 18:11 |
berick |
(should really move that to $logger) |
| 18:11 |
Bmagic |
stdout? |
| 18:11 |
berick |
yeah |
| 18:11 |
Bmagic |
I really gtg |
| 18:11 |
Bmagic |
thanks again! |
| 18:11 |
berick |
same |
| 18:11 |
Bmagic |
berick++ |
| 18:11 |
berick |
later |
| 18:25 |
|
abowling1 joined #evergreen |
| 18:42 |
|
abowling joined #evergreen |
| 19:04 |
|
dbwells joined #evergreen |
| 19:07 |
|
dbwells joined #evergreen |
| 19:12 |
pinesol |
[evergreen|Jane Sandberg] Docs: asciidoc syntax correction for 3.4 release notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=8a97bfa> |
| 19:12 |
pinesol |
[evergreen|Jane Sandberg] Docs: asciidoc syntax correction for 3.5 release notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=43bf6d1> |
| 19:30 |
|
mrisher_ joined #evergreen |
| 20:00 |
pinesol |
[evergreen|Jane Sandberg] Forward-port 3.5.1 upgrade script - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=0fb3cd8> |
| 20:49 |
|
mrisher joined #evergreen |
| 23:05 |
|
khuckins joined #evergreen |
| 23:55 |
|
sandbergja joined #evergreen |
